Proefschrift voorgelegd tot het behalen van de graad van doctor in de Sociale

Wetenschappen: Communicatiewetenschappen aan de Universiteit Antwerpen te

verdedigen door:

NATHALIE CLAESSENS

CELEBRITY, MEDIA, AND AUDIENCES

Social and Cultural Meaning in Contemporary Western Societies

UNIVERSITEIT ANTWERPEN
FACULTEIT POLITIEKE EN SOCIALE WETENSCHAPPEN
DEPARTEMENT COMMUNICATIEWETENSCHAPPEN

Promotor: Prof. Dr. Hilde Van den Bulck

Antwerpen, 2013

1
 Faculteit Politieke en Sociale Wetenschappen
Departement Communicatiewetenschappen
Onderzoeksgroep Media, Policy & Culture

CELEBRITY, MEDIA, AND AUDIENCES

SOCIAL AND CULTURAL MEANING IN CONTEMPORARY WESTERN SOCIETIES

Proefschrift voorgelegd tot het behalen van de graad van doctor in de Sociale
wetenschappen: Communicatiewetenschappen aan de Universiteit Antwerpen te
verdedigen door:

Nathalie CLAESSENS

Promotor: Prof. Dr. Hilde Van den Bulck

Antwerpen, 2013
2
Promotor
Prof. Dr. Hilde Van den Bulck

Doctoral committee
Prof. Dr. Alexander Dhoest
Prof. Dr. Hans Verstraeten

Doctoral jury
Prof. Dr. C. Lee Harrington
Prof. Dr. Charlotte De Backer

Copyright 2013 Nathalie Claessens

All rights reserved. No part of this publication may be reproduced in any form or by any means,
without written permission from the author.

Cover design: Nathalie Claessens

Printed by Universitas Digital Printing, Antwerp

ISBN 9789057284236
EAN 9789057284236

D/2013/12.293/16

Acknowledgments
This dissertation is part of the research project Celebrity activism: An empirical study of the
cooperation between social movements and Flemish celebrities, financed by an external four
year grant from the Flemish scientific fund FWO (application number: G.0315.09N) and an
internal four year grant from the University of Antwerp BOF-NOI funds, both obtained after a
competitive, peer-reviewed process.

nathalie.claessens@ua.ac.be | www.ua.ac.be/nathalie.claessens

3
 A Word of Thanks

My sincerest gratitude goes out to my promoter, Hilde Van den Bulck, who has been an
extraordinary guide throughout the past four years. She has taught me invaluable things
about being an excellent academic while being the most humane and kind boss that one
can imagine. Thank you, Hilde.

Further, I would like to thank the members of my doctoral committee: Alexander

Dhoest and Hans Verstraeten, who have helped me a great deal with their constructive
comments throughout the years. I also thank Charlotte De Backer and C. Lee Harrington
for being interested in my PhD and willing to be part of my doctoral jury.

This PhD would not have been possible without the University of Antwerp and
the funding source that has allowed me to work here during the past four years: BOF
(Bijzonder Onderzoeksfonds Universiteit Antwerpen).

I also thank my CW colleagues who have been essential in making the CW floor
not only the coolest floor of the Meerminne building but also the kindest. Special thanks
go out to Sil and Jasmijn, who introduced me to celebrity studies as a student, to Koen,
who helped me to get settled and become familiar with our project, to Nele, my
conference buddy, to Sara, for the pleasant walks to the station, and to all other
landscape and hallway colleagues.

Further, I want to express my gratitude to all the staff members and respondents
that have cooperated in the empirical studies of this PhD. In particular, I want to thank
the managers of both nursing homes for their warm welcome and cooperation. The
activities staff members of these nursing homes have also been very cooperative and
friendly which I am very grateful for. Most importantly, though, I thank the nursing
home residents who have been so nice to invite me in their rooms for interviews and the
residents who wanted to be part of the focus groups.

4
 Finally, I am extremely grateful to my parents, who have always supported and
stood by me since the day I was born. Mama and Papa, thank you for everything, I could
not have done this without you. The same goes for my sisters, Shana, Lise, and Nele:
Thank you for everything and especially for the fun evenings, weekends, and holidays. A
similar thanks goes out to my friends for making my free time both exciting and relaxing.
Here, Stijn has played a very important role, thank you! Lastly, I want to dedicate this
PhD to my grandfather, who was very proud when I started my first year as a PhD
student, but sadly, is no longer here to witness this moment.

Antwerp, June 2013

DEDICATED TO BOMPA
FRANS VAN NASSAUW

5
 Table of Contents

A Word of Thanks..................................................................................................................... 4

Table of Contents...................................................................................................................... 6

Introduction................................................................................................................................ 8

Theory ........................................................................................................................................ 16
POSITIONING THE PHD ........................................................................................................................ 16
Celebrity Studies ................................................................................................................................. 16
Cultural Studies ................................................................................................................................... 19
Communication Studies.................................................................................................................... 21
Psychology and Gerontology .......................................................................................................... 22
THE CELEBRITY APPARATUS............................................................................................................. 25
Conceptualizing Celebrity ................................................................................................................ 25
The Meaning of Celebrity for Audiences: An Academic Debate ........................................ 29
CELEBRITY AND MEDIA ....................................................................................................................... 30
Expansion of Celebrity News .......................................................................................................... 30
Celebrity and Entourage................................................................................................................... 32
CELEBRITY NEWS MEDIA AND AUDIENCES ................................................................................ 33
CELEBRITY AND AUDIENCES ............................................................................................................. 35
CONTRIBUTION OF PhD ....................................................................................................................... 37
REFERENCES ............................................................................................................................................. 39

Methodology............................................................................................................................ 44
COMBINING METHODS ......................................................................................................................... 44
FRAMING ANALYSIS ............................................................................................................................... 49
ONLINE SURVEY ...................................................................................................................................... 54
INTERVIEWS ............................................................................................................................................. 56
REFERENCES ............................................................................................................................................. 62

6
Chapter 1................................................................................................................................... 66
OF LOCAL AND GLOBAL FAME: A COMPARATIVE ANALYSIS OF NEWS ITEMS AND
AUDIENCE REACTIONS ON CELEBRITY NEWS WEBSITES PEOPLE, HEAT, AND HLN 66

Chapter 2................................................................................................................................... 90
CELEBRITY SUICIDE AND THE SEARCH FOR THE MORAL HIGH GROUND:
COMPARING FRAMES IN MEDIA AND AUDIENCE DISCUSSIONS OF THE DEATH OF A
FLEMISH CELEBRITY ............................................................................................................................. 90

Chapter 3................................................................................................................................ 114

GUESS WHO TIGER IS HAVING SEX WITH NOW? CELEBRITY SEX AND THE FRAMING
OF THE MORAL HIGH GROUND. ......................................................................................................114

Chapter 4................................................................................................................................ 134

PARASOCIAL RELATIONSHIPS WITH AUDIENCES FAVORITE CELEBRITIES: THE
ROLE OF AUDIENCE AND CELEBRITY CHARACTERISTICS IN A REPRESENTATIVE
FLEMISH SAMPLE .................................................................................................................................134

Chapter 5................................................................................................................................ 158

AN EXPLORATORY STUDY OF THE SOCIAL ROLES OF CELEBRITIES IN THE
EVERYDAY LIVES OF NURSING HOME RESIDENTS.................................................................158

Chapter 6................................................................................................................................ 188

NURSING HOME RESIDENTS AND CELEBRITIES: A TALE OF MORALITY......................188

Conclusion ............................................................................................................................. 212

ADDRESSING THE RESEARCH QUESTION...................................................................................212
Conceptualizing Celebrity ..............................................................................................................213
Celebrity as Commodity or Fulfilling Social/Cultural Roles? ...........................................219
ADDRESSING GAPS IN CELEBRITY STUDIES ..............................................................................226
ISSUES FOR FUTURE RESEARCH.....................................................................................................228
SUMMING IT UP......................................................................................................................................230
REFERENCES ...........................................................................................................................................231

Nederlandse samenvatting ............................................................................................ 234

7
 Introduction

Whether embodied in Angelina Jolies double mastectomy, Kim Kardashians pregnancy,

or Robert Pattinson and Kristen Stewarts tumultuous break-up, the celebrity
phenomenon is omnipresent in contemporary Western societies, including Flanders. It
is not only an important feature of the world of entertainment, but the underlying
mechanisms of celebrity culture also extend to almost all domains of society. Indeed, the
defining elements of celebrity culture and the resulting celebrities penetrate not just
the sector of entertainment, but politics, charity, business, and so on.

BUZZ: Has Tiger Woods Checked into Sex Rehab?

Cheryl and Ashley's marriage officially over today
Alexa Chung: I DONT have an eating disorder
Taylor Swift Admits Being 'Obnoxiously' Driven to Succeed
Monaco's Prince Albert to Marry
Leonardo DiCaprio Uses Social Networking to Lobby for the Environment
Kim Kardashian Has a New Man!
Brad Pitt & Angelina Jolie Sue Over Split Story
Tearful Osmond Family Remembers Marie's Son at Funeral
Brad Pitt has shaved off his horrible beard! HALLELUJAH!
(All quotes from the Heat and People websites in 2010)

The omnipresence of the celebrity phenomenon is mirrored by and closely related to the
ubiquitous nature of the media coverage hereof, as the titles above illustrate the
celebrity news that is disseminated across the globe day in day out. Celebrity coverage is
to a larger or smaller extent part of the content of all media, from the specialized press
gossip magazines, celebrity gossip websites, televised talk shows to the more
mainstream and even quality media, as almost all newspapers, magazines, radio and
television shows, news websites, blogs, and so on cover celebrities and their life events
on a daily basis. While some of this coverage is critical of celebrity culture, condemning
its dominance and other medias celebrity coverage, overall media embrace celebrities
as a worthy topic of attention, sometimes celebrating stars, at other times criticizing
their life choices. The omnipresence of celebrity culture and of the media coverage
hereof, in itself, is a firm basis and justification for this PhD.
8
 At first glance, much celebrity news appears quite trivial, characterized by
frivolity and superficiality, gossiping about celebrities appearance, personality,
professional successes and failures and the in and outs of their private lives. A quick look
at audience online reactions to average celebrity news items, seems to confirm this
triviality as audience responses seem equally superficial. This can be illustrated with a
typical item on the celebrity website Heat about Michelle McGee, the mistress of
Sandra Bullocks ex-husband Jesse James, who is apologizing to Sandra Bullock through
the media which leads to seemingly superficial audience reactions:

Michelle McGee says sorry to Sandra Bullock

As if that Michelle is still in her 20's she looks a lot older!
ugly twat ( that michelle one)
eww tattoos like that are disgusting!! :L (Heatworld, 2010)

This apparent triviality is one of the main reasons why the domain of celebrity studies is
still considered the most populist end of the popular in the academic world (Holmes
& Redmond, 2010: 3) and celebrity (news) a form of low culture in contemporary
Western societies. The example also seems to support the celebrity-as-commodity
thesis, which is part of a larger debate in celebrity studies on the social or cultural
meaning of celebrity. This thesis considers celebrity to be a consumer product that is
created by capitalist society to promote (and ensure the continued dominance of)
consumer culture and distract audiences from real life and societal issues. By diverting
audiences attention to the trivialities (appearances) of celebrities, real social debate can
be avoided and consumer culture remains a dominant force in society. In this PhD we
wish to explore to what extent and in which circumstances this thesis holds true.
However, the superficial reactions listed above are not the only ones that are
posted to the article, as other readers respond in a very different manner, expressing
their personal feelings with regard to the celebrities:

What a horrible situation. Poor Sandra.

cheap little hoe bag! Cant stand these cows that ruin families!
(Heatworld, 2010)

These highly personal and emotional responses indicate that the celebrity-as-
commodity thesis is not sufficient to explain the whole variety of audience reactions to
celebrity (news). It appears that there are more intense or complex relationships
involved in the way people deal with celebrities and the coverage hereof. Celebrity thus
9
appears to be more than mere entertainment or commodity as it evokes real emotions
among audience members that stem from a strong form of involvement with celebrities,
one that goes beyond mere distraction from real issues. In this PhD we wish to obtain a
better understanding of this relationship of general audiences with celebrities.
What is more, reading celebrity coverage, there are many articles that go well
beyond the trivial to include facts and opinions on social and moral themes. Likewise,
audiences often transcend the level of the superficial to discuss complex moral and
social themes such as adultery and publicity-seeking behavior:

Media article
As if having a prolonged affair with her husband isnt a bitter enough pill for
Sandra to swallow, she now has to endure listening to the other woman
talking about her marriage live on TV
Anyone else think Michelle is simply making herself feel better by
apologising? (Heatworld, 2010)

Audience reactions
Maybe she should have kept her trampy legs shut in the first place.......and
learnt to keep her mouth shut at the same time!
She said she's sorry it's so public but she didn't say she was sorry for fucking a
married man, skanky slagbag!
she's very sorry (on telly) no such thing as bad publicity .. (Heatworld, 2010)

These types of media content and audience reactions undermine a straightforward

celebrity-as-commodity thesis by demonstrating that, despite their apparent triviality
and beyond their role as entertaining commodities, celebrities can evoke discussions on
social or moral themes that can be considered real life and social issues. This too will
be further explored conceptually and empirically in this PhD.
Given the relevance of the celebrity phenomenon, not just as entertainment or
commodities but as serving social roles, in combination with its omnipresence in
contemporary Western societies, there is a need to better understand and to further
empirically explore this debate on celebrity culture. Therefore, the main research
question of this PhD is: What is the social or cultural meaning of the mediated
phenomenon of celebrity for contemporary audiences? This research question does
not only address the need to examine a phenomenon that strongly characterizes
contemporary Western media and societies but also the need to expand the academic
knowledge of celebrity culture and, especially, the way in which audiences make
meaning of celebrities (Turner, 2010). The focus of this PhD is thus on the relationship

10
between celebrity and audiences, which is essentially mediated. Indeed, important
actors in the construction of celebrity are the media who are the intermediary between
celebrities and audience members, as the latter two hardly ever have direct contact. This
is why the mediated aspect of celebrity embodied in celebrity news coverage is an
important element in this PhD research. Making a PhD always involves making hard
choices about what to include and what not. As such, the focus on the mediated
relationship between celebrity and audiences has come at the expense of an analysis of
the production side of celebrity culture which is not dealt with in this PhD.
The main research question of this PhD is addressed by means of six empirical
chapters that consist of studies that are (to be) published in academic journals or books.
The first three chapters examine the media coverage of and audience reactions to
celebrity (news) and the last three focus on audience members. The empirical studies
are included in the form that they are (to be) published in, preceded by a brief
introduction to situate the chapter with regard to the main research question. Only the
final chapter consists of an extended version of what will be published as a book
chapter.
The first empirical article is titled Of Local and Global Fame: A Comparative
Analysis of News Items and Audience Reactions on Celebrity News Websites People,
Heat, and HLN and includes a large-scale systematic, framing-based content analysis of
online media coverage and accompanying audience reactions on three celebrity
websites. Here, globalization is important, as contemporary celebrity culture is believed
to contain a combination of globally famous celebrities (e.g., the Angelina Jolies and
Brad Pitts of our world) and locally well-known celebrities (e.g., Flemish singers Koen
Wauters and Helmut Lotti). This study explores whether media and audiences react
differently to global and local celebrities and looks at the relevance of cultural proximity.
In chapter 2, titled Celebrity Suicide and the Search for the Moral High Ground:
Comparing Frames in Media and Audience Discussions of the Death of a Flemish
Celebrity, a case study is presented of the media coverage of and audience reactions to
the suicide of a Flemish lesbian celebrity. By conducting a combined qualitative and
quantitative framing analysis, a variety of viewpoints become apparent with regard to
several social and moral themes in the media articles and audience reactions: suicide,
divorce, homosexuality, and abuse. Interestingly, this study shows that audiences do not
just adopt media frames, but often develop counter frames based on their parasocial

11
relationship with the deceased celebrity, personal experiences, and interaction with
peers (the other readers).
Chapter 3 consists of a study titled Guess Who Tiger Is Having Sex With Now?
Celebrity Sex Reporting and the Framing of the Moral High Ground which includes a
framing-based analysis of the online media coverage of celebrity sex news and the
accompanying audience reactions. This study consists of two segments: the first part
comprises a quantitative content analysis of both media articles and audience reactions
on celebrity sex stories. The second part presents a qualitative case study of the theme
of adultery in which both the media coverage and audience reactions are subjected to a
framing analysis, resulting in the (re-)construction of seven adultery frames. Here, the
relevance of parasocial relationships, personal experiences, and interaction with peers
for the adoption of frames is confirmed.
The fourth empirical chapter is titled Parasocial relationships with audiences
favorite celebrities: The role of audience and celebrity characteristics in a representative
Flemish sample and consists of a quantitative online survey study. Here, the concept of
parasocial relationships that has been found to be important in the context of celebrity
news is explored further. Indeed, the nature of parasocial relationships is examined and
a new scale is developed. The concept of cultural proximity is found to be relevant for
the selection of favorite celebrities, especially for older and lower-educated audience
members. Further, parasocial relationships are found to be stronger among older, lower-
educated audience members with a high interest in celebrity news.
Chapter 5 An Exploratory Study of the Social Roles of Celebrities in the
Everyday Lives of Nursing Home Residents builds on the previous chapter as it
focuses on one particular group of audience members that has been found to have
strong connections with celebrities but is largely neglected in academic research on
media or celebrities: older adults. In particular, nursing home residents are the group
under study as they have a lot of free time, are limited in terms of mobility, and have
smaller social networks all factors which can increase their involvement with
celebrities. Indeed, the study in chapter 5 qualitatively explores the social and cultural
functions of celebrities for nursing home residents (and their celebrity preferences) by
means of individual and focus group interviews. This study found that celebrities can be
an easy lead into social conversations, a means to start moral or social discussions, a
form of social companions, and an aide-memoire.

12
 The final empirical chapter is titled Nursing Home Residents and Celebrities: a
Tale of Morality and focuses on one particular social role of celebrities for nursing
home residents (found in chapter 5), namely as a lead into moral or social discussions
and a means for meaning-making. By means of focus group interviews, in which
celebrity pictures are presented to nursing home residents, this study found that
celebrities can easily trigger discussions of moral or social themes. A framing analysis of
the discussions of adultery and homosexuality demonstrates the value of this social role
of celebrities in the nursing home.
In conclusion, these six chapters and the combination of methods in these studies
aim to provide a (relatively) comprehensive view of the meaning of celebrities in the
everyday lives of audience members. The order of the chapters quite accurately
represents the development of the PhD throughout the past four years. The only
exception is that chapter 2 preceded chapter 1 chronologically, but the order in the PhD
is more logical as chapter 1 contains a more general and systematic analysis and chapter
2 a more specific case study approach. The development of the focus of this PhD is also
represented well by the order of the chapters, starting with an interest in the content
and reception of celebrity news, going further into the concept of parasocial
relationships as it proved to be an important element in the framing analyses, and
ending with a focus on older adults as a specific group that has a strong connection with
celebrity but is largely neglected academically.
To properly situate the PhD in the academic world and present some of the main
theoretical concepts and viewpoints, the following part includes a broad theoretical
background sketch. Here, the PhD research is positioned in the academic field, by
relating it to the larger academic domains of celebrity and star studies, cultural studies,
communication studies, psychology, and gerontology. Afterwards, the main theoretical
elements of celebrity studies are outlined, with a new definition of celebrity that will be
empirically tested throughout the PhD and descriptions of each relationship within the
celebrity apparatus. This is followed by a methodological framework, in which the
combination of methods is discussed and motivated. Further, the different methods are
outlined, their (dis)advantages are presented, and the application in the PhD is
explained. Afterwards, the six chapters present the empirical studies and, finally, a
general conclusion will summarize the most important findings and link these to the
main research question.

13
REFERENCES

Heatworld (2010). Michelle McGee says sorry to Sandra Bullock. Heat [online]. 13 April.
Available at: http://www.heatworld.com/Celeb-News/2010/04/Michelle-
McGee-says-sorry-to-Sandra-Bullock/.
Holmes, S., & Redmond, S. (2010). A journal in Celebrity Studies. Celebrity Studies, 1(1),
1-10.
Turner, G. (2010). Approaching celebrity studies. Celebrity Studies, 1(1), 11-20.

14
15
 Theory

POSITIONING THE PHD

This PhD is predominantly situated within the domain of celebrity studies as it examines
the phenomenon of celebrity in contemporary Western societies. In line with the
multidisciplinary character of celebrity studies, multiple academic fields are combined
in this PhD which is very beneficial as it allows to address the main research question
from different angles, hereby providing a more comprehensive understanding.
Throughout the six empirical chapters, links are established between cultural studies
(the academic field in which celebrity studies is mainly rooted) on the one hand and
communication studies, psychology, and gerontology on the other. The specificities of
the positioning of this PhD and the linkages between the different academic fields are
described below.

Celebrity Studies
This PhD can be positioned, first, within the academic field of celebrity studies, a
domain that has been growing extensively in recent decades, especially in the 2000s and
2010s (Beer & Penfold-Mounce, 2010; Holmes & Redmond, 2010; Turner, 2010).
Although the first article discussing celebrity dates from 1958, the real expansion or
mainstreaming of the study of celebrity culture started in the 1990s and 2000s, with
more than half of the articles published until 2008 dating from 2002-2008 (Beer &
Penfold-Mounce, 2010). After the publication of Dyers (1979) book Stars, the academic
study of celebrity has become accepted within academia. In the 2000s, most of the key
books for celebrity studies have been published, enhancing the status of celebrity
studies in academia (e.g., Cashmore, 2006; Evans & Hesmondalgh, 2005; Giles, 2000;
Marshall, 2006; Rojek, 2001; Turner, 2004). In 2010, the first journal exclusively
committed to celebrity, Celebrity Studies, published its first issue and, in December 2012,
the inaugural Celebrity Studies Conference was hosted in Melbourne. The domain of
celebrity studies has thus witnessed an exponential growth during the last decade and

16
this evolution is expected to proceed as celebrity cultures and news presence continue
to increase in contemporary (Western) societies (Holmes & Redmond, 2010; Turner,
2010).
In the first issue of Celebrity Studies, Su Holmes and Sean Redmond (2010: 6)
describe the field as multidisciplinary, bridging with media sociology, psychology,
journalism, law, arts, literature, and sport studies, but having its most energetic hub in
media and cultural studies. This strong embeddedness in media and cultural studies
can be explained by the fact that academics already interested in popular culture and
representation have readily applied themselves to the discussion of particular
celebrities as texts (Turner, 2010: 12). As a result, most celebrity research consists of
case studies of particular celebrities or celebrity events as well as textual and discursive
analyses of celebrity (Turner, 2010: 13, 15). Both Holmes and Redmond (2010) and
Turner (2010) stress the need for complementary research, namely large-scale
systematic analyses in addition to case studies and studies of the production and
consumption of celebrity in addition to the text. Indeed, whereas celebrity can be
conceptualized as a genre of representation, a discursive effect, a commodity, and a
cultural formation, the focus of the current body of celebrity studies is on the first two
conceptualizations, which can be linked to the nature of cultural and media studies,
which focus on the textual (Turner, 2010: 13-14). This causes a clear gap in celebrity
studies, not only with regard to the production of celebrity, but also (and especially)
with regard to the consumption and social functions of celebrity: the aspects we
understand least at the moment [] are also the aspects about which we should be most
legitimately concerned in the long term (Turner, 2010: 14; see also Austin & Barker,
2003; Duits & van Romondt Vis, 2009; Feasey, 2008; Ferris, 2010; Gamson, 1994). There
is thus a need for research that examines celebrity audiences on the one hand and
celebrities social or cultural functions on the other. This is not just a task of American
Mass Communication Studies (as often assumed by cultural studies scholars), but also of
more cultural studies-oriented scholars, who conceptualize audiences as having a
degree of agency rather than being passive victims of media effects (Turner, 2010: 17).
A distinction should be made between the societal cultural or social functions of
celebrity and individual celebrity-audience involvement. On the one hand, the relevance
of celebritys social or cultural functions goes beyond individual audience members, as it
also includes social groups and society at large. On the other hand, the study of

17
individual audience members consumption or reception of celebrities is not limited to
the cultural or social functions of celebrities, but also comprises more psychological
(identification, parasocial relationships) and political economical (audience as
consumers and commodity) aspects. The current body of audience research in the
context of celebrity is mainly focused on fans, a very specific segment of the audience
(Duits & van Romondt Vis, 2009; see Hills, 2002; Sandvoss, 2005), which is why this PhD
wishes to address the larger not necessarily fan audience.
Within the relatively new domain of celebrity studies, this PhD wishes to play a
valuable role, filling in some of the existing gaps in research, by conducting audience
studies in the context of celebrity culture. More specifically, the social or cultural
functions of celebrity are examined from the perspective of individual audience
members, by analyzing the potential of celebrities to evoke social interactions, promote
moral discussions, and help meaning-making processes. In addition, one psychological
concept parasocial relationships is included in the PhD as it can be described as a
social or cultural role of celebrity. Further, the role of the media as the intermediary
between celebrity and audiences is considered, by analyzing the celebrity as a
representation (in celebrity news) and the accompanying audience reactions. Celebrity
scholars have often stressed the importance of social or cultural functions of celebrities
(Boorstin, 1961; Marshall, 1997, 2006; Rojek, 2001; Turner, 2004, 2006) but empirical
audience studies testing these assumptions are rare (Holmes & Redmond, 2010). This is
where this PhD wishes to contribute. In addition, the unique value of this PhD is not only
the study of social and cultural functions of celebrity from individual audience members
perspectives, but also the combination of systematic, larger scale analyses of celebrity
news and its reception with more profound case studies which allow in-depth insight
into the societal value of celebrity and celebrity news and the way audiences deal with
this. The inclusion of large-scale systematic analyses wishes to help fill the gap of
knowledge on the structural aspects of celebrity (news) (Turner, 2010).
When comparing celebrity and star studies, it is important to acknowledge the
important role of star studies such as Dyers (1979; 1986) to inspire other academics
and introduce the star or celebrity as a valuable subject of research. Star and celebrity
studies are united by their shared conceptualization of a star or celebrity as a discursive
construction (Holmes & Redmond, 2010). However, the studies of stars and celebrities
differ in multiple ways. First, star studies are ultimately focused on film stars, limiting its

18
applicability to contemporary celebrities who are often working in multiple domains
and to the contemporary celebrity landscape in which there is a convergence of media
forms which makes it difficult to distinguish various domains. Celebrity studies, in
contrast, conceptualize celebrity as a broad category which defines the contemporary
state of being famous (Holmes & Redmond, 2010: 4) and thus apply to the whole
variety of celebrity that exists now. Second, star studies are mainly interested in the text,
the representational interaction between the on/off screen persona (Holmes &
Redmond, 2010: 4), whereas celebrity studies (ideally) combine textual analyses with
that of production and consumption and focus on the private lives of celebrities rather
than their professional personae (Holmes & Redmond, 2010). Considering its scope, this
PhD can best be situated within the domain of celebrity studies, as it is interested in the
whole variety of celebrities in contemporary celebrity culture as well as the (media
coverage of and audience reactions to the) private lives of celebrity, more than their
professional representations.

Cultural Studies
This PhD can further be positioned within the domain of cultural studies as it (1)
adopts a social constructivist attitude and conceptualizes the audience as active, (2)
focuses on processes of meaning-making rather than media effects, (3) explores a form
of popular culture, and (4) examines the relationship between the text (celebrity and
celebrity news) and the reception or consumption by the audience within the broader
societal and cultural context. In other words, it is interested in how people make sense
of their social worlds and act on these understandings (Murdock, 1997: 90). Following
the tradition of cultural studies, culture can be described as an activity, a set of values
and practices, undertaken by particular people who live in particular lives in particular
setting and try to make sense of them, to reach particular goals, solve particular
problems, express particular sentiments (Gitlin, 1997: 25). In this regard, celebrity can
be conceptualized as a construction of both dominant culture (celebrity industry and
media) and the active audience (Marshall, 1997: 44-48).
In addition to the adoption of the concept of an active audience, this PhD can be
situated within cultural studies as it focuses on a form of popular culture celebrity
and its different aspects, more specifically, the (media) text and the audience within the
broader cultural and societal context. With regard to the text, celebrity news is

19
examined as a media text in this PhD. Although the cultural studies tradition often
considers popular culture or, in this case, celebrity to be a text in itself, this PhD focuses
on the mediated form of celebrity, namely celebrity news, as the textual aspect under
study. Despite the cultural studies orientation toward celebrity news, the analyses will
be conducted in a more communication studies-oriented style, namely by means of a
large-scale systematic framing(-based) analysis, combined with framing analyses of two
case studies (see infra). Given the clear lack of systematic analyses of celebrity in the
domain of celebrity studies (Turner, 2010), this PhD will prove to be a valuable
contribution. In terms of consumption or reception, this PhD investigates the
relationship between audience members and celebrity news (in the large-scale
systematic analysis and the two case studies) by comparing media and audiences and
looking at the role of cultural politics herein, namely whether audiences reject the
dominant media frames or viewpoints and introduce their own (Rojek, 2007). This is
mainly rooted in the ideas of Halls encoding/decoding model (1980) where polysemy
and actively giving meaning are central elements. Indeed, whereas the media aim at
encoding a certain frame or viewpoint, audiences can decode it in various ways: by
accepting the dominant media frame, negotiating between the media frame and their
own viewpoints, and rejecting the media frame and constructing a new frame.
The field of cultural studies has certain specificities, such as the traditional focus
on small-scale research which is limiting as it only considers a small segment of society
(Kellner, 1997: 116). This PhD wishes to move beyond this by conducting broader,
larger-scale (quantitative) research in chapters 1 and 4 (see further). Second, the notion
of the active audience is often romanticized in cultural studies, which is dangerous as it
may understate the power of the dominant media (and celebrity) culture (Kellner, 1997:
114; Morley, 1997: 121). Although the notion of an active audience is accepted
theoretically in this PhD, it acknowledges that the level of activity differs according to
the circumstances of media consumption which implies that audience members are not
always as active or critical while consuming media (Nabi & Oliver, 2009: 23).
To move beyond these specific characteristics of cultural studies, inspiration is
gathered in other academic fields, namely communication studies, psychology, and
gerontology. Methods (e.g., framing analysis), concepts (e.g., parasocial relationships),
and insights (e.g., characteristics of older adults) are borrowed from these disciplines,
but the main themes and interpretations remain within the perspective of cultural

20
studies. The bridging of these different domains with cultural studies in the multi-
disciplinary context of celebrity studies (Holmes & Redmond, 2010; Turner, 2010) is one
of the main contributions of this PhD.

Communication Studies
To provide a more systematic view of celebrity culture in the form of media coverage
and the accompanying audience reactions the academic domain of communication
studies is of importance here. Berger and Chaffee (1987: 17) define communication
science as a scientific domain which seeks to understand the production, processing
and effects of symbol and signal systems by developing testable theories, containing
lawful generalizations, that explain phenomena associated with production, processing
and effects. Communication studies are often discussed in contrast to media studies, as
the former are rooted in social sciences (cf. the dominant paradigm) and the latter in
cultural studies (cf. the alternative paradigm) (McQuail, 2005). These roots entail that
communication studies mostly consist of systematic, large-scale empirical (statistically
processed) studies, whereas media studies are typically more interested in small-scale,
case study approaches with a strong theoretical bias (cf. McQuail, 2005; Turner, 2010).
Combining the large-scale empirical focus of communication studies with the (mainly
theoretical) perspectives of cultural studies allows this PhD to go beyond the choice for
one domain and to complement qualitative, small-scale, case studies with quantitative,
large-scale research.
In particular, this PhD adopts one important concept from communication
studies for the study of celebrity (news), namely framing. Framing both as a concept
and a method is rooted and mainly used in the field of political communication and
rarely transferred to more cultural studies-oriented domains, such as celebrity studies.
However, the concept of framing and the method of framing analysis are very suitable to
conduct more systematic large-scale analyses of celebrity news. Indeed, the definition of
framing as the process by which a frame, on the one hand, indicates which elements
from a perceived reality are selected, excluded, stressed, and complemented in the news
production and, on the other hand, by which it provides receivers with a context and
suggested meaning (Van Gorp, 2004: 16) easily applies to celebrity news coverage.
Framing should be distinguished from processes of agenda setting and priming as, in the
case of framing, media do not just tell us what to think about, but also what to think

21
(Shah et al., 2009: 85). Communication science in general and framing research in
particular are often criticized because of a lack of attention to the cultural, political, and
economic context in which media production and consumption take place. This is why
this PhD aims at combining the systematic and large-scale nature of communication
studies and the main elements of framing (i.e., the reasoning devices) with the principles
and viewpoints of cultural studies, incorporating the wider context.

Psychology and Gerontology

Another domain that is relevant for this PhD is media psychology, especially because
certain psychological concepts have been demonstrated to provide an additional value
for the analysis of media reception (cf. Billig, 1997: 260). The bridging of cultural studies
and media or social psychology is not self-evident, considering the apparent
incompatibility of the domains, as a result of the broad societal view and qualitative case
study approach of the former and the focus on the individual and rigorous quantitative
systematic approach of the latter (McQuail, 2005). However, this PhD aims at
overcoming this strict distinction between the two academic domains by testing the
potential of a social-psychological concept for the study of active audiences and
incorporating the systematic nature of social psychology in the analysis of a cultural
studies-oriented topic, namely the relationship between celebrity and audiences. The
specific psychological concept that is relevant for research of celebrity culture is that of
parasocial relationships (Horton & Wohl, 1956; Giles, 2002, 2003). A parasocial
relationship can be described as an illusion of a long-term friendship between an
audience member and a celebrity, which is one-sided and maintained through the media,
but encompasses an emotional connection and is similar to social relationships.
Parasocial relationships are one of the forms in which audience members can establish a
connection or involvement with celebrities, next to identification (Cohen, 2009) and
fandom (Hills, 2002; Sandvoss, 2005). Parasocial relationships can play an important
role in the reception of celebrity news, as they filter the information-processing of
audience members and mediate the adoption of media frames (Brown, Basil & Bocarnea,
2003: 589, 601; Boon & Lomore, 2001; Klimmt, Hartmann & Schramm, 2006; Schiappa
Gregg & Hewes, 2005). This is what this PhD is interested in: how parasocial
relationships play a role in the reception of celebrity news and the meaning-making of
the news and broader social and moral themes. This PhD further wishes to help

22
compensate for the shortcomings in the current body of research into parasocial
relationships such as (1) the focus on television celebrities, (2) the lack of a clear
distinction between short-term parasocial interactions and long-term relationships and
the lack of a scale focused on the latter, (3) the use of student samples, and (4) the lack
of consideration of socio-demographic variables. The analysis of the parasocial
relationships between audience members and celebrities is conducted with the rigorous
mechanics of psychological research in mind (i.e., with regard to sampling, scale
development, and statistical analysis).
Finally, the sociological domain of social gerontology the study of aging from a
life course perspective is of interest here as the latter two articles in this PhD focus on
nursing home residents. In the early 1940s, the multi-faceted field of gerontology
started to develop and it has matured considerably over the past several decades
(Lowenstein, 2004: 129). Gerontology as an academic discipline is characterized by its
interdisciplinarity (Wilmoth & Ferraro, 2007) and the aim to understand the
heterogeneity of the older population, time-related changes on micro, mezo, and macro
levels (Lowenstein, 2004), and the links between biological, behavioral, and social
structure factors influencing human aging (Ferraro, 1990: 5). Gerontology typically
consist of three areas of research: (1) the physical aspects of aging, (2) the social aspects
of aging, and (3) public policy (Wilmoth & Ferraro, 2007). This PhD is mostly interested
in the second area of research, namely the social view on aging and the role of celebrity
herein. As a critique on the mainly negative view of aging (i.e., ageism) focusing on
decline in society and early academic work, contemporary gerontologists aim at
integrating both the losses and gains that come with aging in their research (Ferraro,
2007; van der Goot, 2009). The global aging of contemporary Western societies
increases the need for research on the older adult population and stimulates the
development of gerontology as an academic domain (Lowenstein, 2004; Wilmoth &
Longino, 2007) as well as the study of older adults in other domains. However, the older
adult population is still largely neglected in both media and celebrity studies which is
striking considering how contemporary Western societies are not just increasingly faced
with processes of global aging but also with an omnipresent celebrity culture
(Harrington, Bielby, Bardo, 2011). In addition, as older adults and, in particular, nursing
home residents are characterized by a large amount of leisure time, restricted mobility,
and decreasing social networks (Drageset, 2004), both media and celebrities are

23
expected to play especially important roles in their daily lives. This is why this PhD
innovatively bridges gerontology and celebrity studies and combines the insights and
concepts from gerontology with those from celebrity studies for the analysis of the
meaning of celebrities for nursing home residents. For instance, the fact that older adults
are characterized by both losses (e.g., physical and memory problems (van der Goot
2009)) and gains (e.g., life experiences, higher emotional complexity or mastery
(Carstensen, Fung, & Charles, 2003; Harrington, Bielby, & Bardo, 2011)) should be
incorporated in the study of the meaning of celebrities for nursing home residents.
In conclusion, this PhD is mainly rooted in celebrity and cultural studies, but aims
at bridging these domains with communication studies, psychology, and gerontology.
Specifically, this PhD will combine cultural studies main principles with the framing
approach from communication studies, the PSR concept from psychology, and the focus
on older adults (nursing home residents) from gerontology, to provide a broad
understanding of the meaning of celebrity (news) in the everyday lives of audience
members in contemporary Western societies.
After positioning the PhD in the academic field, the rest of this theoretical
framework will focus on the conceptualization of celebrity by distinguishing the defining
aspects and examining the construction of celebrity in a celebrity apparatus, consisting
of a famous person, his/her entourage, media, and audiences. Finally, the relationships
between these actors will be discussed and the applications in this PhD will be described
briefly.

24
 THE CELEBRITY APPARATUS

In this part, the theoretical background of celebrity culture will be discussed, by

reviewing existing literature on the definition of celebrity, the different actors that help
construct celebrity in the celebrity apparatus (see infra), and the relationships between
these actors.

Famous person Entourage

CELEBRITY

Audience Society Media

Conceptualizing Celebrity
There is large variety of conceptualizations and definitions of celebrity in both star and
celebrity studies and they share a number of elements which are incorporated in the
definition of celebrity in this PhD. The validity of this celebrity definition will be tested
throughout the empirical studies.
First, one of the main defining elements of celebrity is the aspect of well-
knownness (Boorstin, 1961: 57) or recognizability (Ferris, 2010) which implies that
celebrities are highly visible, famous, or well-known. The concepts of fame or well-
knownness are rather vague and subjective and there is no generally accepted threshold
which has to be crossed in order to be considered famous or well-known. Giles (2000)
description is the most detailed one as he describes fame as a degree of public
knowledge above what would be expected of an individual, given his or her social status
and the type of relationship network s/he would be expected to have (p. 109). Ferris
(2010) term of recognizability is more clear than those of fame or well-knownness as
she defines it as the experience of being recognized by far more people that one can
recognize back (p. 393). In a study on the mobility of fame, van de Rijt, Shor, Ward, &

25
Skiena (2013) measure fame as the number of appearances of that persons name in
newspaper records (p. 272). However, as celebrity coverage is not limited to
newspapers, but includes magazines as well as radio and television shows, online news
websites, and so forth, fame should be described as the number of appearances in media
coverage. Giles (2000) further distinguishes between fame and celebrity, where the
former can also apply to limited contexts (e.g., schools, cities) and the latter is explicitly
linked to the mass media.
Second, celebrities are conceptualized as constructions, the result of
negotiations within a celebrity apparatus (Holmes, 2005b; Marshall, 1997; Dyer 1979,
1986) The latter is conceptualized in this PhD as a complex, but relatively stable
trapezium-shaped relationship between a person aspiring celebrity, his/her entourage,
the media, and the audience. Holmes (2005b: 11) describes it as a triangle an arena of
negotiation between the star, the media and the audience but in this PhD, the triangle
is transformed into a trapezium shape, as the side of the star is divided into the famous
person (or the person aspiring fame) and the entourage because they fulfill different
functions within the apparatus. The four actors of the celebrity apparatus are
inextricably connected so that celebrity can only be constructed when all actors
participate in the process. However, there is a continuous tension between these actors
as their goals or modus operandi differ (notwithstanding some similarities) (Gamson,
1994: 81). They are thus situated in a negotiation position toward each other which
entails that the celebrity industry and culture can be considered the scene of constant
battles for control (Gamson, 1994: 85). Indeed, the celebrity needs the media to reach
audiences and needs audiences to maintain fame. In turn, the media need celebrities to
attract audiences and need audiences to remain viable. Finally, the audience needs
media to get information about celebrities (as there usually is no direct contact) and
needs the celebrity to provide entertainment and fulfill certain social roles (see infra). As
explained in the introduction, the studies in this PhD will not explicitly focus on the
production side of the celebrity apparatus namely the famous person, entourage, and
media axes because it would distract from the central research question.
Third, the celebrity construct typically consists of three personae: a public
persona which focuses on the professional activities of the celebrity films, music,
sports, TV, and so on , a private persona which consists of the official and managed
private life of the celebrity, and a real persona which provides insight into the off-

26
guard, unkempt, unready private life, as captured by, for instance, the paparazzi
(Holmes, 2005a: 24; see Dyer, 1986; Turner, 2004). Recent processes of
commodification and tabloidization have increased the emphasis of both media and
audiences on the private and, especially, real personae of celebrities, rather than their
public personae. In addition, celebrity coverage often employs a rhetoric of
authenticity to stimulate the audiences focus on the private and real celebrity personae
(Holmes, 2005a: 24; see also Dyer, 1986). The definition of celebrity in this PhD is linked
to the growing emphasis on the private and real personae of celebrities in contemporary
societies as it is conceptualized that a famous person becomes a celebrity when both
media and audiences pay (at least) as much attention to the private and real celebrity
personae as to the public persona. This is similar to Turners (2004) definition of the
celebritization process, where a focus on famous peoples private lives is the crucial
element to become a celebrity (cf. Geraghty, 2000; Gorin & Dubied, 2011; Holmes,
2005b; Holmes & Redmond, 2010; Van den Bulck & Tambuyzer, 2008). Here, processes
such as commodification a key process in celebrity culture (Cashmore, 2006: 72;
Turner, 2004) in which celebrities are transformed into consumer products and
tabloidization the growing dominance of soft over hard news and entertainment over
information (Johansson, 2006: 344) play important roles as these processes usually
ignore concepts of privacy and anonymity and, hence, turn celebrities into public
property (Cashmore, 2006; Cowen, 2000).
Fourth, celebrities are characterized by the ordinary-extraordinary paradox
(Dyer, 1979; Gorin & Dubied, 2011; Holmes, 2005a). This paradox entails that, on the
one hand, celebrities are similar to us normal people as they experience similar life
events (mainly in the private sphere: relationships, health, loss, death, ), while, on the
other hand they are special in terms of their talent, wealth, and lifestyle.
Fifth, celebrity is considered to be a mediated persona (Boorstin, 1961; Drake &
Miah, 2010; Duits & van Romondt Vis, 2009; Evans & Hesmondhalgh, 2005; Giles, 2000;
Gorin & Dubied, 2011; Holmes, 2005a, 2005b; Holmes & Redmond, 2010; Marshall,
2006; Rojek, 2001; Turner, 2004), as celebrities need media to reach their audiences
(which they almost never meet in person). In turn, media need celebrities to reach their
commercial purposes and create a loyal connection with their audiences (Drake & Miah,
2010; Turner, 2004).

27
 Combining these aspects in a definition of celebrity, it can be described as a well-
known mediated persona, constructed by the celebrity apparatus (consisting of the
famous person, entourage, media, and audiences), whose private and real personae
attract at least as much attention as the public persona. The value of this definition and
its different aspects will be examined throughout the PhD by testing its suitability in
light of the empirical findings of the six chapters. By empirically studying celebrity
(news) and audience reception, this PhD wishes to provide more information about the
criterion of well-knownness, the mediated character of celebrity, its constructed nature
and the relevance of the actors of the celebrity apparatus herein, as well as the different
personae and the prevalence of the private and real in a definition of celebrity. The
appropriateness of this definition will be discussed in the general conclusion at the end.
Several academics have differentiated between various types of celebrity.
Cashmore (2006) makes a distinction between achievement-based and media-driven
based celebrity and concludes that, in contemporary Western societies, the latter
prevails: doing is less important than being in the public gaze (p. 10). Giles (2000: 110-
116) distinguishes between public figures (whose fame is related to their role in
society), meritocratic fame (through merit), show business stars (whose fame is related
to their off-screen lifestyle and personality rather than their professional abilities), and
accidental fame (momentary fame). Monaco (1978) further makes a distinction between
a hero (whose fame is related to actively doing something), star (celebrities playing
themselves instead of a role), and quasars (who have no control over their image which
is fully constructed by the media). Finally, Rojek (2001) differentiates ascribed (fame by
birth), achieved (fame by merit) and attributed (fame by media attention) celebrity, with
a celetoid type as a short-term and concentrated form of attributed celebrity. With
regard to the different domains in which celebrities work, Marshall (1997) distinguishes
between film (emphasis on distance), television (emphasis on closeness), and popular
music celebrities (emphasis on familiarity). Turner (2010: 12) further states that
celebrity is infiltrating almost all domains in society, such as politics, literary publishing,
sport, and business. In addition, celebrity has become fundamental for contemporary
news media, not only for the popular forms that are traditionally associated with more
soft news and personalization of stories, but also for the quality media, who have
dismissed celebrity as trivial in the past but report on it in the present (even if only on
their websites) (Turner, 2010).

28
The Meaning of Celebrity for Audiences: An Academic Debate
With regard to the celebrity-audience relationship, scholars agree that celebrities need
an audience to be famous, but there is a disagreement about the meaning of celebrities
for audiences. Indeed, one of the central debates in celebrity studies is between the
proponents of the celebrity-as-commodity thesis and those who go beyond this thesis
and consider celebrity to serve social or cultural roles in society (Rahman, 2006). The
former group of scholars believes that celebrity is a commodity (a consumer product)
and a means to mask inequalities and distract from the real problems in capitalist
societies (Cashmore, 2006; Gamson, 1994; Johansson, 2006; Morin, 2005; Rojek, 2001).
By directing attention to the trivialities of celebrities, audiences can deal with their
capitalism-related frustrations in a relatively harmless manner, namely without
explicitly questioning the consumerist nature of society. In addition, celebrity culture
stimulates audience members to embrace consumer society and follow the capitalist
logic. Other scholars believe that celebrities are more than just trivial distractions from
capitalist problems as they also serve important social roles in society (Austin &
Barker, 2003; Boorstin, 1961; Butler Breese, 2010; Cashmore, 2006; Couldry &
Markham, 2007; Cowen, 2000; Duits & van Romondt Vis, 2009; Dyer, 1979, 1986; Evans
& Hesmondalgh, 2005; Feasey, 2008; Gorin & Dubied, 2011; Holmes, 2005a, 2005b;
Holmes & Redmond, 2010; Johansson, 2006; Marshall, 1997, 2006, 2010; Rojek, 2001;
Turner, 2004, 2006). First, celebrities can provide their audiences with food for gossip
and stimulate social conversations among audience members. Indeed, celebrity news
can elicit social interaction on a myriad of topics such as celebrities appearances,
personality, relationships, family life, professional activities, health, crime, and so forth.
A celebrity is an easy lead into social conversations as most people are familiar with
him/her (cf. criterion of well-knownness) and there are no real-life consequences when
criticizing or bad-mouthing the celebrity (Feasey, 2008). This turns celebrities into the
ideal gossip subjects. The social conversations about celebrities can also go beyond mere
gossip and touch upon more profound themes. Indeed, a second role of celebrities
consists of the potential to evoke social or moral discussions among audience members.
By discussing the shared but safe acquaintance embodied in the celebrity, audience
members can make meaning of the world and their own identity in a comfortable but
relatively harmless manner, for instance by freely discussing a commonly known
celebritys adultery or homosexuality. Third, celebrities can provide a form of social

29
companionship, in which the audience members are involved with and feel like they
know the celebrities. One type of involvement (and the most widespread) is the
development and maintenance of parasocial relationships (see definition supra).
Celebrities thus provide a variety of social roles for audience members (Alexander,
2010; Butler Breese, 2010; Dyer, 1979; Giles, 2002, 2003; Holmes, 2005a; Holmes &
Redmond, 2010; Marshall, 2006; Rojek, 2001; Turner, 2004, 2010). This PhD will look
into this debate by investigating the meaning of celebrity (news) in the lives of audience
members. On the one hand, this PhD will examine whether the celebrity-as-commodity
role and/or the social and cultural roles of celebrity are present in the empirical studies
of both media coverage and audience reception of celebrity (news). Can these
paradoxical roles co-exist or do they contradict (and eliminate) each other? On the other
hand, as the co-existence of these celebrity roles is expected (based on the literature in
celebrity studies that points to both roles), it may be interesting to distinguish between
the public (society) level and the private (personal) level. Indeed, whereas the celebrity-
as-commodity thesis is mainly focused on issues of a public level (the fundaments of
Western capitalist, consumerist societies), the (theoretically assumed) cultural or social
roles of celebrity are mainly situated in the personal or private sphere. This PhD thus
wishes to shine light on both the (co-)existence of the celebrity-as-commodity and the
social or cultural roles of celebrity as well as the levels on which they operate, either
public or private.

CELEBRITY AND MEDIA

Expansion of Celebrity News

This PhD is not only interested in the phenomenon of celebrities an sich but also in the
media coverage: the celebrity news stories. In the 21st century, celebrity content has
become a fundamental part of all news media and is no longer restricted to the
specialized press (Turner, 2010: 11; see also Austin & Barker, 2003). With the growth of
new media and online journalism, new ways are developed to represent, consume, and
produce celebrity (news). Interestingly, the so-called quality media mainly tend to
report on celebrities online rather than in their printed versions. Another development
coupled with the increase of celebrity news in media is the mainstreaming of celebrity
scandal stories and paparazzi photography, which can be linked to the increasing need

30
to capture the real celebrity, behind the faade (Turner, 2010; see also Cashmore,
2006; Holmes, 2005a).
The increase of celebrity news in all media can be explained by its excellent
suitability to the fast news cycles, the limited length of news stories and the preference
for entertainment in contemporary media (Turner, 2010: 11). Further, the growth of
celebrity news coverage can be related to a variety of additional changes in the news
media. One interesting revolution related to the increase of celebrity news is
tabloidization or the media turning to sensationalism, entertainment and the realm of
private affairs (Johansson, 2006: 344). Tabloidization is considered to be a broad
cultural movement, enabled by the increase of commercialization and the decrease of
traditional values, which is most visible in certain forms of media, for instance celebrity
news (Turner, 2006: 491). This process of tabloidization is combined with a growing
personalization of news and a preference for soft rather than hard news within news
values. The combination of these processes enables an exponential growth of celebrity
news in media coverage (Cashmore, 2006; Johansson, 2006; Turner, 2004: 75-76;
Whannel, 2010).
Critical views on tabloidization or the trivialisation of media content in general
(Turner, 2006: 491) describe it as a process where actual information is sacrificed for
entertainment and accuracy for sensation and where misleading techniques of
representation are employed to trap the audience members and exploit them which
ultimately leads to a dumbing-down of the public sphere (Turner, 2004: 71, 76, see
Gitlin, 1997). Franklin (1997: 4; see Holmes, 2005a) describes tabloidization in the
context of celebrities as the prioritization of the intimidate relationships of celebrities
at the expense of significant issues and events of international consequence. On the
other side, proponents of tabloidization state that this evolution is part of a demotic
turn which entails the mass medias emancipation from its obsession with the public,
the institutional and the masculine and a democratization of media access (Turner,
2004: 72). Further, Turner (2004: 77) states that tabloids personalization of the news
actually provides a more effective means of demonstrating the importance of the
political. One of the consequences of tabloidization is the development of a particular
form of celebrity journalism by the popular media which focuses on the details that
celebrities do not want to share themselves, namely the real persona (Austin & Barker,
2003; Johansson, 2006).

31
Celebrity and Entourage
Tabloidization can further be related to the growth of the celebrity industry. Indeed, it
entails that celebrity news has a larger share in news coverage which appears to
increase the need for celebrities to work with a manager, agent, or publicist (Gamson,
1994; Rojek, 2001). Celebrities no longer seem able to control media coverage
themselves and are in need of an entourage to promote and control publicity (Gamson,
1994: 63). In addition, commodification plays a central role in the development of
celebrity industries (Cashmore, 2006). Turner (2004: 31) describes the celebrity-
commodity as a famous person who is reduced to a consumer product by production
companies, managers, him- or herself, and the media (Marshall, 1997; Turner, 2004).
Gamson (1994: 62) further describes the star system as an industry in which agents and
managers sell celebrities and a successful sale is guaranteed by the loyalty of an
audience.
In the Anglo-Saxon world, several studies have been conducted with regard to the
celebrity industry and entourage (see Gamson, 1994; Cashmore, 2006; Turner, Bonner &
Marshall, 2006). These studies consider the early 1930s in Hollywood as a period of
industrialization of celebrity culture, with the development of professions aimed at
image management and of a broad and strongly controlled production system that
enabled a mass production of celebrities for a large group of consumers (Cashmore,
2006: 63). Turner, Bonner and Marshall (2006: 771-786) distinguish between three
roles within the celebrity entourage: (1) the agent (or talent scout) who functions as the
intermediary between the celebrity and employers, (2) the manager who has both
financial and persona-related responsibilities, and (3) the publicist who controls the
media coverage and gives the celebrity media training (see Gamson, 1994: 74). Turner,
Bonner and Marshall (2006) note that there are large differences between large
established celebrity industries (e.g., Hollywood) and relatively small and young
industries (e.g., Australia), where the structure of celebrity management of the latter is
far less developed than that of the former.
As stated above, the actors in the celebrity apparatus can be situated in complex
relationships and the connection between celebrity entourage and media is also one of
constant negotiations. On the one hand, media and entourage are interdependent: The
media guard the gates to exposure, the publicists guard the gates to access (Gamson,
2006: 704, zie ook Gamson, 1994; Evans & Hesmondalgh, 2005; Marshall, 2006). Both

32
actors would thus benefit from a symbiotic relationship (Gamson, 2006: 704) but this
is hindered by the fact that celebrities, entourages, and media have different goals: from
one point of view they [media] are simply predators, keen to exploit any item of
scandalous news to the full and at whatever cost to those concerned. From another point
of view, their [celebrities] commercial power makes them almost irresistible as the
quickest route to the public. (Turner, 2004: 75-76). Although this PhD does not
specifically examine the relationship between the celebrity, its entourage and the media,
it is important to keep these processes in mind as a broader context throughout the
analyses .

CELEBRITY NEWS MEDIA AND AUDIENCES

A second aspect of the celebrity apparatus is the relationship between the celebrity
news media and the audience. In addition to providing information, media are an
important source to help construct contemporary society and culture. Indeed, the
conflicts, tragedies, loves, desires, and struggles of celebrities which are presented to
the audience through media coverage can be a starting point to discuss these themes
in society (Alberoni, 1972, Cashmore, 2006; Cowen, 2000; Evans & Hesmondhalgh,
2005; Johansson, 2006; Marshall, 1997; 2006; Rojek, 2001). This is how the lives of
celebrities become part of audience members everyday lives and help them to
understand their own identity and the world around them: Warfare, tragedy, love,
desire, struggle, relationship all are mediated for us and implicated in our everyday
experiences [] They become part of who we are and how we understand the world
around us. (Lewis, 2008: 4). As stated above, the audience member as a news consumer
is highly dependent on the media to gather information on celebrities which turns media
into effective information filters: What do you actually know about celebrities? Only
what youve gleaned from the media, which act as effective filters on information
(Cashmore, 2006: 80; Lewis, 2008; Rojek, 2007). Indeed, as Rojek (2007: 71) states:
fans seldom meet celebrities, let alone form friendships with them, they rely on the
representations produced in magazines like Heat, Closer and Hello in the UK and, in the
USA, People, In Touch and Star; the celebrity gossip columns of tabloid newspapers;
celebrity bulletin boards; the news; biographies; and television and radio interviews.
Celebrity coverage is then considered to be an extension of a form of public therapy

33
culture in which ordinary men and women are encouraged to make sense of problems
and difficulties in their own lives by identifying with the experience of stars (Rojek,
2007: 73). In turn, media use celebrities to create loyalty with their audience and, hence,
reach their financial goals (Turner, 2004: 31). It should be noted that the audience is not
passively played by the media, but actively decodes the encoded celebrity news
messages (cf. Hall, 1980). This decoding can take three forms following the work by Hall
(1980), which can be translated to the framing approach as follows: dominant decoding
means adopting the media frame, negotiated decoding consists of adapting the media
frame to ones own attitudes and ideas, and oppositional decoding can be described as
rejecting the media frame and introducing ones own (counter) frame. Indeed, framing
consists of interactions between the textual level (media frames), the cognitive level
(audience frames or schemata), the extramedia level (frame sponsors), and the shared
cultural ideas in society as a macro-structure (Van Gorp, 2006: 247). Audience activity
can be defined differently according to the form of media consumption: active selection,
utilitarianism, intentionality, resistance to influences, or involvement (Biocca, 1988). In
this PhD, the three latter forms of activity will be examined: the cognitive processing of
media content, the adoption or resistance of dominant media frames, and the emotional
involvement with media content, namely celebrities. It is important to acknowledge that
the level of audience activity tends to differ, as audience members will be more or less
active with regard to different media content and in different circumstances: this does
not necessarily mean that viewers are consistently making meaning out of what they
watch (Gorton, 2009: 34; see Nabi & Oliver, 2009).
In this PhD, the relationship between media and audiences and, in particular, the
way in which celebrity coverage provides frames or viewpoints is examined by means of
framing(-based) analyses (chapters 1, 2, & 3). Here, the most valuable aspects of the
framing approach are what Entman (1993: 51) describes as the four functions of a
media text, coupled to four reasoning devices that can be found in the text: problem
definition, causal diagnosis, moral evaluation, and solution suggestion. These reasoning
devices are employed as the main principles for the framing analyses in this PhD.
Framing analysis can focus on the news production process, namely the construction of
media frames, or on the news consumption process, namely the reception and
negotiation of media frames and the introduction of audience frames (Van Gorp, 2004:
266; Scheufele, 1999). The latter is what this PhD is interested in: how media frames are

34
received by audience members and whether audiences adopt these frames or introduce
new ones to give meaning to (celebrity) news events. Here, a variety of factors is found
to mediate the relevance of media frames for audience members, such as personal
experiences, attitudes, interaction with peers, and so on (DAngelo, 2002; Iyengar, 1991;
Pan & Kosicki, 1993; Scheufele, 2004; Shah et al., 2009; Van Gorp, 2004). These factors
can be described as framing moderators as they affect the direction/strength of framing
effects, or define the conditions under which framing effects will or will not occur (Shah
et al., 2009: 90). The framing analyses in this PhD are also interested in these framing
moderators so these will be included in the analyses.

CELEBRITY AND AUDIENCES

As stated above, the axis between celebrity and audiences is mediated as audience
members rarely meet celebrities face-to-face. Most of the social roles that celebrities
fulfill for audiences and society are thus made possible by the media (Cashmore, 2006:
80; Lewis, 2008; Rojek, 2007). Celebrities are often considered to be trivial and
superficial, as mere entertainment or commodities (Cashmore, 2006; Marshall, 2006;
Rojek, 2001; Turner, 2004). However, celebrities have a variety of social roles for the
audience and society in general that go beyond the superficial or trivial (Alberoni, 1972;
Alexander, 2010; Butler Breese, 2010; Cashmore, 2006; Dyer, 1979; Giles, 2003; Holmes,
2005b; Holmes & Redmond, 2010; Marshall, 1997, 2006; Rojek, 2001; Turner, 2004,
2010). Indeed, celebrities can be easy leads into social interactions, as a shared
acquaintance, but without any real-life consequences when you criticize them (in
contrast to family and friends) (Feasey, 2008). Social conversations can thus be
stimulated by celebrity (news) as celebrities life events are generally shared knowledge
among audience members and can be a smooth way to start a conversation if there are
no other discussions points. The role of celebrity as a shared but distant acquaintance
further makes it possible for celebrities (or celebrity news) to trigger or evoke profound
moral discussions in which they represent what it is to be human (Dyer, 1998: 99). By
being the lead into moral discussions, celebrities give a broad moral theme a familiar
face which makes it easier to discuss the issue, once again without the risk of criticizing
and hurting anyone in real-life social networks (Feasey, 2008). Celebrities can evoke a
large variety of moral issues, as all their life events from their public, private, and real

35
personae are covered in the media going, for example, from money, talent, and
achievements, over the status of marriage, family issues, and health-related themes, to
drug abuse or coming out of homosexual celebrities. This is how celebrities can help
audiences make meaning of the world and their own identities (Alberoni, 1972,
Cashmore, 2006; Cowen, 2000; Evans & Hesmondhalgh, 2005; Johansson, 2006;
Marshall, 1997; 2006; Rojek, 2001). Further, celebrities can be a form of social
companions in the form of parasocial relationships between a celebrity and an audience
member. These parasocial relationships can be described as relationships in which
audience members have the illusion of friendship with a celebrity, which is ultimately
mediated and one-sided, but entails an emotional connection and is similar to social
relationships (see Chapter 4; Giles, 2002, 2003; Horton & Wohl, 1956). Celebrities thus
have multiple social roles and, hence, important meaning for audiences. These roles or
meanings are generally accepted on a theoretical level within the domain of celebrities
studies, but there is a scarcity of empirical research on the meaning of celebrity for
audiences (Holmes & Redmond, 2010; Turner, 2010). In the domain of fan studies,
empirical research has been conducted, but considering how fans are a specific segment
of the audience, with a very intense and emotional involvement and higher levels of
community and (textual) productivity (Fiske, 1992; Hills, 2002; Sandvoss, 2005), the
transferability of these findings to the general audience is limited. Therefore, this PhD
wishes to go beyond the focus on fans and include the broader not necessarily fan
audience to explore how they make meaning of celebrity. In addition, although there are
several empirical case studies that look at the broader audience in the context of
celebrity (Butler Breese, 2010; Feasey, 2008), there is a pressing need for more
(systematic) studies of these social functions, which is where this PhD wishes to
contribute by conducting a large-scale systematic analysis of celebrity news and
accompanying audience reactions (Chapter 1) and two case studies (Chapters 2 & 3), an
analysis of parasocial relationships (Chapter 4), and two studies on the social roles of
celebrity for a specific segment of society: nursing home residents (Chapters 5 & 6).

36
 CONTRIBUTION OF PhD

In conclusion, there are several gaps in celebrity studies which is not surprising
considering the fairly new character of the domain of which some are addressed in this
PhD. First, due to the dominance of textual analysis in celebrity studies, there is a lack of
audience research, in particular with regard to the social or cultural roles of celebrities
for audiences. This PhD focuses on the audience, while incorporating the wider celebrity
apparatus (celebrity & media), within the broader context of contemporary Western
societies. Second, celebrity research mainly consists of case studies which means that
there is a need for more large-scale systematic analyses. Here, the first empirical chapter
consists of a (quantitative) systematic analysis of celebrity news coverage and
accompanying audience reactions and the fourth empirical chapter includes a
(quantitative) analysis of a representative survey in the context of parasocial
relationships. To complement these systematic analyses with more profound
knowledge, two (qualitative) case studies are conducted as well as two (qualitative)
studies on the meaning of celebrity for one specific group in society, nursing home
residents. Third, due to the focus on fan and case studies in celebrity audience research,
several groups have been strongly neglected. This PhD wishes to contribute here by
focusing on a specific group in society, nursing home residents, whose main
characteristics indicate that the meaning of celebrities is especially important for them.
Finally, this PhD wishes to honor and strengthen the multidisciplinary character of
celebrity studies by combining the main principles and concepts from cultural studies
with methods, concepts and insights from framing research, psychology, and
gerontology.
In sum, this PhD addresses the part of the celebrity apparatus that is indicated by
the triangle in the figure below: the relationships between celebrity, media, and
audience. Here, the value of the celebrity concept (as defined above) will be examined
throughout the empirical studies to develop a more complex and reality-based view of
celebrity. In addition, the views of celebrity as a commodity or as fulfilling social or
cultural roles in society are both examined as to whether or not they are present in the
empirical studies and thus (co-)exist in celebrity (news) and audience responses.
Further, the levels (public/private) on which these roles are operating will be discussed
throughout this PhD, especially in the final conclusion.

37
 Famous person Entourage

CELEBRITY

Audience Society Media

38
 REFERENCES

Alberoni, F. (1972). The Powerless Elite: Theory and Sociological Research on the
Phenomenon of Stars. In D. McQuail (Ed.), Sociology of Mass Communications (pp.
75-98). Harmondsworth: Penguin Books.
Alexander, J. C. (2010). The Celebrity-Icon. Cultural Sociology, 4, 323-336.
Austin, T. & Barker, M. (2003). Contemporary Hollywood Stardom. London: Arnold.
Beer, D., & Penfold-Mounce, R. (2010). Researching glossy topics: The case of the
academic study of celebrity. Celebrity Studies, 1(3), 360-365.
Berger, C.R., & Chaffee, S.H. (1987). The study of communication as a science. In C.R.
Berger, & S.H. Chaffee (Eds.), Handbook of Communication Science (pp. 15-19).
London: Sage.
Billig, M. (1997). From codes to utterance: Cultural studies, discourse and psychology. In
M. Ferguson, & P. Golding (Eds.), Cultural Studies in Question (pp. 205-226).
London: Sage.
Biocca, F. A. (1988). The breakdown of the canonical audience. In J. Anderson (Ed.),
Communication Yearbook 11 (pp. 127-32). Newbury Park, CA: Sage.
Boon, S.D., & Lomore, C.D. (2001). Admirer-Celebrity Relationships Among Young
Adults: Explaining Perceptions of Celebrity Influence on Identity. Human
Communication Research, 27(3), 432-465.
Boorstin, D. J. (1961). The Image: A Guide to Pseudo-Events in America. New York:
Vintage.
Brown, W. J., Basil, M. D., & Bocarnea, M. C. (2003). Social influence of an international
celebrity: Responses to the death of princess Diana. Journal of Communication,
53(4), 587-605.
Butler Breese, E. (2010). Meaning, Celebrity and the Underage Pregnancy of Jamie Lynn
Spears. Cultural Sociology, 4, 337-355.
Cashmore, E. (2006). Celebrity Culture. New York: Routledge.
Cohen, J. (2009). Mediated Relationships and Media Effects: Parasocial Interaction and
Identification. In R. L. Nabi & M. B. Oliver (Eds.), The SAGE Handbook of Media
Processes and Effects (pp. 223-236). London: Sage.
Couldry, N. & Markham, T. (2007). Celebrity Culture and Public Connection: Bridge or
Chasm? International Journal of Cultural Studies, 10(4), 403-421.

39
Cowen, T. (2000). What Price Fame? Cambridge: Harvard.DAngelo, P. (2002). News
framing as a multiparadigmatic research program : a response to Entman. Journal
of Communication, 52 (4), 870-888.
Drageset, J. (2004). The importance of activities of daily living and social contact for
loneliness: a survey among residents in nursing homes. Scandinavian Journal of
Caring Sciences, 18, 65-71.
Drake, P. & Miah, A. (2010). The cultural politics of celebrity. Cultural Politics, 6(1), 49-
64.
Duits, L., & van Romondt Vis, P. (2009). Girls make sense: Girls, celebrities and identities.
European Journal of Cultural Studies, 12(1), 41-58.
Dyer, R. (1979) (reprinted 1998). Stars. London, BFI.
Dyer, R. (1986). Heavenly Bodies: Film Stars and Society, London: BFI.
Entman, R. M. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of
Communication, 45, 51-58.
Evans, J., & Hesmondhalgh, D. (2005). Understanding Media: Inside Celebrity.
Maidenheid: Open University Press.
Feasey, R. (2008). Reading heat: The Meanings and Pleasures of Star Fashions and
Celebrity Gossip. Continuum, 22(5), 687-99.
Ferraro, K.F. (1990). Gerontology: Perspectives and issues. New York: Springer.
Ferraro, K.F. (2007). The evolution of gerontology as a scientific field of inquiry. In J.M.
Wilmoth, & K.F. Ferraro (Eds.), Gerontology: Perspectives and Issues (pp. 13-34).
New York: Springer.
Ferris, K.O. (2010). The next big thing: Local celebrity. Society, 74(5), 392-395.
Fiske, R. 1992. The cultural economy of fandom, in The Adoring Audience: Fan Culture
and Popular Media, edited by L.A. Lewis. London: Routledge, 30-49.
Franklin, B. (1998). Newszak and News Media. London: Edward Arnold.
Gamson, J. (1994). Claims to Fame: celebrity in contemporary America. Berkeley:
University of California Press.
Geraghty, C. (2000). Re-examining Stardom: questions of texts, bodies and performance.
In C. Gledhill, & L. Williams (Eds.), Reinventing Film Studies (pp. 183-201).
London: Arnold.
Giles, D. C. (2000). Illusions of immortality: A psychology of fame and celebrity. Palgrave
Macmillan.

40
Giles, D. C. (2002). Parasocial Interaction: A Review of the Literature and a Model for
Future Research. Media Psychology, 4, 279-305.
Giles, D. C. (2003) Media Psychology. London: Routledge.
Gitlin, T. (1997). The Anti-Political Populism of Cultural Studies. In M. Ferguson & P.
Golding (Eds.) Cultural Studies in Question (pp. 25-38). London: Sage.
Gorin, V., & Dubied, A. (2011). Desirable people: Identifying social values through
celebrity news. Media, Culture & Society, 33(4), 599-618.
Gorton, K. (2009). Media Audiences: Television, Meaning and Emotion. Edinburgh:
Edinburgh University Press.
Hall, S. (1980). Encoding/decoding. In S. Hall, D. Hobson, A. Lowe, & P. Willis (eds.),
Culture, Media, Language: Working Papers in Cultural Studies (pp. 123-38).
Manchester: Hutchinson/CCCS.
Harrington, C. L., Bielby, D. B., & Bardo, A. R. (2011). Life course transitions and the
future of fandom. International Journal of Cultural Studies, 14(6), 567-590.
Hills, M. (2002). Fan Cultures. London: Routledge.
Holmes, S. (2005a). Off-guard, Unkempt, Unready?: Deconstructing Contemporary
Celebrity in Heat Magazine. Journal of Media and Cultural Studies, 19(1), 21-38.
Holmes, S. (2005b). Starring ... Dyer?: Re-visiting Star Studies and Contemporary
Celebrity Culture. Westminster Papers in Communication and Culture, 2(2), 6-21.
Holmes, S., & Redmond, S. (2010). A journal in Celebrity Studies. Celebrity Studies, 1(1),
1-10.
Horton, D., & Wohl, R. (1956). Mass Communication and Parasocial Interaction:
Observations on Intimacy at a Distance. Psychiatry, 19, 215-229.
Iyengar, S. (1991). Is anyone responsible ? How television frames political issues. Chicago:
University of Chicago Press.
Johansson, S. (2006). Sometimes you wanna hate celebrities : tabloid readers and
celebrity coverage. In S. Holmes, & S. Redmond (eds.), Framing Celebrity (pp. 343-
358). New York: Routledge.
Kellner, D. (1997). Overcoming the Divide: Cultural Studies and Political Economy. In M.
Ferguson, & P. Golding (Eds.), Cultural Studies in Question (pp. 102-120). London:
Sage.

41
Klimmt, C., Hartmann, T., & Schramm, H. (2006). Parasocial Interactions and
Relationships. In J. Bryant, & P. Vorderer (Eds.), Psychology of Entertainment (pp.
291-313). London: Lawrence Erlbaum Associates.
Lewis, J. (2008). Cultural Studies: The Basics. London: Sage.
Lowenstein, A. (2004). Gerontology coming of age: The transformation of social
gerontology into a distinct academic discipline. Educational Gerontology, 30(2),
129-141.
Marshall, P. D. (1997). Celebrity and Power: Fame in Contemporary Culture. Minneapolis:
University of Minnesota Press.
Marshall, P. D. (2006). The Celebrity Culture Reader. London: Routledge.
Marshall, P.D. (2010). The promotion and presentation of the self: celebrity as marker of
presentational media. Celebrity Studies, 1(1), 35-48.
McQuail, D. (2005). McQuails mass communication theory (5th ed). London: Sage.
Monaco, J. (1978). Celebrity. New York: Dell.
Morin, E. (2005 [1972]). The Stars. Translated by Richard Howard. Minneapolis:
University of Minnesota Press.
Morley, D. (1997). Theoretical orthodoxies: Textualism, constructivim and the new
ethnography. In M. Ferguson, & P. Golding (Eds.), Cultural Studies in Question (pp.
121-137). London: Sage.
Nabi, R. L., & Oliver, M. B. (2009). Introduction. In R.L. Nabi & M.B. Oliver (Eds.), The
SAGE Handbook of Media Processes and Effects (pp. 1-5). London: Sage.
Pan, Z., & Kosicki, G. M. (1993). Framing Analysis: An Approach to News Discourse.
Political Communication, 10(1), 55-75.
Rahman, M. (2006). Is straight the new queer? David Beckham and the dialectics of
celebrity. In P.D. Marshall (Ed.), The Celebrity Culture Reader (pp. 223-228).
London: Routledge.
Rojek, C. (2001). Celebrity. London: Reaktion Books.
Rojek, C. (2007). Cultural Studies. Cambridge: Polity Press.
Sandvoss, C. (2005). Fans. Cambridge: Polity.
Scheufele, B. (2004). Framing-effects approach : a theoretical and methodological
critique. Communications, 29, 401-428.
Scheufele, D. A. (1999). Framing as a theory of media effects. Journal of Communication,
49 (1), 103-122.

42
Schiappa, E., Gregg, P. B., & Hewes, D. E. (2005). The parasocial contact hypothesis.
Communication Monographs, 72(1), 92-115.
Shah, D. V., McLeod, D. M., Gotlieb, M. R., & Lee, N. J. (2009). Framing and Agenda Setting.
In R. L. Nabi & M. B. Oliver (Eds.), The SAGE Handbook of Media Processes and
Effects (83-93). London: Sage.
Turner, G. (2004). Understanding Celebrity. London: Sage.
Turner, G. (2006). Celebrity, the Tabloid and the Democratic Public Sphere. In P.D.
Marshall (Ed.), The Celebrity Culture Reader (pp. 487-500). London: Routledge.
Turner, G. (2010). Approaching celebrity studies. Celebrity Studies, 1(1), 11-20.
Turner, G., Bonner, F., & Marshall, P. D. (2006). Producing Celebrity. In P.D. Marshall
(Ed.), The Celebrity Culture Reader (pp. 770-798). London: Routledge.
van de Rijt, A., Shor, E., Ward, C., & Skiena, S. (2013). Only 15 minutes? The social
stratification of fame in printed media. American Sociological Review, 78(2): 266-
289.
Van den Bulck, H. ,& Tambuyzer, S. (2008). De celebritysupermarkt. Berchem: EPO.
van der Goot, Margot J. 2009. Television Viewing in the Lives of Older Adults. Nijmegen:
Radboud University.
Van Gorp, B. (2004). Frames in de nieuwsmedia. Ph.D dissertation, Department of
Communication Studies, University of Antwerp, Antwerp. [Frames in News
Media]
Van Gorp, B. (2006). Een constructivistische kijk op het concept framing. Tijdschrift voor
Communicatiewetenschap, 34 (3), 246-256.
Whannel, G. (2010). News, Celebrity, and Vortextuality: a Study of the Media Coverage of
the Michael Jackson Verdict. Cultural Politics, 6(1), 65-84.
Wilmoth, J.M., & Ferraro, K.F. (2007). The fountain of gerontological discovery. In J.M.
Wilmoth, & K.F. Ferraro (Eds.), Gerontology: Perspectives and Issues (pp. 3-12).
New York: Springer.
Wilmoth, J.M., & Longino, C.F. Jr. (2007). Demographic perspectives on aging. In J.M.
Wilmoth, & K.F. Ferraro (Eds.), Gerontology: Perspectives and Issues (pp. 35-56).
New York: Springer.

43
 Methodology

COMBINING METHODS

In this PhD, a combination of methods is used to obtain a more profound and

comprehensive understanding of the workings of celebrity culture in peoples everyday
lives. Further, considering the lack of systematic analyses in the domain of celebrity
studies, especially with regard to audiences, this PhD wishes to contribute by combining
large-scale systematic analyses with case studies and quantitative (framing analysis,
online survey, and scale development) with qualitative research (framing analysis and
interviews).
The methods in this PhD can be situated within the ontological paradigm of social
constructivism (versus objectivism), in which social categories and their meanings are
considered to be continuously molded by the actors, as people give meaning to the
world in social interactions (Berger & Luckmann, 1966; Bryman, 2012). Further, this
PhD adopts the interpretative epistemological approach with a focus on verstehen
(Weber) or understanding rather than mere explaining human behavior (cf. Bryman,
2012; Hijmans & Wester, 2006). In this approach, social reality is not considered to be
pre-given or external but actively constructed by social actors. The research strategies
employed in this research are a combination of inductive - developing theory on the
basis of the data and deductive strategies testing theory by means of data (Bryman,
2012). For each study, the most appropriate strategy with regard to the specific research
questions is employed. For instance, considering how framing research has mainly been
situated in political communication studies, which entails that there are very few frames
that can be applied to the context of celebrity news or the moral themes of adultery and
homosexuality, the framing analyses here are mainly inductive and construct frames on
the basis of the data. Further, the lack of research on older adults and celebrities makes
it more appropriate to conduct an inductive study as there are no real hypotheses to be
tested. On the contrary, the large body of literature with regard to the concept of
parasocial relationships makes it more appropriate to employ a deductive strategy,

44
testing hypotheses by means of (representative) data. In sum, the research strategy as
well as the design and specific method are always selected in such a manner that they fit
the research questions as well as possible. This PhD thus aims at overstepping the
boundaries of specific paradigms (e.g., cultural studies versus social sciences) and
combine the best-fitted methods for each research topic and question, while remaining
respectful towards the specificities and demands of each method (and paradigm).
In terms of design, this PhD combines (large-scale) cross-sectional, comparative,
and (small-scale) case study approaches (Bryman, 2012). The first study in this PhD
(chapter 1) is a quantitative cross-sectional, comparative content analysis conducted by
means of a framing-based analysis. The second study (chapter 2) is a combined
qualitative and quantitative case study in which framing analysis is employed. The third
study (chapter 3) consists of a quantitative cross-sectional and comparative content
analysis, with a quantitative framing-based analysis and a qualitative framing analysis of
a case study. The fourth study (chapter 4) comprises a cross-sectional design and uses a
large-scale quantitative survey method. The fifth study (chapter 5) employs a cross-
sectional design with a qualitative combination of in-depth and focus group interviews.
Finally, the sixth study (chapter 6) can be situated within a cross-sectional design in
which the selected method contains qualitative focus group interviews.
When situating the methods in the communication model, the framing analyses
combines the aspects of text and audience, while the survey and interviews focus
exclusively on the audience side. The element of culture is taken into account in all
methods, and thus in all articles of this PhD, to contextualize the research and link the
results to the broader contemporary Western culture and society.
The combination of these complementary methods is highly beneficial as it aids
to fill the gap of systematic analyses in celebrity studies and helps to shine light on the
different aspects of the meaning of celebrity culture for contemporary audiences. First,
the quantitative and qualitative framing(-based) analyses provide insight into the
media coverage of celebrities and accompanying audience reactions in one systematic
analysis (Chapter 1) and two case studies (Chapters 2 & 3). However, as the audience
reactions in these studies are limited to a selection of online reactions to three celebrity
websites in chapters 1 and 3 and to a selection of online and in-paper audience
comments in response to Flemish media coverage of a single case study in chapter 2,
other methods are needed to gain wider knowledge of the meaning of celebrities in the

45
everyday lives of audience members. This need for more general knowledge is why, in
chapter 4, a representative online survey is conducted with a thousand Flemish
respondents in which they are asked about their favorite celebrities and the parasocial
relationships they maintain with these celebrities. This quantitative analysis, which
includes the development of a new scale, provides more information about the meaning
of celebrities in audience members lives, namely whether it goes beyond mere
entertainment into the selection of favorite celebrities and the development of
parasocial relationships. This study then functions as a test of the prevalence of
involvement with celebrities in society (which is indicated by the previous three
studies) as well as an attempt to better understand the concept of parasocial
relationships. Indeed, by means of a newly developed scale, information is collected on
the nature of these parasocial relationships and the relative importance of audiences
and celebrities characteristics for its strength. As the survey data show that parasocial
relationships are stronger for (among others) older respondents, the latter part of this
PhD focuses on a group of people that is thought to have a higher-than-general media
use and, hence, potentially more interest in celebrities, but into which little research has
been conducted: older adults, and more specifically, nursing home residents. Since the
survey data are quantitative and provide little information on the deeper meaning of
celebrities for audience members and in particular nursing home residents ,
qualitative methods are used to bring more profound knowledge to the surface in
chapters 5 and 6. A combination of in-depth semi-structured interviews and focus group
interviews allows to gain insight into the meaning or social roles of celebrities in the
lives of nursing home residents from a qualitative point of view. In sum, this PhD
demonstrates how qualitative research (part of the framing analyses) can inspire
quantitative research (online survey), which in turn can inform further qualitative
research (interviews and focus groups) (cf. Bryman, 2012; Ritchie, 2003). This is exactly
where the value of a combination of methods lies, as the interplay between the different
methods both quantitative and qualitative provides a more comprehensive and
profound understanding of the research topic as it offers different types of intelligence
(Ritchie, 2003: 38). In addition, combining these methods brings new questions to the
surface, for instance the high degree of parasocial relationships among older people
stresses the need for research on this particular group with regard to celebrity.

46
 Regarding the relationship between media texts and audiences in the context of
the wider contemporary Western culture, the combination of quantitative and
qualitative framing analyses provides more profound insight both in the extent to which
frames are present in media coverage and audience reactions as well as in the societal
norms and values that are often the basis of the frames. Whereas the quantitative
framing analysis shows whether media frames are adopted by audiences or whether
counter frames are developed, the qualitative framing analysis can shine light on the
framing moderators, for instance, peoples personal experiences, interactions with
peers, or parasocial relationships with celebrities. Combining quantitative and
qualitative methods further allows us to examine the audience side of the
communication process from multiple angles. Indeed, the last three chapters help gain
an understanding of the relationship between (mediated) celebrities and audiences,
either in the form of parasocial relationships (chapter 4), the broad social roles of
celebrities for nursing home residents (chapter 5), or celebrities as leads into moral
discussions in the nursing home context (chapter 6).
For the study of celebrity in the nursing home context (chapters 5 and 6) a field
which is relatively unexplored in the existing body of research -, this PhD employs
qualitative approaches which are rooted in the principles of grounded theory. Grounded
theory is an explorative-descriptive approach which is ultimately aimed at
reconstructing meaning by means of a phased, iterative process (Glaser & Strauss, 1967;
see also Hijmans & Wester, 2006). It typically consists of four cumulative phases:
exploration (sensitizing concepts, developed on the basis of theory and previous
findings or experiences, are used to explore the issue), specification (dimensions and
variables are developed on the basis of the sensitizing concepts), reduction (determining
the main concepts for the research), and integration (developing the theory) (Glaser &
Strauss, 1967). The last chapters of this PhD do not include a grounded theory approach
as there is some literature that can inspire the study of celebrity and nursing home
residents -, but follow its main guiding principles. Indeed, the exploration phase in
chapters 5 and 6 starts developing sensitizing concepts on the basis of the existing
literature in both gerontology and celebrity studies. These sensitizing concepts are
brought into the practice of nursing homes by conducting paired or triad interviews
with the animation staff at the two nursing homes under study. In these interviews, the
relevance of the sensitizing concepts for the nursing home residents (according to the

47
staff members) is explored. Despite the bias that is created by asking the staff members
about nursing home residents experiences, interesting results are found in the paired or
triad interviews. With issues of power and bias in mind, the specification phase is based
on the findings of the interviews, combined with the literature of gerontology and
celebrity studies. Here, the sensitizing concepts are specified into a set of dimensions
social roles that celebrities can serve in the lives of nursing home residents. Semi-
structured interviews with the residents are then conducted to ask some initial
questions about the role of celebrities in their lives (e.g., by identifying favorite
celebrities, gauging their celebrity news interest and celebrity media use). The analysis
of the data gathered in these interviews allows to come to the reduction phase in which
the main research concepts are developed and selected. With these main concepts in
mind, focus group interviews with nursing home residents are prepared and conducted.
The findings of these focus group interviews allow to develop a theoretical view the
social roles of celebrities for nursing home residents on the basis of the data, namely
the integration phase, together with the data from the interviews with staff members
and the semi-structured interviews with the residents. The research processes of the
studies in chapters 5 and 6 are thus inspired by the guiding principles of grounded
theory, without explicitly consisting of a pure form of grounded theory (because of the
existing literature that can be transferred unto the research topic).
In the following sections of this methodological framework, the different methods
that are used in this PhD are discussed in terms of their main characteristics, advantages
and disadvantages, and the application in this PhD. The first three chapters are
concerned with the relationship between media coverage (text) and the accompanying
audience reactions (consumption). Instead of using the (in cultural studies more widely
accepted) method of discourse analysis, these studies employ framing analysis because,
in contrast to the former, the latter provides direct insight into the main elements of a
message (problem, cause, judgment, and solution) and this is especially valuable when
analyzing issues in the context of morality. Further, framing analysis allows for both
deductive (looking for previously developed frames in the data) and inductive
(developing frames from the data) and quantitative as well as qualitative research
(Reese, 2007; Scheufele, 1999). Finally, framing analysis can be employed for both the
analysis of media coverage and audience reactions (Scheufele, 1999). The final three
chapters of the PhD are focused on the audiences of celebrities and the combination of a

48
quantitative online survey and qualitative interviews and focus groups allows for a more
comprehensive understanding of both the concept of parasocial relationships and the
meaning of celebrity for nursing home residents. The focus on older adults and nursing
home residents in particular is a result of the findings of the study in chapter 4. Indeed,
as the online survey results showed that older adults tend to have stronger parasocial
relationships with celebrities, the interest in this particular group was raised.
Considering how little research has been conducted on older adults in the context of
media and celebrity which is striking in times of global aging this group became the
focus for the final two studies which are mainly inductive as a result of the limited
body of literature.
As stated above, the following segments include descriptions of the different
methods that are employed throughout this PhD in order of appearance in the studies
of this PhD: first, framing analysis (chapters 1-3), then online survey (chapter 4), and,
finally, interviews and focus groups (chapters 5-6).

FRAMING ANALYSIS

One influential way that the media may shape public opinion is by framing events and issues in
particular ways (de Vreese, 2005: 51).

In news research, the concept of framing has become increasingly popular in the last
decades (de Vreese, 2005). However, framing research has so far been almost
exclusively confined to the world of political communication (see de Vreese, 2005 for an
overview) which is why the application of framing to other realms here, popular
culture is one of the innovatory aspects of this PhD. The applicability of framing in the
context of popular culture is demonstrated by the successful adoption of framing
analysis as a research method to investigate celebrity news and the accompanying
audience reactions. To offer more insight into the concept of framing, an overview of the
main framing literature is provided below.
There are several definitions of framing (see Gamson & Modigliani, 1989; Gitlin,
1980; Goffman, 1974; Iyengar, 1991; Neuman, Just, & Crigler, 1992; Scheufele, 2004; Van
Gorp, 2004) which share several common aspects that are best combined in Entmans
(1993) definition: framing is a process whereby a frame suggests which aspects of

49
reality are selected, rejected, emphasized, or modified in the production of a media text
and, at the same time, provides the audience with context and suggested meaning. For
him, to frame is to select some aspects of a perceived reality and make them more
salient in a communicating text, in such a way as to promote a particular problem
definition, causal interpretation, moral evaluation and/or treatment recommendation
for the item described (Entman, 1993: 52). In his definition, the four functions of frames
are described problem definition, causal diagnosis, moral judgment, and remedy
suggestions which serve to highlight and select particular elements from reality to
convey an argument (Entman, 1993). In this PhD, the four functions of framing are
employed as the main guiding principles in the framing analyses. In this context, we turn
to Gamson & Modigliani (1989: 3) who introduce framing and reasoning devices as the
condensing symbols suggesting the core frame, displaying the frame package as a whole.
Framing devices consist of metaphors, exemplars, catchphrases, depictions, and visual
images. Reasoning devices are linked to the roots (causal analysis), appeals to principle
(moral claims), and consequences (effects) of a frame (Gamson & Modigliani, 1989: 3-4).
The framing and predominantly reasoning devices described above are the main
elements of the coding scheme in this PhD, for the analysis of both media and audience
frames.
Framing research entails a variety of advantages (Van Gorp, 2007). First, it
explicitly allows comparisons between media and audience frames and thus creates a
link between production and consumption processes (Carragee & Roefs, 2004: 215; Pan
& Kosicki, 1993; Van Gorp, 2007). In other words, it allows for the possibility to examine
both encoding and decoding processes (cf. Hall, 1980). In this regard, audience members
are found to both adopt media frames (the dominant reading) and develop counter
frames by reframing a message (negotiated or oppositional) (Entman, 1993). Second,
framing analysis provides direct insight into the essential elements of a message by the
use of reasoning devices which are explicitly linked to the four functions of framing,
namely problem, cause, moral judgment, and solution (Entman, 1993; Van Gorp, 2007).
As such, it provides a holistic view of the media and audience texts, rather than just
counting the amount of times a word is mentioned, and allows for inductive research
developing frames on the basis of the data especially in qualitative framing analyses.
Third, the deductive form of framing analysis is well-equipped to systematically deal
with larger datasets and can be applied to a variety of issues as it is a very flexible and

50
transparent method (Pan & Kosicki, 1993; Van Gorp, 2007). Inductive framing analysis
is not as suitable for large datasets because of the difficulty to construct frames on the
basis of strongly varied data. Fourth, framing analysis provides insight in the underlying
norms and values of a society as frames are always rooted in society (Van Gorp, 2007).
There are also certain disadvantages when using framing analysis. First, the
quality of the findings is strongly dependent on the quality of the selected data. This is
why the data for the framing analyses in this PhD are selected carefully, namely by
working with a sample of ten constructed weeks for three celebrity websites from the
US, UK, and Flanders for chapters 1 and 3. In the second chapter, the selected data
consist of all media coverage of the event in the most important Flemish media (both
online newspapers and magazines). Second, it is a highly subjective method as the (re-
)construction of frames strongly depends on the interpretation of the researcher (Van
Gorp, 2007). However, this disadvantage can be met by being reflexive about this
subjectivity and aiming at linking the frames to (moral) viewpoints in the society under
study and documenting the latter. Third, framing may seem to simplify media texts by
assigning one frame to a text, whereas one text may contain different frames (Matthes &
Kohring, 2008). In this PhD, this pitfall is compensated by not considering the entire text
as a unit for analysis, but rather distinguish between the different voices in a text and
attribute a frame to each voice. This allows to grasp the complexity of an item without
losing too much structure and comparability. Fourth, although the qualitative framing
analyses provide some information on the underlying mechanisms of frames, it is still
difficult to grasp the deeper meaning of celebrities for audiences. This is why interviews
are employed in chapters 5 and 6, namely to allow for a more profound understanding
of the meaning or social roles of celebrities. Fifth, as the data in the samples are almost
always found on celebrity websites, the analysis of audience reactions cannot be
generalized to the wider population as the use of online audience reactions only covers a
limited group in society (Van Zoonen, 2007). In addition, only a specific group of people
highly interested in celebrities visits celebrity websites and even a smaller portion is
likely to react on articles (Beddows, 2008). Moreover, we often do not know who is
behind the posting, why the individual sent in a comment or how postings relate to
peoples real-life politics (Van Zoonen, 2007: 535; see also Beddows, 2008; Wright,
2005). However, the latter two also occur in face-to-face communication while online
comments provide useful information about how people present themselves to others

51
(Van Zoonen, 2007: 535; Wright, 2005). In addition, as it is not our objective to
generalize our findings to the wider population, it is not that problematic for this
research. Still, it is important to be aware of the limitations of the selected methods.
In this PhD research, framing analysis is used in the first three chapters. In the
first chapter, featuring a broad analysis of online media coverage and online audience
reactions on celebrity news, a framing-based analysis is employed. The data for this
research consist of a sample of ten constructed weeks of celebrity websites HLN
Showbizz (Flanders), Heat (UK) and People (US) during the first ten months of 2010.
These websites are selected because of their high number of users and the opportunity
for online audience reactions. Further, the selection of these three websites allows
international comparisons of celebrity news and accompanying audience reactions, in
both a small local (Flanders) and two larger global (UK and US) celebrity industries. As
this large sample of 1,289 articles and 16,000 reactions includes a great variety of topics
or issues, it is practically impossible to do a proper framing analysis which is why a
framing-based analysis is employed instead. Here, the reasoning devices (problem,
cause, moral judgment, solution) related to the four functions of framing are used as the
main guiding principles for the codebook (see Chapter 1, Appendix 1) and the analysis.
This allows to maintain the benefits of framing without having to reconstruct proper
frames. In practice, each item (media article and audience reaction) is subjected to a
series of questions based on the reasoning devices: Who is discussed? (in terms of
celebrity name, gender, nationality, ethnicity, sexual orientation, and domain), What is
the issue? What caused the issue? How is the issue judged? and What is the
proposed solution? Further, a first look at the data showed that there are references to
peoples parasocial relationships with celebrities, personal experiences, and interaction
with peers (namely the other readers) in the audience reactions so these potential
framing moderators are also included in the topic list. The results are then analyzed and
a comparison is made between the treatment of local and global celebrities. The basic
tables for these analyses are not all included in the article itself, which is why they are
provided in Appendix 1 on the CD-ROM that is provided along with this PhD.
The second chapter, on the media coverage of and audience reactions to the
suicide of a Flemish celebrity, does include a proper framing analysis. The data
comprise of all Flemish media coverage and readers responses regarding the celebrity
death between June 25 2009 date of the celebrity suicide and September 25 2009

52
when news died down. The media sample contains the largest players in the Flemish
media landscape to provide a view as broad as possible, with two quality (De Standaard
and De Morgen) and two popular (Het Laatste Nieuws and Het Nieuwsblad) Flemish
daily newspapers, two Flemish celebrity gossip magazines Dag Allemaal and Story, and
magazine Humo. The corpus of audience responses contains all reactions as they appear
on the selected medias websites (respond to this article) and in the printed editions
(letters to editor). This study combines quantitative and qualitative work, with an
emphasis on the latter. In practice, each article is analyzed to see how many different
voices are present (e.g., journalist, celebrity, entourage, commentators) and the
segments for each voice are subjected to a range of questions, based on framings
reasoning devices: Who is the problem? What is the problem? What caused the
problem? What is the moral judgment? What is the proposed solution? (see Chapter 2,
Appendix 1). These answers are put in a matrix to look for similarities and patterns,
which are then combined into frame packages. Starting from the data, frames are thus
inductively and qualitatively (re-)constructed and analyzed in detail and linked to
viewpoints in society -, after which a quantitative analysis is conducted to see how often
the frames were mentioned in the media coverage and audience reactions. The
qualitative analysis shows that certain framing moderators are at work for the audience
members: personal experiences, interaction with peers (the other readers) and
parasocial relationships with the celebrity.
In the third chapter, there is a combination of a framing-based analysis for the
media coverage of and audience reactions to celebrity sex stories and a framing analysis
of celebrity adultery stories. The data for this study consist of a selection of the data in
the first chapter, namely all articles and reactions concerning sex in the sample of ten
constructed weeks throughout 2010 of the celebrity websites HLN Showbizz (Flanders),
Heat (UK) and People (US), resulting in a total of 51 articles and 685 reactions. The use
of the quantitative framing-based analysis is motivated by the large variety of issues or
topics discussed within this sample of celebrity and sex, making it difficult to
(re)construct different frames. This analysis employs a similar codebook as the first
chapter, with a focus on the reasoning devices (problem, cause, moral judgment,
solution) (see Chapter 3, Appendix 1). In this framing-based analysis, however, the focus
lies on the problems and moral judgments of the media articles and audience reactions,
as little information is provided with regard to causes or solutions. To gain more

53
profound knowledge, a qualitative framing analysis is conducted with regard to celebrity
adultery stories by far the most-discussed topic in the sex sample. A total of 17 articles
and 366 reactions are subjected to a qualitative, inductive framing analysis (following
the same procedure as in chapter 2). Here, the relevance of certain framing moderators
is once again confirmed, namely personal experiences, interaction with peers (i.e., other
readers), and parasocial relationships with the celebrities.

ONLINE SURVEY

As the first studies (chapters 1, 2, and 3) demonstrate the relevance of parasocial

relationships with celebrities, a quantitative online survey research is conducted in
chapter 4 to gain more systematic knowledge of this role of celebrity: do audiences have
favorite celebrities and do they maintain parasocial relationships with these celebrities?
In addition, we are interested in the factors (both celebrity and audience characteristics)
that correlate with the selection of favorite celebrities and the strength of parasocial
relationships. This study differs from the other studies in this PhD as it employs an
exclusively quantitative method and works deductively, by constructing survey items
and hypotheses on the basis of theoretical insights.
The choice for a survey is motivated by the fact that it, more than other methods,
provides valid and reliable findings on the occurrence of phenomena at one moment in
time with a multitude of subjects (Oomens, Scheepens, & Vergeer, 2006; Rea & Parker,
2012) . This is ideal for measuring the dispersion and relevance of audience members
favorite celebrities and parasocial relationships with these celebrities. The use of
standardized questions allows to measure a large amount of characteristics and easily
compare the subjects on the basis of background variables (Oomens, Scheepens, &
Vergeer, 2006; Rea & Parker, 2012) such as age, gender, education, and celebrity news
interest. Statistical analysis is used to process the data and, more specifically, to
construct a scale measuring the strength of parasocial relationships on the basis of
the survey data. Although it is not possible to measure actual behavior in surveys, it is a
good method to examine attitudes, opinions, feelings, motives, and so forth (Oomens,
Scheepens, & Vergeer, 2006;). The main advantages of an online survey are its fast speed
and low cost (Oomens, Scheepens, & Vergeer, 2006; Wright, 2005). In addition,
respondents can fill in the survey whenever they want and at their own pace. The use of

54
online surveys also entails some disadvantages as they usually have a lower response
rate and there are more issues with anonymity and confidentiality (than face-to-face,
telephone or paper surveys) and are limited to the online population, omitting large
groups in society (Wright, 2005). However, these pitfalls are outweighed by the
advantages provided by online surveys (as described above) and the use of a research
facilitator (iVOX) helps to address the first two pitfalls. The limitation to the online
population is something that should be acknowledged and kept in mind throughout the
analysis.
Considering the advantages described above, survey research can be considered
an adequate method to study parasocial relationships with celebrities. Some pitfalls
when conducting surveys consist of the fact that they are less suited to examine causal
questions (Rea & Parker, 2012). However, as this particular study is not interested in
causal questions (for instance, the influence of parasocial relationships with celebrities
on media enjoyment) but in the nature of parasocial relationships and the correlation
with both respondents (age, gender, education, and celebrity news interest) and
celebrities characteristics (gender, nationality, and domain), this is not problematic
here. Further, standardized instruments may not be available for all research topics.
This does apply to the study in chapter 4, as there is no appropriate measure to examine
parasocial relationships. This is why a new scale is developed by combining items from
existing scales on parasocial interaction and parasocial relationships which are
valuable but not fully adequate to measure long-term parasocial relationships with a
variety of celebrity with new items.
To conduct a representative survey, researchers should ensure both a high
population validity and external validity (Rea & Parker, 2012). Both are accomplished in
the fourth chapter by obtaining a representative sample of the adult population in
Flanders (northern part of Belgium). The representativeness of the survey is ensured by
administering iVOX, a Belgian research facilitator, for the distribution of the survey.
Here, a random sample of the Flemish population was contacted by iVOX and filled in the
survey. The final respondents are 1000 Flemings, consisting of 501 men and 499
women. The 30-49 age group is represented most often (40,9%), followed by people
over 50 (39,0%) and people from 18 to 29 (20,1%). 27,0% of the sample had a higher
education, while 35,8% has a higher secondary diploma and 37,1% a lower secondary or
less. Our sample is representative of the Flemish population which consists of 49.3%

55
men and 50.7% women (FOD Economie, 2010a); 14.4% Flemings are between 18 and
29, 28.7% is between 30 and 49, and 37.1 is older than 50 (FOD Economie, 2010b);
27.3% had a higher education, 33.2% a higher secondary, and 39.4% a lower secondary
education or less (FOD Economie, 2010c). The detailed procedures of the development
of the online survey data and the scale as well as the accompanying analyses are
provided in chapter 4.

INTERVIEWS

Interestingly, the study in chapter 4 demonstrates that older adults tend to have
stronger parasocial relationships with their favorite celebrities than their younger
counterparts. Combined with the global aging and omnipresent celebrity culture in
contemporary societies, there is a pressing need for research on the meaning of
celebrities for older adults, which is why the latter part of the PhD explores this topic.
Instead of looking at the general group of older adults, the specific segment of nursing
home residents is focused on because they tend to have more free time, are less mobile,
and have smaller social networks than independent-living older adults (Drageset, 2004).
These factors are often linked to higher media use and, consequently, more occurrences
in which the nursing homes residents come in contact with celebrities. Qualitative
methods are ideal to examine this particular topic as they are ultimately aimed at
describing or interpreting an issue in its specific, day-to-day environment and at gaining
insight in the meanings that it entails for the research subjects (Baarda, de Goede &
Teunissen, 2005; Bryman, 2012). The research design is typically flexible and aimed at
studying the natural environment in which the phenomenon takes place, understanding
it holistically within the larger context.
Qualitative interviews are employed as a method here as we are interested in the
meaning of celebrities for the specific group of nursing home residents. Further, the
choice for interviews is motivated by the fact that is puts the perspectives of the
research subjects central instead of those of the researcher and the limited structure of
interviews allows for great flexibility (Bryman, 2012; Snape & Spencer, 2003). It also
provides very rich information, for instance by the fact that peoples natural word use
comes to the surface. It is important to acknowledge that meaning is constructed
throughout the interview and thus the role of the researcher should be reflected upon

56
(Bryman, 2012; Snape & Spencer, 2003). The latter is also one of the pitfalls that come
with using qualitative interviews, but as long as the centrality of the researcher is
discussed in the research and there is transparency about the research process, this is
not problematic. Other disadvantages of the method of qualitative interviews are the
risk to create a detailed plan beforehand, hereby limiting the flexibility of the interviews,
the practical difficulties due to the lack of structure, the risk to fragment the data and
losing sight of the holistic approach and the context, the fact that is does not always
result in the development of theories and that there are many different ways in which to
conduct interviews, rendering it a subjective method (Bryman, 2012; Snape & Spencer,
2003). These pitfalls are avoided by respecting the flexible nature of interviews, using
the analytical steps of open, axial, and selective coding to provide some structure
(Strauss, 1987), the awareness of the wider context throughout the analyses to allow for
a holistic approach, the focus on the interplay between the findings and theory, and,
finally, the recognition of the subjectivity of interviews and transparency about the
procedures and process.
The sampling in the last two chapters of this PhD is a form of theoretical or
purposive sampling in which relevant respondents are selected who represent certain
characteristics of interest for the study (Bryman, 2012; Glaser & Strauss, 1967; Hijmans
& Wester, 2006). Indeed, the sample for both studies consists of residents of two nursing
homes, one in a rural area (Essen) and one in an urban area (Antwerp) in Flanders. The
rural nursing home is selected because of the connections of the researcher as her
mother works at the nursing home, her grandmother lives there, and she lives in Essen
herself which facilitates the access to the nursing home (cf. Tulloch, 1991). The
manager of the rural nursing home in Essen further helped to find an urban nursing
home in Antwerp that is similar to the one in Essen in terms of size and population.
Three different forms of qualitative interviews are used in this PhD: individual in-
depth semi-structured interviews (chapter 5), paired or triad interviews (chapter 5),
and focus group interviews (chapters 5 & 6) (Ritchie, 2003; Lewis, 2003). The individual
interviews are semi-structured in the sense that there is an interview guide, containing a
topic list or protocol to create a certain level of structure without losing the flexibility of
qualitative interviews (Hijmans & Wester, 2006). The level of structure is important for
the analysis as it promotes comparability of the different respondents. The choice for
interviews instead of surveys is also motivated by the specific characteristics of the

57
older adult group, where issues with writing, concentration, memory, and so on make it
difficult to let them fill in a survey by themselves (Barrett & Kirk, 2000). When
conducting individual interviews, it is important to make it as natural as possible by
creating a dialogue and showing empathy for the situation and life world of the
respondents (Bryman, 2012; Legard, Keegan, & Ward, 2003). To obtain high quality data
on the basis of semi-structured individual interviews, researchers should construct a
logical order by which to guide the respondents (without losing flexibility) and make the
employed concepts understandable for the respondents. Semi-structured interviews
should further be tested in trial interviews, on the basis of which adaptations can be
made (Hijmans & Wester, 2006). Finally, interviews should be audio-recorded and
transcribed in order to start coding the data profoundly. The coding procedures are also
based on those in grounded theory, starting with an explorative phase of open coding on
the basis of the sensitizing concepts, followed by a specification phase of axial coding in
which codes become more specific and patterns are examined, and finally, a phase of
reduction and/or integration in which the coding is more specified based on the main
concepts of the study (Strauss, 1987; see also Hijmans & Wester, 2006). This process is
cumulative and iterative so that concepts are developed scrupulously. In addition to the
transcripts, memos taken during the interviews can be of assistance in the coding and
analysis process. The use of individual interviews allows researcher to gain new insights
as well as more profound information on the underlying meanings of phenomena
(Lewis, 2003). The main disadvantage of individual interviews is the subjectivity as the
researcher functions as the research instrument (Bryman, 2012; Hijmans & Wester,
2006). However, this risk can be solved by creating a systematic and transparent form of
structure for instance, by using topic lists or protocols which allows for the
interviews to be reproduced by other researchers with other respondents. In addition,
the method of individual interviews is time-consuming, not only in terms of conducting
interviews but also in terms of transcription, coding, and analysis (Hijmans & Wester,
2006; Ritchie, 2003). This disadvantage is outweighed by the benefits of using
interviews for this PhD. In chapter 5, twenty residents participated in the individual
interviews for each nursing home, consisting of (in total) 35 women and 5 men between
the ages of 68 and 101 (M=85), which is representative of the nursing home population
in Flanders (Vanden Boer & Pauwels 2005). Due to practical considerations and the
residents own willingness to participate, these are the most healthy residents with

58
relatively good mental capacities, which is often the case in nursing home research
(Hajjar, 1998). The procedures of the interviews are described in detail in chapter 5.
A distinction should be made between paired or triad interviews and focus group
interviews. Whereas the former are similar to in-depth interviews, with the exception
that they are conducted with more than one person simultaneously, the latter differ
from individual interviews as they are focused on one theme and ultimately aimed at
creating and examining social interaction. In the fifth chapter of this PhD, two paired or
triad interviews are conducted with the animation staff members of each nursing home.
For the rural nursing home, three female staff members are interviewed and two (one
male, one female) for the urban home. As stated above, these interviews served an
important role in the specification phase putting the sensitizing concepts found in the
literature in practice and provided a first (but biased) insight into the meaning of
celebrities for nursing home residents. In chapter 5, the details of these interviews are
provided.
Focus group interviews allow researchers to gain insight in the collective
dimensions of human action (Bryman, 2012; Lewis, 2003; van Selm & Wester, 2006) and
it can be defined as a research technique that collects data through group interaction on
a topic determined by the researcher. It is the researchers interest that provides the
topic, whereas the data themselves come from the group interaction (Morgan, 1997).
Focus groups are selected here as the research focuses on a phenomenon that takes
place on a social, group level. Indeed, the meaning of celebrities is not only constructed
by the individual audience member but also in interaction with other people, for
instance other nursing home residents, visitors (family, friends), or staff members. The
sampling method here is theoretical with a focus on heterogeneous and natural groups
(i.e., people living in the same nursing home) because the aim is to explore the
differences in meaning of celebrities as well as to (re-)create natural interactions. A
topic list is used in the focus groups as well as stimulus material in the form of celebrity
pictures. These are standardized to ensure and enhance replicability as well as
comparability. All focus groups are audio-recorded and later transcribed. Memos are
taken during the focus groups by the researcher and a student who assisted with the
focus groups and later transcribed them. The coding process is similar to that for the
individual interviews with phases of open, axial, and selective coding (Strauss, 1987). In
the analysis of the focus groups, special attention is paid to shared versus conflicting

59
opinions within the groups. The use of focus groups can be motivated by the fact that
they in contrast to participatory observation allow for a lot of interactions about a
topic in a limited time frame, provide more insight in the attitudes and decision-making
processes, and, in the case of focus groups, it is less difficult to get access to the field (van
Selm & Wester, 2006). Further, focus group interviews provide information about
meaning and social interactions, are informal, and people do not have to read or write
anything which is often a difficulty for older adults and in particular nursing home
residents. In addition, focus groups can be conducted with traditional non-respondents,
such as older adults or nursing home residents and it is relatively easy to get into a field
through focus group interviews (Barrett & Kirk, 2000; Bloor et al. 2001; Morgan 1997).
Focus groups benefits further include the fact that new ideas can be generated because
of the flexibility and that researchers get a clear view on attitudes and conflicting or
shared opinions in the groups (van Selm & Wester, 2006). Finally, focus groups are
selected here because it provides insight in the processes of group interaction in which
opinions and meaning are built (Lewis, 2003). However, the use of focus groups also
entails several disadvantages such as the fact that the data are limited to the verbal and
non-verbal statements of the respondents rather than their actual behavior and to group
interactions rather than individual opinions (van Selm & Wester, 2006) but this is not
problematic as the studies in chapters 5 and 6 are interested in the meaning of celebrity
rather than residents behavior and in social interactions rather than individual views.
The latter are covered in the individual interviews in chapter 5. In focus group
interviews, group effects such as conformity or polarization can cause a gap between the
silent and loud members and may cause people to express more culturally excepted
opinions (van Selm & Wester, 2006). This is something which should be acknowledged
and kept in mind throughout the analyses. A focus group typically is a created artificial
situation and allows for less depth than individual interviews which makes it difficult to
investigate taboo issues (van Selm & Wester, 2006). The former is true for every
research method as research almost always intrudes in everyday life and is never fully
natural and the latter is compensated by combining focus groups with individual
interviews in chapter 5. Further, researchers have less control over focus group
interviews, compared to individual interviews, and they are hard to set up as people
have to show up (van Selm & Wester, 2006). The lack of control is acknowledged by the
researcher but is also one of the strengths of the focus group method as it allows for

60
real group interactions. Further, in chapters 5 and 6, the hard-to-set-up pitfall is
avoided by conducting the focus group interviews in the nursing homes. Focus group
interviews are very time-consuming, for instance the transcriptions take longer but, as
stated above, this is outweighed by the valuable information that is provided by this
method. For chapters 5 and 6 of this PhD, four focus groups are conducted with six to
seven nursing home residents and they lasted approximately one hour and a half. The
participating residents live in the two nursing homes described above. Only the
residents who are able and willing to participate in the research are selected, resulting
in a total of 27 residents (13 for the rural and 14 for the urban nursing home) with an
average age of 85,6 (ranging from 65 to 97). There was a large overlap between the
participants in the semi-structured individual and focus group interviews. The practical
details of the interviews and analyses in chapters 5 and 6 are discussed in the chapters
themselves. The Dutch transcriptions of the focus group interviews are included in
Appendix 2 that can be found on the CD-ROM provided with this PhD.

61
 REFERENCES

Baarda, D. B., de Goede, M. P. M., & Teunissen, J. (2005). Basisboek kwalitatief onderzoek:
handleiding voor het opzetten en uitvoeren van kwalitatief onderzoek. Groningen:
Stenfert Kroese.
Barrett, J., & Kirk, S. (2000). Running Focus Groups with Elderly and Disabled Elderly
Participants. Applied Ergonomics, 31, 621-29.
Beddows, E. (2008). The methodological issues associated with Internet-based research.
International Journal of Emerging Technologies and Society, 6(2), 124-139.
Berger, P. L., & Luckmann, T. (1966). The social construction of reality: A treatise in the
sociology of knowledge. New York: Anchor Books.
Bloor, M., Frankland, J., Thomas, M., & Robson, K. (2001). Focus Groups in Social Research.
London: Sage.
Bryman, A. (2012). Social research methods (4th ed). Ofxord: Oxford University Press.
Carragee, K. M., & Roefs, W. (2004). The neglect of power in recent framing research.
Journal of communication, 54, 214-233.
de Vreese, C. H. (2005). News framing: Theory and typology. Information Design Journal
+ Document Design, 13(1), 51-62.
Drageset, J. (2004). The Importance of Activities of Daily Living and Social Contact for
Loneliness: A Survey among Residents in Nursing Homes. Scandinavian Journal of
Caring Sciences, 18, 65-71.
Entman, R. M. (1993). Framing : towards clarification of a fractured paradigm. Journal of
Communication, 43(4), 51-58.
FOD Economie (2010a). Bevolking van Belgische en vreemde nationaliteit naar geslacht.
Retrieved August 24, 2011 at
http://statbel.fgov.be/nl/binaries/N%2000.01%20y-2009%20AAE_tcm325-
128579.xls.
FOD Economie (2010b). Bevolking per geslacht, leeftijdsgroep en klasse 1990-2008.
Vlaams Gewest. Retrieved August 24, 2011 at
http://statbel.fgov.be/nl/binaries/Structuur%20v.%20d.%20bevolking%20per
%20leeftijd%20en%20geslacht%20-%20Vlaams%20Gewest%20-
%20Per%20jaar_finale_tcm325-114594.xls.

62
FOD Economie (2010c). Onderwijsniveau van de bevolking van 15 jaar en ouder in
procenten (1987-2010). Retrieved August 24, 2011 at
http://statbel.fgov.be/nl/binaries/Onderwijsniveau1987-2010_tcm325-
44615.xls.
Gamson, W.A., & Modigliani, A. (1989). Media discourse and public opinion on nuclear
power : a constructionist approach. The American Journal of Sociology, 95(1), 1-
37.
Gitlin, T. (1980). The whole world is watching: mass media in the making and unmaking of
the new left. Berkeley: University of California Press.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for
qualitative research. Chicago: Aldine.
Goffman, E. (1974). Frame Analysis. Harmondsworth: Penguin Books.
Hajjar, W. J. (1998). Television in the Nursing Home: A Case Study of the Media
Consumption Routines and Strategies of Nursing Home Residents. London:
Routledge.
Hijmans, E., & Wester, F. (2006). De kwalitatieve interviewstudie. In F. P. J. Vester, P.
Scheepers, K. Renckstorf, & F. Wester (Eds.), Onderzoekstypen in de
Communicatiewetenschap (pp. 507-532). Alphen aan de Rijn: Kluwer.
Iyengar, S. (1991). Is anyone responsible ? How television frames political issues. Chicago:
University of Chicago Press.
Legard, R., Keegan, J., & Ward, K. (2003). In-depth interviews. In J. Ritchie, & J. Lewis
(Eds.), Qualitative research practice: A guide for social science students and
researchers (pp. 138-169). London: Sage.
Lewis, J. (2003). Design issues. In J. Ritchie, & J. Lewis (Eds.), Qualitative research
practice: A guide for social science students and researchers (pp. 47-76). London:
Sage.
Matthes, J., & Kohring, M. (2008). The content analysis of media frames: Toward
improving reliability and validity. Journal of Communication, 58(2), 258-279.
Morgan, D. L. (1997). Focus Groups as Qualitative Research (2nd ed). London: Sage.
Neuman, W. R., Just, M. R. & Crigler, A. N. (1992). Common knowledge: News and the
construction of political meaning. Chicago: University of Chicage Press.

63
Oomens, S., Scheepers, P., & Vergeer, M. (2006). Survey-onderzoek. In F. P. J. Vester, P.
Scheepers, K. Renckstorf, & F. Wester (Eds.), Onderzoekstypen in de
Communicatiewetenschap (pp. 75-100). Alphen aan de Rijn: Kluwer.
Pan, Z., & Kosicki, G. M. (1993). Framing Analysis: an approach to news discourse.
Political Communication, 10(1), 55-75.
Rea, L. M., & Parker, R. A. (2012). Designing and conducting survey research: A
comprehensive guide (3rd ed.). San Francisco: John Wiley & Sons.
Reese, S. D. (2007). The Framing Project: A Bridging Model for Media Research
Revisited. Journal of Communcation, 57(1), 148-154.
Ritchie, J. (2003). The applications of qualitative methods to social research. In J. Ritchie,
& J. Lewis (Eds.), Qualitative research practice: A guide for social science students
and researchers (pp. 24-46). London: Sage.
Scheufele, B. T. (2004). Framing-effects approach : a theoretical and methodological
critique. Communications, 29, 401-428.
Scheufele, D. A. (1999). Framing as a theory of media effects. Journal of Communication,
49(1), 103-122.
Snape, D., & Spencer, L. (2003). The foundations of qualitative research. In J. Ritchie, & J.
Lewis (Eds.), Qualitative research practice: A guide for social science students and
researchers (pp. 1-23). London: Sage.
Strauss, A. L. (1987). Qualitative Analysis for Social Scientists. Cambridge: Cambridge
University Press.
Tulloch, J. (1991). Approaching the audience: The elderly. In E. Seiter, H. Borchers, G.
Kreutzner, & E.-M. Warth (Eds.), Remote Control: Television, Audiences & Cultural
Power (pp. 180-203). London: Routledge.
Vanden Boer, L., & Pauwels, K. (2005). 75-Plussers: een Profielschets. Accessed
September 17, 2012.
https://wvg.vlaanderen.be/applicaties/kenniscentrum/pdf/webbijdragen/2005
_75-plussers_profielschets.pdf
Van Gorp, B. (2004). Frames in de nieuwsmedia (Doctoraatsthesis). Antwerpen:
Universiteit Antwerpen.
Van Gorp, B. (2007). The constructionist approach to framing: Bringing culture back in.
Journal of Communication, 57, 60-78.

64
van Selm, M., & Wester, F. (2006). Focusgroep-onderzoek. In F. P. J. Vester, P. Scheepers,
K. Renckstorf, & F. Wester (Eds.), Onderzoekstypen in de Communicatiewetenschap
(pp. 533-560). Alphen aan de Rijn: Kluwer.
Van Zoonen, L. (2007). Audience reactions to Hollywood politics. Media, Culture &
Society, 29, 531-547.
Wright, K. B. (2005). Researching Internet-based populations: Advantages and
disadvantages of online survey research, online questionnaire authoring software
packages, and web survey services. Journal of Computer-Mediated
Communication, 10(3), article 11.
http://jcmc.indiana.edu/vol10/issue3/wright.html

65
 Chapter 1

This first empirical chapter provides systematic insight in everyday celebrity news coverage and
the accompanying audience responses. This allows to situate the following chapters
(particularly the case studies in chapters 2 and 3) within the larger, systematic, whole of
everyday celebrity news. The large-scale nature of this study aims to compensate for the lack of
structural analyses in the academic domain of celebrity studies. This study combines the aspects
of text and reception and systematically analyzes online celebrity news coverage and
accompanying audience reactions from three celebrity websites (People, Heat, and HLN) by
means of a framing-based approach. This approach rooted in (political) communication studies
allows for direct insight into the essential elements of an item, is suited for large datasets, and
provides the possibility to combine the analysis of (media) encoding and (audience) decoding
processes. This chapter thus provides structural insight into the mediated phenomenon of
celebrity and the ways audiences react to it which serves as a larger background for the other
studies in this PhD. Important concepts are globalization and cultural proximity, especially in
terms of how media coverage and audience reactions differ with regard to local or global
celebrities.

OF LOCAL AND GLOBAL FAME: A COMPARATIVE ANALYSIS OF NEWS ITEMS AND

AUDIENCE REACTIONS ON CELEBRITY NEWS WEBSITES PEOPLE, HEAT, AND HLN

Hilde Van den Bulck and Nathalie Claessens

(Published in Journalism, http://jou.sagepub.com/content/early/2013/05/28/1464884913488725.full.pdf+html)

Worldwide, people follow global celebrities lives almost in real time, while communities
create their own celebrities, of great local fame but unknown internationally. Examining
the People (US), Heat (UK), and HLN (Flanders) websites, this article provides a
quantitative framing-based content analysis of a sample of online celebrity news and the
accompanying audience responses. General results show that media focus on celebrities
professional and love lives, while audiences cover more superficial topics (appearances).
Further, media tend to be less and readers more judgmental in discussing celebrities.
Examining global/local distinctions, HLN articles are shown to focus on global celebrities,
while readers comment more on local stars. Reactions also cover more personal topics
when discussing local celebrities, and both HLN articles and reactions judge global
celebrities more negatively than locals. Finally, most positive parasocial relationships are
maintained with local and most negative with global celebrities. The cultural proximity
hypothesis can help explain these differences.

Keywords: celebrity, online news, content analysis, global/local, audience reactions,

globalization, cultural proximity
66
Introduction
This article aims to fill a gap in current studies of celebrity journalism by providing a
systematic empirical analysis of day-to-day celebrity reporting and audience reactions
to this celebrity news. In doing so it examines claims regarding celebrity (news) as a
mere instrument in the continuation of consumer society versus claims that it provides a
forum to discuss social and moral topics. At the same time it wishes to explore the
impact of cultural proximity on the potentially multiple meanings of celebrity news.
In recent decades, celebrities have become omnipresent in Western society. Even
though the celebrity phenomenon predates the twentieth century, the celebrity system
as we know it has been made possible by the rise of modern mass media (popular
newspapers, cinema, radio, and television) and the (postmodern) Internet (Drake &
Miah, 2010). For a long time, celebrity news was restricted to specialized gossip
magazines, but today it is a fundamental part of all (news) media including daily and
weekly popular and quality media and their online equivalents, current affairs
programming, and so forth (Turner, 2004: 11, 71; see also Gorin & Dubied, 2011).
In the context of tabloidization, celebrity news is a frequently discussed
journalistic genre (Turner, 2004) and thus of interest for journalism studies. However,
specific content analyses are scarce (an exception is the work of Gorin & Dubied, 2011).
This paper aims to fill this gap by offering a framing-based analysis of a large sample of
online celebrity news articles (N=1,289) and, not addressed in the past, their online
readers comments (N=16,000). The sample consists of all articles and accompanying
audience reactions on three celebrity news websites: the Flemish HLN, the British Heat,
and the American People website. While most research on celebrity news focuses on
particular case studies (e.g., Butler Breese, 2010), we choose to examine the everyday
celebrity news coverage which readers encounter on a daily basis.

Theoretical Framework
Celebrities, news media, and audiences
The celebrity construct. The theoretical starting point of this paper is the notion of
celebrity as constructed within the celebrity apparatus, consisting of four actors: the
famous person and his/her entourage (i.e., manager, publicist, agent), the media, and the
audience. All actors need to be present to guarantee the (ongoing) construction of a
celebrity (i.e., celebritization; Turner, 2004), whereby the public figures private life and

67
behavior increasingly becomes (at least) as important as his/her professional activities
to media and audiences. Media exposure negotiated by the famous persons entourage
- is like oxygen to a celebrity who, in turn, helps media compete for shares of audiences
eager to read about the rich and famous (Drake & Miah, 2010; Gamson, 1994; Marshall,
2006a; Rojek, 2001). At the same time, celebrities cannot exist without the attention and
interest from audiences. The latter, in turn, use celebrities to fulfill multiple needs: as
mere entertainment or food for gossip, representations of what it is to be human
(Dyer, 1998: 99), or means to discuss social/moral issues.
The celebrity construct typically unites three personae: the professional (the
roles they perform, the music they make, the products and causes they endorse), the
private (the official private life as presented to the outside world), and the real persona
(the paparazzi glimpses of the real person behind the celebrity). Furthermore,
celebrities distinguish themselves by the paradoxical combination of ordinariness and
extraordinariness (Dyer, 1998) and a certain habitus that separates them from other
people (Gorin & Dubied, 2011: 600-1). At the same time, celebrities represent societal
ideas, norms, and values (Dyer, 1987). As such, the growing centrality of celebrities has
been linked to societys celebration of individuality - as celebrities represent individual
success and transformation and to (the rise and perseverance of) consumer capitalism
(Marshall, 2006b: 317). As exponents of the culture and entertainment industries,
celebrities are commodities, created to embody capitalist society as both objects and
vehicles of consumption, encouraging and validating consumer culture (Cashmore,
2006: 269).
This constructed, multi-image, exceptional, and ideological character of celebrity
does not prevent a notion of authenticity (Tolson, 2001), but points to the fact that
authenticity is not just determined by the celebrity itself, but by all meaning-creating
players in the apparatus, including media and audiences. Indeed, virtually all the
communication with and about these celebrities is essentially mediated, turning
celebrities into mediated personae (Evans & Hesmondalgh, 2005).
Celebrity news. Celebrity news can be defined as news that is produced according
to certain news criteria and routines and that focuses more on the private than the
public life and behavior of celebrities (Gorin & Dubied, 2011: 600-1). From a journalistic
perspective, the success of celebrity news can be explained by its suitability for short-
term news cycles, short news stories, and contemporary medias preference for

68
entertainment (Turner, 2010: 11). Celebrity journalism at its core is aimed at revealing
the real celebrity and at creating intimacy with their everyday lives rather than
focusing on distance and the celebritys aura (Holmes, 2005; Marshall, 2006b). Celebrity
scandals and paparazzi photographs have thus become mainstream news content, tying
in with shifts towards sensationalism, entertainment, and the realm of private affairs, in
short tabloidization (Turner, 2004).
Indeed, several case studies demonstrate this emphasis on the private and real,
either in topics like teenage pregnancies (Butler Breese, 2010) and sex (Dyer, 1987), or
in the rhetoric of authenticity present in most celebrity coverage (Holmes, 2005).
However, little systematic empirical work has been done on overall, day-to-day celebrity
reporting, a shortcoming this paper wishes to overcome by analyzing an extensive
corpus of celebrity news. This enables the examination of claims such as Dyers (1998:
35) that the general image of stardom consists of consumption, success and
ordinariness combined with love, marriage and sex, Halls (1980) original contention
that media express the dominant ideology of market capitalism, and the role of news
media in defining celebrity authenticity.
Audiences. As one of the four actors in the celebrity apparatus, the audience is a
crucial part of the construction of celebrity. Even though audiences depend on the media
for news about celebrities they almost never meet in person (Cashmore, 2006), they
have the freedom and power to construct their own definitions of (authentic) celebrity.
This echoes Halls (1980) encoding/decoding theory in which media encode a message
by presenting a dominant interpretation, which can be adopted, negotiated, or rejected
by the audience. Celebrity news thus functions as information filters for audiences and is
central in the development of emotional and cognitive connections between audience
members and celebrities. Such involvement with celebrities can take the form of
identification processes (Tian & Hoffner, 2010) or parasocial relationships (Horton &
Wohl, 1956; Giles, 2002). A parasocial relationship (PSR) can be described as the illusion
of a long-term friendship with a celebrity, which is one-sided and enabled by the media,
but entails an emotional connection and is analogous to social relationships.
Identification refers to a more intense involvement, defined as psychologically merging
with the [celebrity] persona, momentarily losing ones self-identity and situational
references (Tian & Hoffner, 2010: 252), occurring more with fans than general

69
audiences. Both types of connections, in turn, can impact on the way audiences decode
celebrity news.
Despite the apparent triviality of the topic, the construction and media coverage
of celebrities help shape audiences understanding of the social world (Evans &
Hesmondalgh, 2005: 14; Gorin & Dubied, 2011) and their cultural identities (Dyer,
1998): the lives of celebrities provide parables of instruction for fans and general
audiences (Rojek, 2007: 73). Some authors interpret this within the celebrity-as-
commodity thesis, arguing that audiences attachment to celebrities and interest in
celebrity news is primarily to compensate for their anonymity and alienation, and for
their frustrations in capitalist society, i.e., for failing to reach the dream of becoming rich
and glamorous. Others argue that the meaning of celebrities goes beyond the consumer
capitalism argument as celebrity coverage provides media and society with a forum for
discussion of a wide range of social and moral topics and issues (Cashmore, 2006; Drake
& Miah, 2010; Evans & Hesmondalgh, 2005; Gorin & Dubied, 2011; Marshall, 2006a).
Yet, with the exception of some specific case studies (e.g. Van den Bulck & Claessens,
2013), there is little empirical work that systematically analyses audience reactions to
celebrity news against these claims. This paper aims to help fill this gap by analyzing
both celebrity news coverage and (online) audience reactions to these items, a
comparative analysis that will also enlighten how celebrity and its authenticity are
conceptualized, defined, and portrayed by both media and audiences.

The global and the local in celebrity culture

Global media and entertainment industries. As both product and motor of the
media and entertainment industries, celebrities are part and parcel of the multi-faced
globalization process (Appadurai, 1996; Turner, 2004). For several decades now, media
and entertainment industries are dominated by trans-national and cross-media
concentrating and converging companies, pushing (cultural and wider) globalization.
Celebrities, according to Turner (2004: 34), prove crucial in this, as they are a branding
mechanism for media products that has assisted their fluent translation across media
formats and systems of delivery and, indeed, across national borders. As has been
extensively demonstrated, and despite the identification of counter-flows, this
globalization has a distinct US flavor (Berger & Harrington, 2002; Miller et al., 2001). As
such, the US celebrity is the foot soldier of globalization (Turner et al., 2000: 176),

70
him/herself becoming of world fame, and news about global celebrities spread easily
through new technologies and converging modes of delivery attracts audiences all
over the world.
The local. While the growing dominance of celebrity culture worldwide may be
the result of globalization processes, creating a vast army of global celebrities, there
clearly are local celebrity cultures, with their own local stars, originating from specific
cultures (e.g. Schlager music celebrities in Germany, The Netherlands, and Flanders; Van
den Bulck & Van Gorp, 2011), or from local versions of global formats such as reality
television (Dhoest, 2005). Ferris (2010: 393) defines a local celebrity as a more narrow-
cast version of celebrity, in which people are treated as famous only by and for their fan
audiences, be it at a local, regional, or even national level (cf. Australian celebrity
culture; Turner et al., 2000). Local media pay ample attention to these home celebrities,
according to some increasingly so. For instance, a diachronic (1949-2005) analysis of
French society magazine Paris Match covers shows increased attention for French rather
than foreign celebrities (Chenu, 2010).
Celebrity audiences and cultural proximity. Audiences are confronted with and
take varying interest in both global and local celebrities. As such, audiences do not only
identify or maintain parasocial relationships with international stars, as relational
dynamics of celebrity operate at others levels as well: and there may also be interesting
variations on these patterns that are observable at these levels (Ferris, 2010: 393). One
element that mediates relationships of audiences with global or local celebrities is
cultural proximity (Tian & Hoffner, 2010), a concept that proved important in research
on news factors (Galtung & Ruge, 1965; Zaharopoulos, 1990) and the popularity of
national TV programmes (De Bens & de Smaele, 2001; Straubhaar, 1991). Coined by De
Sola Pool (1977), Straubhaar (2003: 85) describes it as the tendency to prefer media
products from one's own culture or the most similar possible culture. Most work in this
regard has been done on news and TV fiction (soap operas), suggesting that cultural
proximity is complex, incorporating a wide range of elements including personal and
group affiliations, lifestyle, language, and social norms and values relating to gender,
religion, ethnicity, and family (Straubhaar, 2003).
While genre proximity may explain the worldwide interest in celebrity news as it
evolves centrally around melodrama familiarity with which has been shown to be the
basis for the success of soaps and telenovelas worldwide (La Pastina & Straubhaar,

71
2005) -, other aspects may explain possible differences in reactions of audiences
towards global or local celebrities. This is further suggested by work into interpersonal
liking that can be extended to identification and parasocial relationships between
audiences and celebrities. This work indicates that homophily or perceived similarity
predicts a strong association between attitude/background similarity and interpersonal
liking (Klimmt et al., 2006; Tian & Hoffner, 2010). As such, this study wishes to
investigate if and how audiences respond differently to coverage of global and local
celebrity.

Analytical Framework
A framing approach to celebrity news and audience reactions
As a framework for analysis, we opt for a framing approach because, more than other
methods for content and audience analysis, it allows for the study of both encoding
(media frames) and decoding (audience reactions, see Hall, 1980) with the same
analytical tool. Framing can be described as a process whereby a frame determines
which aspects of reality are selected, rejected, emphasized, or modified in the
production of a media text and, at the same time, provides the audience with a context
and suggested meaning (Van Gorp, 2006: 46). Frames serve as meta-communicative
messages that are present in a text through framing devices, such as word or image
choice, metaphors, symbols, and stereotypes (Entman, 1991: 7). Reasoning devices relate
to the four functions of a frame: problem definition, causal interpretation, moral
evaluation, and treatment recommendation (Entman, 1993: 52).
Yet, the influence on audiences is not guaranteed by the mere presence of a frame
in a text. Researchers have pointed to a list of framing moderators, such as interaction
with peers (Neuman et al., 1992), personal experiences and cognitions that audiences
bring into play to make sense of media content (DAngelo, 2002), parasocial
relationships that people maintain with celebrities (Schiappa et al., 2005), and a
successful fit between audience and media values, norms, and practices (Gamson &
Modigliani, 1989). The latter is tied in with the distinction between global and local
celebrities that may represent varying levels of this cultural fit, affecting audiences
reading of celebrity news. As a result, media frames can be consumed by audiences in
different ways.

72
 Due to the large variety of topics and issues in our sample, it proved impossible to
identify general celebrity news (reception) frames. However, the principles of framing
theory, particularly the reasoning devices, are employed to allow for a systematic
comparative analysis of celebrity news coverage and the audience reactions to these
stories.

Online audiences
In this study, audiences varied readings of celebrity news are measured analyzing
online reactions to media articles as they appear on celebrity websites. This method can
be motivated by the more natural results that arise from mundane research contexts
(such as online discussion sites) rather than those obtained in unnatural settings like
experiments, interviews, or focus groups (Van Zoonen, 2007: 534-535). One
disadvantage is that only certain types of audience members are likely to engage in
online discussions, while a majority of online news consumers are lurkers, following
but not actively taking part in online discussion (Beyers, 2004). Others are not online at
all. Moreover, we often do not know who is behind the posting, why the individual sent
in a comment or how postings relate to peoples real-life politics (Van Zoonen, 2007:
535). However, the latter two issues also occur in face-to-face communication while
online comments provide useful information about how people present themselves to
others (Van Zoonen, 2007: 535).
Earlier work on online audience behavior can provide additional insights
significant to our study. Research showed that people display higher levels of self-
disclosure (i.e., revealing personal information) on the Internet than in face-to-face
situations, which can be explained by the psychological effects of anonymity and
heightened self-awareness (Joinson, 2001). In addition, Walther (1996) found that while
computer-mediated communication (CMC) can foster impersonal contact over face-to-
face communication (FtF), in some hyper-personal communication cases CMC has
surpassed the level of affection and emotion of parallel FtF interaction (Walter, 1996:
17). Bayms (2000) analysis of the online r.a.t.s. community on Usernet shows high
levels of self-disclosure and community among soap fans resulting from a need to share
the soap-viewing experience by discussing it with other fans. Similarly, we expect that
celebrity news readers want to share their personal interpretations with other celebrity
news website followers.

73
Operationalization
Data. Data come from three websites devoted to celebrity news: HLNs showbizz
page (Flanders, i.e. the northern part of Belgium), Heat (UK), and People (US). These
websites are selected because of their high number of users (see below) and the
opportunity for online audience reactions. We do not take into consideration varying
degrees of tabloidization as this is not our primary interest. The showbiz channel of HLN
website (www.hln.be/showbizz) is selected as Flemish celebrity website. HLN is both a
Flemish popular newspaper and a news website. The HLN newspaper can be described
as the most tabloid-like daily in Flanders (Van den Bulck, 2011). The HLN website
reaches 599.220 unique users a day (CIM, 2011) while its showbiz channel, the part we
are interested in, reaches 99.254 unique users a day (K. Hellemans, personal
communication, 18 January 2011). Heat website (www.heatworld.com) was selected as
British celebrity website. Heat is a British magazine and a celebrity website. Heat
magazine is described by Turner (2004: 74) as slightly trashier, cheekier, less
sophisticated, and more gossip-oriented than its British counterparts Hello! and OK, but
as still very positive and sympathetic. Heat website targets British women aged 16 to 34
and reaches 800.000 unique users (Bauer Media, 2010: 9). People is selected as the
American celebrity website in our sample. People is not only a celebrity website, but also
a magazine (established in 1974), built from Time magazines People section
(Marshall, 2006b: 320). Turner (2004) describes People magazine as less scandalous
than Star and the National Enquirer, other American celebrity weeklies. The People
website reaches 13.2 million unique users a month and approximately 430.000 users a
day (People, 2011a). The median age of People users is 36, and 66 percent of its visitors
are female (People, 2011b).
The sample consists of ten constructed weeks from January to October 2010 (one
week/month), in which all celebrity articles and the first one hundred audience
reactions to each article on all three websites were selected. The first hundred audience
reactions were chosen out of practical considerations, although we realize that limiting
the number of audience reactions may cause loss of information. A total of 1,289 articles
and 16,000 audience reactions were collected. Most audience reactions were posted in
response to People articles (56.3 %), followed by Heat (30.6%) and HLN (13.1%). Heat
has the highest reactions per article ratio (M=58.7; SD=72.5), followed by People
(M=40.7; SD=82.2) and HLN (M=5.3; SD=12.0).

74
 Topic List. The topic list is based on previously developed theoretical insights
within framing literature, and is applied to both the articles and audience reactions (see
Appendix 1). First, the characteristics of the celebrity discussed are coded: i.e., gender,
nationality, ethnicity, sexual orientation, and domain. Second, investigating the frames
reasoning devices, the problem or issue is coded, as well as the cause, judgment, and
solution proposed in the article/reaction. Third, based on previous findings and the
theoretical framework, audience reactions are additionally analyzed for expressions of
parasocial relationships (directly addressing the celebrity or describing him/her as a
friend), references to personal experiences (cf. higher self-disclosure), and interaction
with other readers (i.e., peers; cf. hyper-personal communication).

Results
Celebrity characteristics
The celebrities that are covered most often in the articles are US actress Lindsay Lohan
(3.1%), US singer Lady Gaga (2.4%), and UK singer Cheryl Cole (2.2%). Audience
reactions talk primarily about Cheryl Cole (3.1%), US golf player Tiger Woods (2.7%),
and Lindsay Lohan (2.3%). Interestingly, nine (all but US socialite Paris Hilton) of the
articles top ten celebrities recur in the audience reactions, indicating great similarity in
the importance attached to these celebrities by media and audiences. All top five
celebrities had an eventful 2010: Lindsay Lohan went in and out of prison/rehab, Lady
Gaga became a world famous star who kept on surprising (shocking) the world with her
work and looks, Cheryl Cole and US actress Sandra Bullock divorced their cheating
husbands, Tiger Woods cheated on his wife and went into sex rehab, and UK TV
personality Katie Price continued to fight with her ex-husband Peter Andre while
marrying Alex Reid.
Gender. Both the articles and audience reactions pay almost as much attention to
male as to female celebrities. Readers react more to news on female (58.6%) than on
male celebrities (41.4%). A slight but significant overrepresentation of women was
found in the Heat and People articles (X=15.134; df=2; p<.01) and audience reactions
(X=139.844; df=2; p<.001).
Nationality. American, British, and Flemish celebrities are covered most
frequently. This is not surprising as we selected media from the US, UK, and Flanders.
Still, the domination of American celebrities is striking. Canada comes in fourth place,

75
probably due to its proximity to the US. Fifth place is for the Netherlands, which may be
explained by its geographical and cultural proximity to Flanders with which the
Netherlands shares a language as well as a border. Interesting in this regard is the
absence of Walloon (the southern, French-speaking part of Belgium) and French
celebrities, indicating the different cultural spheres extent to celebrity cultures. Looking
at the celebrities nationalities per medium, both British Heat and, especially, American
People are oriented towards their own celebrities, whereas the Flemish HLN almost
neglects its local celebrities (only 9 percent) for an international focus on American and
British celebrities. The reactions replicate the articles results for Heat and People, as
their audiences focus on home celebrities. This seems to confirm the cultural
proximity thesis, an interpretation furthered by the case of HLN where audience
members differ significantly (X=132.705; df=2; p<.001) from the medium in that they
discuss Flemish celebrities in 32.5 percent of their comments while the HLN articles
mostly cover international stars. Still, HLN readers mimic the articles in that their
comments are mostly on Americans (44.3%).
Ethnicity. The media articles and audience reactions discuss white celebrities in
almost 90 percent of the items. Almost eight percent of the articles and almost nine
percent of the reactions cover black celebrities, followed by Latin stars (3.4% articles,
4.0% reactions). The smallest percentage is that of Asian celebrities (1.4% - 0.8%). The
celebrity world thus seems to be predominantly white, especially in HLN (85.6% whites)
and Heat (88.6% whites) articles. Most Latin celebrities (5.2%) appear in People articles
and reactions, which is not surprising as the Latin population in the US is larger than in
the UK or Flanders. HLN has a relative high percentage of Latin celebrities (2.7%)
compared to the Latin population in Flanders. As discussed above, HLN is internationally
(mainly US) oriented, which explains this percentage as well as the relatively high
percentage of black and Latin celebrities in the HLN sample.
Sexual orientation. More than 90 percent of the articles and reactions cover
heterosexual celebrities, indicating a lasting dominance of heteronormativity for these
media and their readers. Interestingly, bisexuals are discussed more often than
homosexuals, which could be explained by the fact that Lindsay Lohan, who is bisexual,
is the most frequently discussed celebrity. When we look at the items on bisexual
celebrities in our sample, items about Lindsay Lohan indeed account for 43.1 percent,
followed by Megan Fox (14.7%) and Lady Gaga (10.5%). Interestingly, Heat (5.1%) and

76
People (5.5%) pay more attention to bisexuals than HLN (3.1%). Homosexuals are
covered most in Heat reactions (3.1%) (X=69.403; df=4; p<.001).
Domain. The media spend most of their articles on music (33.6%), film (31.4%),
and TV/radio celebrities (16.6%). These domains recur in audience reactions, although
they are ranked significantly (X=66.660; df=15; p<.001) different: first film (31.0%),
then music (30.4%), and then TV/radio (22.4%). The fourth place significantly
(X=2191.161; df=30; p<.001) differs depending on the medium: HLN (9.7%) and People
(5.1%) pay more attention to sports celebrities, while Heat is more interested in
celebrities coming from the sex industry (6.5%). The fifth spot is devoted to sports
celebrities in Heat (3.4%), to stars from the fashion world (models, designers, ) in
People (3.0%), and to fashion celebrities (3.6%) and royals (2.8%) in HLN. Although the
UK is renowned for its interest in its royal family (McGuigan, 2000), Heat only covered
royals in two articles and four reactions. So, overall, it appears that the celebrities
grabbing the attention of both media and readers come from the world of film, music
and TV/radio, are mainly heterosexual, slightly more white and female, and despite an
overall US dominance show a certain geographical proximity to media and audiences.

Framing analysis
The media articles and audience reactions are further examined by means of the four
reasoning devices of a frame: problem definition, causal attribution, moral judgment,
and solution recommendation (Entman, 1993: 52).
Issue/problem. The most discussed issues/problems differ significantly for the
articles and reactions (X=509.641; df=21; p<.001). The articles cover, in more than 28
percent of the cases, the celebrities professional activities (28.3%), including their new
work (album, film, concert, ), award show attendance, and the like. More than thirteen
percent of the media articles cover love stories, including flirtations, romance, a new
relationship, engagement, marriage, relationship problems, divorce, life as a single,
etcetera. Articles further deal with the following topics: appearances (celebrity looks:
8.5%), children (pregnancy, birth, good/bad parenting: 7.3%), and crime (drugs, theft,
driving under influence, fraud, burglary, violence: 6.4%).
Audience reactions discuss the more superficial topic of celebrities appearances
the most (20.1%), followed closely by celebrities professional life (19.4%), love life
(8.6%), personality (8.6%), and children (7.0%).

77
 Overall, it is clear that the selected media and particularly their readership focus
more on private (media: 61.5%; readers: 73.7%) than on public (media: 38.5%; readers:
26.3%) issues. This confirms Gorin and Dubieds (2011) findings that celebrity news
media focus more on intimate biographical details (love, children, and health).
However, the issues/problems identified differ significantly according to medium
(X=1872.542; df=42; p<.001), despite the fact that all articles and reactions, except for
Heat reactions, primarily discuss the celebrities professional occupations. The Heat
readers mostly react to appearances or how the celebrity looks (28.1%). Appearances
are also the second most discussed topic in HLN reactions (11.3%), People reactions
(17.8%), and Heat articles (16.0%). Another oft-mentioned topic is love, especially in
HLN (11.1%), People (17.2%), and Heat articles (12.2%), and in People reactions
(10.6%). Celebrities relationships to their children are also covered frequently in all
media and in Heat (6.1%) and People audience comments (7.8%). An interesting
difference between Flemish and other media is the large amount of articles and
reactions spent on crime: it takes third place in the HLN articles (8.0%) en reactions
(9.0%).
Cause. Both articles and reactions point most frequently to the celebrity (80.6%)
rather than the outside world as the cause. The cause is either situated in the public
sphere (HLN & Heat reactions; e.g., pursuit of profit, career) or in private sphere
(People, Heat & HLN articles; People & Heat reactions; e.g., out of love, lust).
Judgment. Media articles are mostly neutral towards the celebrities they discuss
(62.3%), whereas audience members tend to be more judgmental, both negative
(42.5%) and positive (36.9%) (X=1155.645; df=2; p<.001). In contrast, audience
members do not seem to hold back in their evaluations as they express judgment in 79.4
percent of their reactions. However, there are considerable differences in the judgments
made by the Flemish, British, and American media and readers (X=969.704; df=4;
p<.001). Celebrities are most often judged neutrally in HLN (79.6%) and People (64.1%)
articles. People articles are the most positive (30.6%) of the three media, followed by
Heat (28.3%). Heat has the most negative judgments (39.7%). Results for audience
reactions are quite different though: People (44.3%) and (especially) HLN readers
(56.8%) are more positive than the media content they are commenting on. HLN
(49.0%), Heat (56.8%) and People readers (33.2%) are also more negative than the

78
media. Heat readers are the most negative of all readers (56.8%) and People readers the
most positive (44.3%).
Interestingly, all medias readers are less neutral than the articles they comment
on. The difference between the articles and reactions can be explained by multiple
factors. First, articles mainly contain factual information and are less focused on
developing judgments, whereas expressing opinions is the core goal of readers reacting
to these articles. Second, the websites want to at least appear to stick to journalistic
standards of impartiality and objectivity in their celebrity reporting. Third, and related,
media are careful in their writings to preserve good working relations with celebrities
(as they depend on them for news) and to avoid costly defamation law suits. Readers
take little risks when writing evaluative comments, because they are not dependent on
celebrities for their income and cannot be punished (as most react anonymously).
Solution. Only 10 percent of the articles and reactions mention a solution or
prospect. When the media and readers present a solution, there is an almost equal
distribution of internal (coming from the celebrity his-/herself) and external solutions
(originating in the outside world).

Parasocial relationships, personal experiences, and interaction with peers

Parasocial relationships. 44 percent of reader responses express a parasocial
relationship (PSR) with the celebrity discussed in the article. Readers comments were
coded as containing parasocial references when directly addressing the celebrity or
speaking of the celebrity as if they know him/her personally. These parasocial
expressions are positive in 26 percent of the cases (I love you, you are the best) and
negative in 18 percent (I hate you, you suck). References to PSR with celebrities are
found most frequently in People readers comments: 36.8 percent of the reactions
express positive PSR with the celebrity and 19.7 percent negative PSR. Heat readers
express the least PSR (26.3%), followed by HLN readers (32%) (X=1587.072; df=4;
p<.001).
Personal experiences. Less than six percent of the readers comments contain
references to peoples personal experiences. About half of the personal experiences
mentioned are positive. Positive experiences can be found most frequently in HLN
(4.2%) and People readers comments (3.5%), and negative experiences in Heat

79
reactions (2.8%) (X=18.053; df=4; p<.01). High online self-disclosure is thus found in
only a small part of the audience reactions.
Interaction with peers. In 80.1 percent of the items, readers do not interact with
peers (i.e. the other commenting readers). In 10.3 percent, readers constitute a friendly
community in which they positively interact with one another. In 9.6 percent of the
cases, virtual disagreements or conflicts are present. Interestingly, there is a difference
between the media when it comes to interaction among peers (X=1260.580; df=4;
p<.001). HLN readers interact the least (7.7%). In 22.1 percent of Heat reactions,
however, people positively interact as if they are part of a community, a group of friends,
reaffirming each others opinions, similar to Bayms (2000) r.a.t.s. group. Most conflicts
among peers could be found in People reactions: here 12.6 percent of the interaction
with peers was negative.

Local versus global

Further investigating the coverage of and reactions to celebrities, the difference between
local and global celebrities is taken into account. We expect that a celebritys cultural
proximity (i.e., nationality) will entail background similarity and, thus, sympathy for the
celebrity. In the context of celebrity news, it is thus interesting to investigate how media
cover and audiences react to local versus global celebrities.
Only the HLN medium is examined here, for two complementary reasons. First,
the Flemish case allows for a distinction between local (only famous in their home
region) and global (American and British) celebrities with worldwide fame. In the case
of People and Heat, such distinction is far more difficult to make, as many of the
celebrities they cover have a double status as local (American or British) and global
celebrity. Second, with regards to the readership, given the worldwide dominance of the
English language and the global interest in many American and British celebrities, there
can be no identification with any degree of certainty of the nationality of the Heat and
People readers. Readership of HLN is restricted to people with a command of the Dutch
language and with at least a basic interest in local, Flemish celebrity culture (the latter
confirmed by the high reader reaction rate to local celebrity news items).
To examine whether the Flemish website and its readership treat local and global
celebrities differently, a range of bivariate analyses are executed. First, correlations
between, on the one hand, the local/global character of celebrities (as independent

80
variable) and, on the other hand, the items problems/issues, judgments, and the
readers parasocial relationships, personal experiences, and interactions with peers (as
dependent variables) are discussed.
Problem/issue. Correlations between nationality and problem/issue are not
significant for the websites articles. For the reactions, however, significant relationships
are found (X=562.855; df=40: p<.001). Interestingly, HLN reactions to local celebrity
news tend to be about personal or emotional themes (alongside professional activities
(19.8%)), such as a celebritys death (22.1%), break-up (11.0%), and children (9.8%),
whereas reactions to global celebrities are more about distant topics such as their
professional life (30.6%), appearances (16.2%), and behavior (6.9%).

Table 1. Local/Global Difference for Problems (Reactions)

Top Five Problems (Reactions)
Local % Global %
1. Death 22.1 1. Professional 30.6
2. Professional 19.8 2. Appearances 16.2
3. Crime 12.2 3. Sex 9.5
4. Break-up 11.0 4. Crime 8.9
5. Children 9.8 5. Personality 6.9
(X=562.855; df=40: p <. 001)

Judgment. It becomes clear that HLN articles are (significantly: X=9.804;

df=4;p<.05) more positive towards the local (15.6%) and more negative towards the
global (American and British) stars (14.6%). HLN readers are less neutral toward local
(10.4%) than to global celebrities (24.0%). These readers are in fact significantly more
positive towards local stars (42.1%) (X=86.867; df=4; p<.001) and equally negative
towards local (47.5%) and global celebrities (49.9%).
Parasocial relationships. A more detailed analysis of the parasocial relationships
of HLN readers according to the celebrities nationality shows that most positive PSR is
expressed towards local celebrities (26.2% local; 15.5% global; X=54.249; df=4;
p<.001). Negative PSR is found least often in reactions to local celebrities (8.1%) and
most often with global stars (17.3%). This seems to point to a relationship between
positive PSR and cultural proximity.
Personal experiences. More personal experiences are mentioned when readers
react to local (7.4%, especially positive experiences (5.8%) compared to global
celebrities (4.2%) (X=11.710; df=4; p<.05). It appears that cultural proximity invites

81
people to feel a great resemblance to what happens in local celebrities lives, provoking
them to discuss personal experiences.
Interaction with peers. There is more interaction with other readers when
reactions are about global (9.7%) rather than local celebrities (3.8%) (X=25.720; df=4;
p<.001). In the case of global celebrities, interactions with peers are both positive
(5.2%) and negative (4.5%).

Discussion and Conclusion

Our framing-based content analysis provides empirical insight into some of the key
claims about celebrities and celebrity news. For one, the combination of news analysis
and analysis of audience reactions confirms the conceptual notion that celebrity is
continuously constructed and reconstructed in a complex interplay between the famous
person, media and audiences. Celebrities covered by articles and audience reactions
make up a similar top ten. Further similarities consist of the celebrities covered by the
media and readers being mostly white, heterosexual, and working in the music, film, or
TV/radio industry. However, when it comes to the core issue of a celebritys
authenticity, differences between articles and reactions with regard to issues and
judgment indicate that media and readers employ a different conceptualization of
(authentic) celebrity. Whereas celebrity news presents celebrities as either authentic or
unauthentic, focusing on their professional and love lives (i.e., public and private
persona), readers often do not accept this and either criticize it (in the form of negative
judgments) or construct their own definition of (authentic) celebrity by introducing
different problems/issues through which they discuss celebrities, often in terms of the
celebritys real persona. As such, the readers tend to employ a rhetoric of authenticity
in terms of the off-guard, unkempt and unready (Holmes, 2005: 24) celebrity with a
large focus on appearances , while the media search for authenticity through a
combination of the celebritys off-guard and more controlled aspects like their
professional and (official) private activities.
Our analysis further confirms Dyers (1998: 35) contention that stardom is
mainly about consumption, success and ordinariness combined with love, marriage
and sex, but this too is dealt with differently by press and audiences. Celebrity news
mainly covers celebrities professional and love lives, whereas reactions focus on
appearances and personality, next to celebrities professional activities, love life, and

82
children. This indicates that while celebrity news has the potential to evoke discussions
of wider societal issues, as indicated by Cashmore (2006) and Evans & Hesmondalgh
(2005), amongst others, and demonstrated for specific high profile cases such as a
celebritys suicide (cf. Van den Bulck & Claessens, 2013), readers reactions often do not
go beyond quite superficial comments on how a celebrity looks or whether (s)he is
likeable. This provides support for the celebrity-as-commodity thesis arguing that
audiences attachment to celebrities and interest in celebrity news is primarily to
compensate for their anonymity and alienation, and for their frustrations in capitalist
society.
However, the results also indicate that audiences attention to celebrity news
goes beyond a passive escapism. Looking at the judgments made by media and readers,
the media tend to provide rather neutral celebrity accounts which can be ascribed to the
medias aim for a status as objective news providers, the factual character of most
articles, and the need to maintain good working relationships with celebrities. Readers,
on the other hand, show a much more strong and complex involvement, judging
celebrities both negatively (Heat) and positively (People). This can be attributed to the
lack of accountability: the possibility to react anonymously creates a freedom to make all
sorts of statements regardless of the consequences. Yet, it can also indicate audiences
involvement is more complex and rich and that entertainment does not necessary
negate a critical stance towards the celebrity phenomenon.
Finally, the articles aimed for a better understanding of the global-local dynamics
with regard to celebrity. The framing-based analysis shows that most articles and
reactions cover American celebrities, followed by British and Flemish stars. This
indicates that globalization of celebrity culture is a complex issue, confirming
Americanization as well as the relevance of other dominant celebrity centers (notably
UK celebrities receiving worldwide attention) and of local celebrity cultures. Focusing
on the comparison between the treatment of local versus global celebrities by the
Flemish HLN showbiz website and its readers, interesting (and significant) differences
arose. First, while HLN articles hardly (-10%) cover Flemish celebrities, readers appear
far more interested in these local stars. The former can be attributed to the cheap and
easy access to an abundance of global celebrity news, which is heavily copied from
foreign celebrity news sources such as celebrity gossip magazines and websites,
compared to the limited pool of local celebrities that does not guarantee a similar

83
constant stream of information. Readers interest in local celebrities seems to point to
the relevance of cultural proximity in attracting audience attention. Second, more
personal, emotional topics (such as death, break-up) are discussed for local celebrities,
while global celebrities tend to be covered in terms of less emotional, less depth-seeking
issues such as appearances and personality. Readers thus appear to be more involved
with local celebrities (who often have similar backgrounds as the readers), confirming
the homophily phenomenon. Third, articles appear far more positive toward local and
more negative toward global celebrities. This may be explained by the need to maintain
good working relationships with local celebrities to guarantee access and, increasingly,
to avoid defamation law suits. News about global stars, on the other hand, is rarely
obtained through direct contact with these celebrities who, in turn, will not display
much interest in (or eagerness to sue) local news outlets, giving the latter a level of
freedom to be judgmental. This seems to be copied by audiences that comment mostly in
a negative fashion on global celebrities. Interestingly, local celebrities seem to evoke
more evaluative judgments, both positive and negative. Again, cultural proximity might
explain this, as local celebrities are considered to operate within the same value system
as the readers, providing the latter with more insight to condemn or praise them for
acting within the common belief system. Finally, in terms of references to parasocial
relationships with celebrities, most positive expressions are found for local and most
negative for global celebrities, again confirming the homophily thesis. Local celebrities
thus appear to have a certain appeal for the Flemish HLN medium and its readers that is
not found in global stars, despite a seemingly homogenized global celebrity culture. The
results, however, question whether this can be attributed to cultural proximity or to
proximity an sich, given that much of the celebrity coverage deals with globally
recognized themes, topics, and emotions that are by and large interpreted within a
Western belief system. It thus appears that readers are more interested in local
celebrities not so much because of cultural similarities but because they live in the same
country, i.e., in their geographical proximity. More in-depth, qualitative analyses are
needed to better understand this.

84
References
Appadurai, A. (1996). Modernity at Large. Minneapolis: University of Minnesota Press.
Bauer Media (2010). Heat Media Pack. [online] Available at:
<http://www.bauermedia.co.uk/Global/mediapacks/Heat.pdf> [Accessed 23
February 2011].
Baym, N. K. (2000). Tune In, Log On: Soaps, Fandom, and Online Community. London:
Sage.
Berger, P. L., & Harrington, S. P. (2002). Many Globalisations: Cultural Diversity in the
Contemporary World. Oxford: Oxford University Press.
Beyers, H. (2004). Interactivity and Online Newspapers: a Case Study on Discussion
Boards. Convergence, 10(4), 11-20.
Butler Breese, E. (2010). Meaning, Celebrity and the Underage Pregnancy of Jamie Lynn
Spears. Cultural Sociology, 4, 337-55.
Cashmore, E. (2006). Celebrity Culture. New York: Routledge.
Chenu, A. (2010). From Paths of Glory to Celebrity Boulevards: Sociology of Paris Match
Covers, 1949-2005. Revue Franaise De Sociologie, 51(1), 69-116.
CIM (2011). CIM Metriweb Public Report 14/02/2011 20/02/2011. [online] Available
at: < http://www.cim.be/downloads.php?files=CIM Metriweb - Public Report
TAGS - 14-02-2011 - 20-02-2011.pdf> [Accessed 23 February 2011].
DAngelo, P. (2002). News Framing as a Multi-Paradigmatic Research Program : A
Response to Entman. Journal of Communication, 52, 870-88.
De Bens, E., & de Smaele, H. (2001). The Inflow of American Television Fiction on
European Broadcasting Channels Revisited. European Journal of Communication
16(1), 51-76.
De Sola Pool, I. (1977). The Changing Flow of Television. Journal of Communication,
27(2), 139-49.
Dhoest, A. (2005). The Pfaffs are not like the Osbournes National Inflections of the
Celebrity Docusoap. Television & New Media, 6(2), 224-45.
Drake, P., & Miah, A. (2010). The cultural politics of celebrity. Cultural Politics, 6(1), 49-
64.
Dyer, R. (1987). Heavenly Bodies: Film Stars and Society. London: MacMillan.
Dyer, R. (1998). Stars. London: British Film Institute.
Entman, R. M. (1991). Framing US coverage of international news: contrasts in

85
 narratives of the KAL and Iran air incidents. Journal of Communication, 41(4), 6-
27.
Entman, R. M. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of
Communication, 43(4), 51-8.
Evans, J., & Hesmondhalgh, D. (2005). Understanding Media: Inside Celebrity.
Maidenheid: Open University Press.
Ferris, K. O. (2010). The Next Big Thing: Local Celebrity. Society, 47(5), 392-95.
Gamson, J. (1994). Claims to Fame: celebrity in contemporary America. Berkeley:
University of California Press.
Gamson, W. A., & Modigliani, A., (1989). Media Discourse and Public Opinion on Nuclear
Power: A Constructionist Approach. The American Journal of Sociology, 95, 1-37.
Giles, D. C. (2002). Parasocial Interaction: A Review of the Literature and a Model for
Future Research. Media Psychology, 4, 279-305.
Gorin, V., & Dubied, A. (2011). Desirable people: Identifying social values through
celebrity news. Media, Culture & Society, 33(4), 599-618.
Hall, S. (1980). Encoding/Decoding. In S. Hall, D. Hobson, D. Lowe, & P. Willis (Eds.),
Culture, media, language: working papers in cultural studies (pp. 128-138).
London: Hutchinson.
Holmes, S. (2005). Off-guard, Unkempt, Unready?: Deconstructing Contemporary
Celebrity in Heat Magazine. Journal of Media & Cultural Studies, 19(1), 21-38.
Horton, D., & Wohl, R. (1956). Mass Communication and Parasocial Interaction:
Observations on Intimacy at a Distance. Psychiatry, 19, 215-29.
Joinson, A. N. (2001). Self-disclosure in computer-mediated communication : the role of
self-awareness and visual anonymity. European Journal of Social Psychology, 31,
177-92.
Klimmt, C., Hartmann, T., & Schramm, H. (2006). Parasocial Interactions and
Relationships. In J. Bryant, & P. Vorderer (Eds.), Psychology of Entertainment (pp.
291-313). London: Lawrence Erlbaum Associates.
La Pastina, A. C., & Straubhaar, J. D. (2005). Multiple Proximities between Television
Genres and Audiences. The Schism between Telenovelas Global Distribution and
Local Consumption. Gazette, 67(3), 271-88.
Marshall, P. D. (2006a). The Celebrity Culture Reader. London: Routledge.
Marshall, P. D. (2006b). Intimately intertwined in the most public way: Celebrity and

86
 Journalism. In P. D. Marshall (Ed.), The Celebrity Culture Reader (pp. 315-323).
London: Routledge.
McGuigan, J. (2000). British identity and the peoples princess. Sociological Review,
48(1), 1-18.
Miller, T., Govil, N., McMurria, J., & Maxwell, R. (2001). Global Hollywood. London: BFI.
Neuman, W. R., Just, M. R., & Crigler, A. N. (1992). Common Knowledge: News and the
Construction of Political Meaning. Chicago: The University of Chicago Press.
People (2011a). Media Kit: Audience Reach. [online] Available at: <
http://www.people.com/people/static/onlinemediakit/audience_reach.html>
[Accessed 23 February 2011].
People (2011b). Media Kit: Audience Demographics. [online] Available at: <
http://www.people.com/people/static/onlinemediakit/audience.html>
[Accessed at 23 February 2011].
Rojek, C. (2001). Celebrity. London: Reaktion Books.
Rojek, C. (2007). Cultural Studies. Cambridge: Polity Press.
Schiappa, E., Gregg, P. B., & Hewes, D. E. (2005). The Parasocial Contact Hypothesis.
Communication Monographs, 72(1), 92-115.
Straubhaar, J. D. (1991). Beyond Media Imperialism: Assymetrical Interdependence and
Cultural Proximity. Critical Studies in Mass Communication, 8, 39-59.
Straubhaar, J. D. (2003). Choosing national TV: Cultural capital, language, and cultural
proximity in Brazil. In M.G. Elasmar (Ed.), The impact of international television: A
paradigm shift (pp. 77-110). Mahwah, NJ: Lawrence Erlbaum Associates.
Tian, Q., & Hoffner, C. A. (2010). Parasocial interaction with characters on a popular TV
series. Mass Communication and Society, 13(3), 250-69.
Tolson, A. (2001). Being yourself: The pursuit of authentic celebrity. Discourse Studies,
3(4), 443-57.
Turner, G. (2004). Understanding Celebrity. London : Sage.
Turner, G. (2010). Approaching celebrity studies. Celebrity Studies, 1(1), 11-20.
Turner, G., Bonner, F., & Marshall, P. D. (2000). Fame Fames: The Production of Celebrity
in Australia. Cambridge: Cambridge University Press.
Van den Bulck, H. (2011). Media: Structuur en werking. Antwerpen: Acco.

87
Van den Bulck, H., & Claessens, N. (2013). Celebrity Suicide and the Search for the Moral
High Ground: Comparing Frames in Media and Audience Discussions of the Death
of a Flemish Celebrity. Critical Studies in Media Communication, 30(1), 69-84.
Van den Bulck, H., & Van Gorp, J. (2011). Eternal Fandom: Elderly Fans, the Media and
the Staged Divorce of a Schlager Singer. Popular Communication, 9(3), 212-26.
Van Gorp, B. (2006). Een Constructivistische Kijk op het Concept Framing. Tijdschrift
voor Communicatiewetenschap, 34, 246-56.
Van Zoonen, L. (2007). Audience Reactions to Hollywood Politics. Media, Culture &
Society, 29, 531-47.
Walther, J. B. (1996). Computer-Mediated Communication: Impersonal, Interpersonal,
and Hyperpersonal Interaction. Communication Research, 32(1), 3-43.
Zaharopoulos, T. (1990). Cultural Proximity in International News Coverage: 1988 U.S.
Presidential Campaign in the Greek Press. Journalism Quarterly, 67(1), 190-4.

88
Appendix
1. Codebook
ARTICLES
V1. Number
V2. Medium
V3. Date
V4. Title
V5. Number of reactions
V6. Which celebrity is discussed?
V7. Celebrity: Gender
V8. Celebrity: Nationality
V9. Celebrity: Relationship status
V10. Celebrity: Ethnicity
V11. Celebrity: Sexual orientation
V12. Celebrity: Domain
V13. Who is talking?
V14. Framing devices
V15. Problem (who, what)
V16. Cause (who, what)
V17. Moral judgment
V18. Solution.
V19. Who is talking? (2)
REPETITION OF V14-V19
V20. Who is talking (3)
REPETITION OF V14-V19
REACTIONS
V1. Number
V2. Reaction to article number X
V3. Medium
V4. Date
V5. Name reader
V6. Gender reader
V7. Which celebrity is discussed?
V8. Celebrity: Gender
V9. Celebrity: Nationality
V10. Celebrity: Relationship status
V11. Celebrity: Ethnicity
V12. Celebrity: Sexual orientation
V13. Celebrity: Domain
V14. Framing devices
V15. Problem (who, what)
V16. Cause (who, what)
V17. Moral judgment
V18. Solution.
V19. Parasocial interaction
V20. Interaction with peers
V21. Personal experiences
89
 Chapter 2

The large-scale, systematic findings of the first chapter are explored in depth by means of two
case studies in chapters 2 and 3. Indeed, the second empirical chapter of this PhD presents a
case study of the media coverage of and audience reactions to a high profile celebrity death in
Flanders. Similar to chapter 1, this study combines the elements of text and reception by
comparing media and audiences. The method of framing analysis mainly employed in
(political) communication studies but transferred here to celebrity studies provides insight in
how the celebrity suicide triggers melodramatic articles about and responses to the beloved
celebrity as well as more profound discussions of moral issues. Indeed, a variety of media and
audience frames are found to discuss divorce, homosexuality, suicide, and abuse. Whereas
chapter 1 indicates that most audience responses to celebrity news are superficial, this case
study demonstrates that celebrity news has an important social role as a lead into social
discussions of moral issues in contemporary Western societies. Here, there is a difference
between the celebrity coverage of popular and quality media contrasting claims of a no brow
culture as well as a difference between media and audience frames. Audience members are
found to adopt, negotiate, or reject media frames depending on their personal experiences,
interaction with peers, and parasocial relationships with the celebrity.

CELEBRITY SUICIDE AND THE SEARCH FOR THE MORAL HIGH GROUND:
COMPARING FRAMES IN MEDIA AND AUDIENCE DISCUSSIONS OF THE DEATH OF A
FLEMISH CELEBRITY

Hilde Van den Bulck and Nathalie Claessens

(Published in Critical Studies in Media Communication, 30(1), 69-84)

Taking a case study and framing approach, this article analyzes media coverage of and
(on- and off-line) audience reactions to a gay celebritys death and its aftermath to see
how it triggers discussion about socially and ethically sensitive issues. It starts from the
assumption that the celebrity construct is not just a commodity but also acts as a forum
to discuss social and ethical issues, norms, and values. Media coverage of the suicide of
Flemish singer Yasmine is revealed to be dominated by frames relating to the soap of life
(esp. divorce), the celebritys status as a lesbian icon, and her position as a victim of
abuse, with framing differing according to media and journalistic style. Audience
reactions do not simply mimic these frames, but are mediated in different ways by
personal experiences, peers, and parasocial relationships, resulting in active negotiations
of the media frames, even creating counter-frames, particularly regarding the act of
suicide and the gay status of the celebrity.

90
Introduction
On June 25th 2009, Famous Fleming1 Yasmine (real name Hilde Rens) died by her own
hand. Yasmine owed her celebrity status to a career as respected singer and television
personality, and a position as gay icon. During her life, media and fans were by and large
respectful of Yasmines public and private life choices. Yet her death created a frenzy
with press and audiences that rivaled the attention for the death of pop icon Michael
Jackson the same week, and that ignited discussions on social and ethical issues, norms,
and values. Adopting a case study and framing approach, this article analyzes how a
celebrity death and its aftermath can lead to social and ethical discussions among media
and audiences: how media frames differ according to the type of news media and how
(online) audience reactions to these media frames are mediated by media preferences,
personal experiences, peers, and parasocial relationships. While wider academic
relevance of a case study is hampered by its idiosyncrasies, it allows for a more in-depth
analysis of the contribution of relationships between celebrity (event), media, and
audiences to social and ethical discussions. Similarly, a shock event may not provide
generalizable insights into the day-to-day role of celebrities in culture and society, yet
other examples such as the death of pop icon Michael Jackson indicate its wider
relevance (cf. Van den Bulck & Panis, 2010).

Celebrity, media and audience

The article starts from the notion of a celebrity as a construct, resulting from
negotiations between the person looking for fame, the media, and the audience. A
celebrity is a mediated persona (Evans & Hesmondhalgh, 2005), critically consumed and
molded by audiences (Marshall, 1997) that engage in parasocial relationships with the
celebrity (Giles, 2002; 2003). This celebrity construct, moreover, is a mediated interplay
between a public persona based on public activities, a private persona based on the
private life as presented to the world and, finally, glimpses of the real person behind the
image off-guard, unkempt, unready (Holmes, 2005: 24). In Flanders, as elsewhere,
the relationship between celebrity, media, and audience is influenced by ongoing
processes of media commercialization and tabloidization (Deuze, 2005; Van den Bulck &

1Term commonly used to refer to local celebrities in the Northern, Flemish-speaking part of
Belgium (60% of Belgiums population of 11 million).

91
Tambuyzer, 2008). This has led to a growing stress on the private and the scandalous,
making obsolete any sense of a celebritys privacy.
Marshall (1997), Rojek (2001), and Cashmore (2006), amongst others, underline
the position of celebrities as commodities, created to embody capitalist society as both
objects and vehicles of consumption, encouraging and validating consumer culture
(Cashmore, 2006: 269). As such, mediated communication with and about celebrities
constitutes returning examples of Halls (1980) original contention that media messages
express the dominant ideology of market capitalism. Audiences, in turn, are seen to
develop strong attachments with celebrities, primarily to compensate their frustrations
in capitalist society, i.e., for failing to reach the dream of becoming rich and glamorous.
Building on Morins work, Rojek (2001: 15) therefore considers celebrity as the
expression of the frustrated desires of the audience. To the extent that audience
reactions differ, this is seen to be rooted in different social/class backgrounds (e.g.
Sandvoss, 2005).
While the celebrity-as-commodity argument is strong, this article wishes to move
beyond this view to study if and how celebrities are more than consumer goods and the
audience reception hereof is more than a class-based reaction to the capitalist system, to
explore the way in which celebrities serve wider social functions as a conduit to discuss
social and ethical issues, norms, and values (Alberoni, 1972; Butler Breese, 2010).
Liberated from the tutelage of state and church and its top-down socialization (Van
Poecke, 1996), contemporary society and its citizens-consumers are expected to actively
create and negotiate their identities and belief systems bottom-up. Celebrities attitudes,
ideas, and behavior can provide a starting point to discuss such meaning systems. This
article analyzes how media and audiences use celebrities ideas and behavior to discuss
what is socially and ethically acceptable or not. In doing so, we wish to explore the ways
in and extent to which audiences diverse readings (i.e., hegemonic, negotiated, or
oppositional (Hall, 1980)) of the mediated celebrity communication are determined by
factors other than social/class background, including media preferences, personal
experiences, interaction with peers, and parasocial relationship with the messages
subject.

92
Celebrity death and social issues, norms and values
A vantage point for such a study is a celebrity death. When a celebrity dies, his/her
status is often elevated to sanctity, stemming from the impossibility to disappoint after
death (Lumby, 2006: 537). Increasingly, though, it is also an occasion to unravel his/her
private life, emphasizing the scandalous, especially with a sudden or unexpected death
(Lumby, 2006: 544). Kitch (2005: 69) hints to the fact that a celebritys death is a
moment for public discussion of shared ideals and identities, particularly when a
celebritys behavior, ideas, and/or values are at the fringe of social norms and values. As
such, the death of Famous Fleming Yasmine is an interesting case as it involves a
celebrity whose private life had gained public attention, who was openly gay, and died
by her own hand. We therefore expect that this celebrity death will provoke social and
ethical debates in media and among audiences about norms and values.
One area of discussion is the (extensively covered) marital break-up of Yasmine
two months before her death. As in many countries, in Flanders divorce has become a
fact of life with approximately 15,000 couples ending their marriage each year, a
European record (SiV, 2011). Only 13% of the population considers divorce as never
justifiable (Halman et al., 2008: 238). Yet, it is still considered problematic as it requires
new interpretations of relationships with former partners, children, and in-laws. We
therefore anticipate Yasmines death to provoke discussions about the institution of
marriage, divorce, and the aftermath.
A further potential element for debate is Yasmines position as lesbian icon. On
the one hand, Flanders is part of a progressive country (Belgium) when it comes to gay
rights: it was the second country to legalize same-sex marriages in June 2003, followed
in April 2006 by a bill allowing legal adoption by same-sex couples.2 This indicates a
move away from the rejection of homosexuality that typified traditional Catholic
societies, a shift reflected in a more positive representation of gays in Flemish (as in
other European) media content since the 1990s (Fejes & Petrich, 1993). Similarly, after a
long history of denial of celebrity homosexuality (Rojek, 2001: 86), there has been an
increase since the 1990s in (the media presence of) positive gay celebrity role models
such as Ellen DeGeneres and, indeed, Yasmine (Dow, 2001). At the same time, prejudices
and discriminations persist. Despite a decrease of 26% since 1981, in 2000 still 26% of

http://www.belgium.be/nl/familie/koppel/huwelijk/homohuwelijk/ and
2

http://www.belgium.be/nl/familie/kinderen/adoptie/

93
the Flemish population considered homosexuality to be never justifiable (Halman et
al., 2008: 235). We therefore expect that a lesbian celebritys death provokes discussion
about the position of gays and lesbians in society.
A final topic for possible debate is the fact that Yasmines death was self-inflicted.
With a suicide rate of 16.1 per 100,000 inhabitants, Flanders scores high in European
perspective (Reynders, Van Heeringen, De Maeseneer, & Van Audenhove, 2009).
According to cross-national research, this should predict more tolerant attitudes toward
suicide (Stack & Kposowa, 2008: 39). However, Reynders et al. (2009) found that
Flemish people justify suicide less often than people in countries with lower suicide
rates. Researchers relate this to Flanders as a former mono-religious catholic society,
where suicide was a mortal sin, unacceptable and not to be tolerated (Kerkhof,
Arensman, Neelenman, & de Wilde, 2003). Despite the secularization of Flanders since
the 1960s (Billiet, Maddens, & Frognier, 2006), 45% of the Flemish population still
believes suicide is never justifiable (Halman et al., 2008: 240). These conflicting data
predict that a celebrity suicide provokes discussion about the acceptability of suicide.
The topics potential sensitivity is further suggested by the development in 2004 of
Recommendations for Responsible Suicide Coverage by the Flemish government, the
Flemish organization of professional journalists VVJ, and the Flemish Press Council
(Werkgroep Verder, 2007). Aimed at media professionals covering suicide, the
guidelines show strong resemblance to those formulated by the WHO and the American
Foundation for Suicide Prevention and include, amongst others, the recommendation to
avoid unnecessary details on the method of suicide to reduce the risk of imitation (see
Sonneck, Etzersdorfer, & Nagel-Kuess, 1994). They result from insights linking
imprudent or extensive media coverage of (celebrity) suicides to a series of imitative
suicides in the past (Chen et al., 2010; Martin, 1998; Stack, 1987).

Framework for Analysis

To understand how the death of Flemish celebrity Yasmine evokes media and audiences
to discuss meaning systems, we opt for a framing approach as, more than other methods
for content and audience analysis, it allows for the study of both encoding (media
frames) and decoding (audience reactions) with the same analytical tool and, moreover,
for the immediate identification of the core aspects (problem, cause, moral judgment,
and solution) relating to belief systems. Framing is understood as a process whereby a

94
frame determines which aspects of reality are selected, rejected, emphasized or
modified in the production of a media text and, at the same time, provides the audience
with a context and suggested meaning (Van Gorp, 2006: 46). Framing, in other words,
allows us to analyze the interplay between the textual level (media frames), the
cognitive level (audience schemata), the extra-media level (discourse of frame
sponsors), and shared cultural ideas as a macrostructure (Van Gorp, 2006: 247). A frame
serves as a meta-communicative message that displays itself in a text through various
framing devices, such as word or image choice, metaphors, symbols and stereotypes
(Pan & Kosicki, 1993: 67). Reasoning devices relate to the four functions of a frame:
problem definition, causal interpretation, moral evaluation, and treatment
recommendation (Entman, 1993: 52).
Our first aim is to analyze the framing (or encoding) of Yasmines death and
aftermath in several Flemish media. Since media frames are related to wider social and
cultural belief systems and since different media adhere to different types of journalism,
we predict that the framing of Yasmines death in meaning systems differs according to
journalistic type.
Media audiences, in turn, are confronted with these frames which prime or
raise the salience or apparent importance of certain ideas [...] that encourage target
audiences to think, feel, and decide in a particular way (Entman, 2007: 164). Yet, the
mere presence of a frame in a text does not guarantee its influence on audiences.
Researchers have pointed to a list of intermediating factors, such as a successful fit
between audience and media values, norms, and practices (Gamson & Modigliani, 1989),
interaction with peers (Neuman, Just, & Crigler, 1992), and personal experiences and
cognitions that audiences bring into play to make sense of media content (Broos & Van
den Bulck, 2008; DAngelo, 2002). As a result, media frames can be consumed (or
decoded) by audiences in different ways. We therefore expect that an audience
members interpretations of media frames is mediated by his/her media preferences,
personal experiences, and interaction with peers.
One further element, not mentioned in framing literature but expected to
influence audiences consumption of media frames, is the parasocial relationship with a
celebrity. This refers to the illusion of a one-sided, long-term friendship between an
audience member and a celebrity in which a large group of people (the audience) knows
a lot about one person (the celebrity) who, in turn, has no knowledge of all those people

95
except in abstract terms. It is the illusion of a face-to-face relationship, created after
multiple mediated encounters in which a high level of attraction is present (Giles, 2002;
Horton & Wohl, 1956). The parasocial bond can be particularly strong and emotional in
the case of fans (Sandvoss, 2005) but also occurs with regular audiences. The
relationship is a determining factor in the impact of celebrity (stories) on audience
members (Brown, Basil, & Bocarnea, 2003). Sometimes a parasocial relationship simply
provides material for gossip (Giles, 2002: 191), at other times it can help alter audiences
attitudes about social and cultural issues and groups, particularly when they have little
knowledge of or direct contact with the issue or group (Schiappa, Gregg, & Hewes,
2006). We anticipate that an audience members interpretations of media frames is
mediated by his/her parasocial relationship with the celebrity.

Methodology
This framing approach is applied to our case study that is part of a wider research
project studying contemporary celebrity culture.3 The data comprise of all media
coverage and readers responses regarding the death of Yasmine between June 25 2009
Yasmines death and September 25 2009 when news of her death and its aftermath
had died down. The media sample contains two quality (De Standaard, (Corelio,
circulation 83,000) and De Morgen (De Persgroep Publishing, 70,000)) and two popular
(Het Laatste Nieuws (De Persgroep Publishing, 288,000) and Het Nieuwsblad (Corelio,
260,000)) Flemish daily newspapers, the two main Flemish celebrity gossip magazines
Dag Allemaal (De Persgroep Publishing, 405,622) and Story (Sanoma, 205,831), and
weekly television guide and magazine Humo (Sanoma, 221,485). The corpus of audience
responses contains all reactions as they appeared on the selected medias websites
(respond to this article) and in the printed editions (letters to editor).
Analyzing online audience reactions can be motivated by the more natural
results that arise from mundane research contexts (such as online discussion sites)
rather than those obtained in unnatural settings like experiments, interviews, or focus
groups (Van Zoonen, 2007: 534-535; Zhou & Moy, 2007: 83). One disadvantage is that
only a certain type of audience member is likely to engage in online discussions, while a

3Funded by an internal grant of the University of Antwerp BOF-NOI funds and an external grant
of the Flemish scientific fund FWO, both obtained competitively and after peer review.

96
majority of online news consumers are lurkers, following but not actively taking part
in online discussions (Beyers, 2004). Others are not online at all. Moreover, we often do
not know who is behind the posting, why the individual sent in a comment or how
postings relate to peoples real-life politics (Van Zoonen, 2007: 535). However, the
latter two also occur in face-to-face communication while online comments provide
useful information about how people present themselves to others (Van Zoonen, 2007:
535).
The sample was coded through inductive framing analysis using a self-composed
topic list4 based on theoretical and empirical insights from previous studies (see
Appendix 1). All answers were put together in a matrix in search for similarities and
patterns and overall consistencies were combined into frame packages (Gamson &
Modigliani, 1989) that were tested for mutual exclusiveness. Frame salience was
operationalised in frequency of occurrence (Zhou & Moy, 2007: 88).

Results
Overall, 176 articles and 1,399 audience reactions relating to Yasmines death and
aftermath were found (see Table 1 for distribution per medium). We should note that
newspapers De Morgen and Het Nieuwsblad, and all magazines do not provide explicit
possibilities for online reactions and thus have a much lower response rate than the
others.

Table 1. Percentages of articles/reactions per medium

Media Media Articles Audience Reactions
De Standaard 29.0% 32.0%
Het Nieuwsblad 23.3% 4.2%
Het Laatste Nieuws 10.2% 60.8%
De Morgen 9.1% 1.1%
Story 18.8% 0.0%
Dag Allemaal 8.5% 1.6%
Humo 1.1% 0.3%
TOTAL 100% (N = 176) 100% (N = 1,399)

41 What/who is the problem? 2 What/who caused this? 3 What moral evaluation is provided? 4
What is the recommended treatment?

97
Framing Yasmines death
Analyzing our data, we found a considerable portion of media and audience response
items that stuck to a general Celebrity Death frame (see Table 2 for percentages of
frames in media articles and audience responses), focusing on Yasmine life as a singer
and television presenter. Vocabulary and tone are predominantly melodramatic (we
will miss you) but positive towards Yasmine who is glorified unconditionally as a
successful and talented performer, a wonderful and strong woman, an angel.

Table 2. Percentages of frames in media articles/audience responses

Frames Popular Media Quality Media All Media
(%) Media Audience Media Audience Media Audience
Celebrity death 49.4 70.4 39.0 33.1 45.1 58.6
All Saints 12.3 2.5 18.6 1.9 15.0 2.3
Saint vs. Evil Ex 11.1 5.1 0.0 0.3 6.4 3.5
Gay Icon 17.3 0.0 27.1 0.8 21.4 0.3
Lesbian Exception 0.0 0.0 0.0 1.6 0.0 0.5
Gay Rejection 0.0 0.0 0.0 4.6 0.0 1.5
Selfish Coward 0.0 6.3 0.0 14.8 0.0 9.0
Victim of Abuse 9.9 15.7 15.3 42.9 12.1 24.3
TOTAL 100.0 100.0 100.0 100.0 100.0 100.0
N 81 791 59 366 140 1,157

Our data further revealed social and ethical discussions on four topics: the soap
of life, lesbianism, suicide, and abuse - resulting in a number of frames. The first topic is
the soap of life, in particular the issue of divorce for which two frames were found. The
All Saints frame tells a story in which Yasmines suicide is seen as the sad outcome of
the separation (see Rothblum (1990) on depression following lesbians break-up) but no
one is blamed. This frame can be compared to the US no-fault divorce system, in which
divorce is seen as a private decision with no identification of victim or victimizer and no
pointing of blame (Adams & Coltrane, 2007). This frame allocates the position of victim
not only to Yasmine but to all involved: former wife Marianne, their daughter, families,
fans, and colleagues. Tone and vocabulary are melodramatic (grieving) but positive: My
commiserations extend to Marianne, who must also be suffering right now (reaction to
Hoefkens, 2009). The second is the highly mythical Saint versus Evil Ex frame in which
Marianne is blamed for the tragic events following the break-up. This echoes the older
and adversarial fault divorce system where victim is pitted against victimizer, focusing
on marital norms rather than individual choice (Adams & Coltrane, 2007). The wicked
ex-wifes alleged behavior (cheating on and dumping Yasmine) is considered the cause
of Yasmines suicide. Vocabulary and tone are melodramatic and fairy-tale like, starring

98
Marianne as the evil villain and Yasmine as the princess-like victim: Yasmine looks
sweet and soft, Marianne looks like a dodgy cold Evil Ex (reaction to D.A. & L.B., 2009).
In both frames, romantic relationships are discussed as central to our lives and
happiness, and the break-up hereof as a main cause of depression. The issue of divorce is
dealt with quite centrally, either to refute any negative connotations (All Saints) or to
promote it as the main reason behind the tragic death (Saint vs. Evil Ex). Yasmines
death thus results in a discussion of relationships and emotional wellbeing.
Three frames discuss Yasmines death in relationship to her life choice as a
lesbian. A dominant Gay Icon frame focuses on the celebritys iconic position as a
lesbian which is seen to represent current acceptance of gays in Flemish society and
Yasmines status as a role model for young gays (see Dow (2001) on gay role models).
This frame is sponsored mainly by fellow celebrities and members of the gay movement
who state that Yasmine was our icon (De Bock, 2009b). The second frame considers
Yasmine as a Lesbian Exception, questioning the acceptance of gay people in society
and viewing this rejection as a problem that can cause emotional and mental
breakdowns and suicide (see Rothblum, 1990; Russell & Joyner, 2001). In this frame, the
celebration of Yasmine as an icon is seen to conceal the problems homosexual people
are dealing with on a daily basis. For instance, her death is used to discuss the higher-
than-average suicide rate among lesbians. An article titled They struggle with their
sexual orientation (De Bock, 2009a) describes how 50 percent of all lesbians between
18 and 23 has seriously considered suicide. In the third, Gay Rejection frame,
Yasmines private life choice as a lesbian and homosexuals in general are criticized as
unacceptable (see Wood & Bartkowski (2004) on homophobia): Statistically, gay
people do not only act ten to twelve times more promiscuously than straight people, but
[] are also more threatening when it comes to the sexual abuse of children (Reaction
to De Bock, 2009a).
Different than expected, the death of Yasmine is hardly ever framed explicitly as a
socially or morally debatable suicide. Instead, the suicide is mostly framed as the
result of other issues such as depression and the end of relationships. This reflects
Szaszs (2002) findings that suicide is no longer considered as a sin but rather a mental
illness. A limited number of audience reactions, however, reveal a Selfish Coward frame
in which Yasmine is criticized for committing what is considered a cowardly and selfish
act, abandoning her fans, friends, family, and two-year old daughter.

99
 Unexpectedly, our analyses revealed a frame discussing the celebrity as a
(posthumous) Victim of Abuse, covering the (problematic) relationship between
Yasmine and the world around her. The focus is not on Yasmine or her death, but on
events happening as a side-effect, negatively affecting Yasmine and the people around
her: an eBay trade in Memoriam cards, the popular medias persecution of Marianne,
and the fact that unauthorized policemen went through Yasmines deaths police file
without permission. All issues are condemned as scandalous practices at the expense of
Yasmine and those who mourn her.

Media framing Yasmines death and its aftermath

Popular vs. Quality Media. Looking at media frames, certain differences between
quality and popular news media can be observed (see table 2). Quantitative framing
analysis shows that popular media tend to cover the death and its aftermath more in
terms of the Soap of Life (23.4% or 19 articles) while the quality media focus more on
the celebrity as Victim of Abuse (15.3% or 9 articles). This is reflected in how the
celebrity is approached: quality media predominantly cover the private persona (73%
or 27 articles) whereas popular media are interested in the private (64.7% or 33
articles) as well as the real (21.6% or 11 articles) celebrity personae.
The first weeks after Yasmines death, all media, and particularly the quality
press, stick to a positive coverage within the Celebrity Death and Gay Icon frame. The
media differ, though, in the discussion of her love life. Quality media report this in an All
Saints frame, with no negative references to the ex-wife while 11.1% (9 articles) of all
popular media coverage blames Marianne within a Saint vs. Evil Ex frame. This frame is
sponsored through allegations from Yasmines family and fellow celebrities. Looking
more closely (and qualitatively), most media portray Yasmine as a victim, except quality
newspaper De Morgen that stresses her personal choice in death without condemning it.
The quality media ignore the cause of Yasmines death by her own hand or attribute it to
situational circumstances, while popular media point to Marianne as the reason for the
break-up and, consequently, Yasmines death. Another notable difference is the presence
of details surrounding the suicide, elaborated on by popular media while almost entirely
absent in quality media.

100
 Whereas the Saint vs. Evil Ex frame remains dominant in some popular media
(Dag Allemaal, Het Laatste Nieuws), quality newspaper De Standaard and the other
popular media (Story, Het Nieuwsblad) join quality media De Morgen and Humo in their
All Saints frame, seeing Marianne as fellow victim, reporting on Mariannes deteriorating
health and portraying her as suffering.
Popular and quality media further differ in their use of the Victim of Abuse frame,
arising in the fourth week after the celebrity death. Apart from some popular
newspapers coverage of the eBay trade in Memoriam cards and of policemen going
through Yasmines files unauthorized, popular media focus on Victim of Abuse stories of
Yasmines private circle. The quality papers, employing this frame much more
frequently, use it to question the popular medias persecution of Marianne. They
extensively discuss and condemn the scandalous popular media coverage of Yasmines
death, pointing to a lack of ethics, a dominance of commercial considerations, and the
publics cry for sensational news as reasons for such degraded journalism. Popular
media do not reflect on journalistic standards. When legally forced to publish Mariannes
Right of Reply, popular magazines Story and Dag Allemaal show no critical distance
towards their coverage but try to clear their name (and to accuse competing magazines).
Finally, both popular and elite newspapers avoid the topic of suicide as such, even
formerly explicitly Catholic newspapers De Standaard and Het Nieuwsblad, and more
conservative gossip magazine Dag Allemaal. All media choose their vocabulary carefully:
they hardly ever use the popular term self-murder in favor of more neutral terms that
avoid any aggressive connotation of the act. This may indicate knowledge of the
journalistic guidelines but also the medias wary attitude (Szasz, 2002).

Audience reactions to media frames.

Media preference. Analysis of audience reactions shows, first, that they differ
according to media type. Reactions clearly echo the difference in tone of various media:
a majority of comments to (melodramatic) popular media are emotional (69.6% or 537
reactions) whereas reactions to the more critical quality media are mostly distant and
neutral (84.1% or 332 reactions).
Most melodramatic audience responses deal with the passing of the celebrity,
either positively within the Celebrity Death frame (93.5% or 544 reactions) or
negatively in a Selfish Coward frame (5.2% or 30 reactions). Critical reactions mostly

101
cover the celebrity as a Victim of Abuse (53.3% or 252 reactions) and the suicide itself
(30.3% or 143 reactions), either in the neutral Celebrity Death (15.9% or 75 reactions)
or critical Selfish Coward frame (14.4% or 68 reactions). Like the media, most reactions
to popular articles focus on Yasmines death and her personal life (84.3% or 667
reactions), while those to quality media generally cover social and ethical issues such as
evolutions in the press (42.9% or 157 reactions) and acceptance of homosexuality (7%
or 26 reactions) (see table 2). Interestingly, 40 out of 60 audience reactions to popular
media articles about the Soap of Life adopt the Saint vs. Evil Ex frame: I never trusted
you, Marianne [] you are a cold, pitiless bitch, shame on you (reaction to D.A. & L.B.,
2009), compared to only one quality media reaction about the Soap of Life. That is not to
say that audiences follow the media uncritically. Analyzing further, it appears that the
way in which audience members deal with media frames is mediated by a number of
factors.
Personal Experiences. When dealing with media frames, audiences bring their
own experiences - and associated views and feelings - into their responses to the media
coverage. Every topic presented by the media, is addressed from different angles with a
myriad of references to audiences personal experiences. For instance, one reader states:
I know what a broken love can do to you. You could call it good (or as I see it, bad) luck
that they found me in time [when I tried to commit suicide] (reaction to De Pourcq,
2009). Other readers refer to their own loss: I know what it is like to lose loved ones
through suicide (reaction to De Pourcq, 2009).
Interestingly, audience responses reveal counter-frames that did not appear in
the media coverage and that are found to originate from personal experiences, resulting
in strong views on a certain issue. These seem to make an audience member more likely
to reject the media frame. For instance, 104 reactions (9.0%) express a counter-frame
criticizing suicide as a cowardly and selfish act (Selfish Coward frame):

I lost someone very precious to me, someone who fought in the hospital for three
months, someone who wanted to go home more than anything. When I read
about young, healthy people who give up on life like this, I cannot feel any
sympathy, ab-so-lute-ly not. (reaction to Hoefkens, 2009)

Some reactions to quality media express counter-frames to the medias positive

or neutral stance on homosexuality. Introducing a Lesbian Exception counter-frame, six
quality media reactions (1.6%) consider homosexuality as not accepted in society,

102
making it tough for gays because they are regarded as different: It is still hard for
people to come out (De Bock, 2009a). 17 Quality media reactions contain a Gay
Rejection counter-frame (4.6%), criticizing the homosexual community and its behavior:

The Gay Movement is the first to complain about others, while nobody points a
finger at them. These days one is expected to be ashamed to admit that one does
not like sharing a dressing room with a gay person. Where will it end? (reaction
to De Bock, 2009b)

In both counter-frames, audience members explicitly condemn the uncritical media

coverage for promoting homosexuality as unproblematic or positive.
Interaction with Peers. Interestingly, such counter-frames often spark heated
audience discussions in which some readers agree with the media frame, while others
adhere to counter-frames and express their arguments to do so, even seeking to win
over other readers. These differences of opinions and views, voiced in the audience
reactions, mediate frame acquisition. For example, a reader of De Standaard reacts to
someone criticizing homosexuality: Mister [X], I have to say that I believe your rational
arguments entail a certain fear of homosexual people, or a sense of discomfort
(reaction to De Bock, 2009b). Such statements evoke subsequent reactions from the
reader under attack and others, culminating in a debate about homosexuality and
freedom of speech in which audience members views become polarized.
Parasocial Relationship. As stated, some audience reactions contain deep
emotional responses to Yasmines death. Some of this melodrama does not simply mimic
that of the coverage but appears to be linked to degrees of parasocial relationships
audience members maintain with the deceased celebrity. People express bereavement in
a fashion similar to reactions to the death of a relative or friend. In reactions, Yasmine is
often talked to as if she will read it, addressing her with Dear Yasmine, Dear Hilde,
and comments like: I hope you will find the peace you could not find here, We will
miss you. The media are given the role of intermediary in the (imaginary)
communication between celebrity and audience. Interestingly, on several occasions
journalists themselves write about their parasocial relationship with Yasmine: Goodbye
Yasmine, we respect your choice, but well miss you so much (Beyers, 2009).

103
Discussion
At one level, our study indicates that the death of Famous Fleming Yasmine and
treatment hereof by media and audiences provides an(other) illustration of the
celebrity-as-commodity thesis of celebrity studies scholars like Rojek (2001). Celebrity
news, particularly of an unexpected, dramatic nature such as a sudden death, provides a
means to sell copies, and so Flemish media pay ample attention to the death and its
aftermath. Quality papers too, under pressure to keep their readership, report
increasingly and quite uncritically about celebrity culture (Van den Bulck & Tambuyzer,
2008). About half of the media items (45.1%) and audience reactions (58.6%) treat the
event within a Celebrity Death frame, focusing on the exceptional talent and fame of the
singer/presenter, melodramatically celebrating and commiserating the commodity that
is the celebrity Yasmine. Audiences deeply emotional expressions of attachment (we
will miss you) can be interpreted as reflecting frustrations with their own position
within the capitalist system. Yet, the other half of the media coverage (54.9%) and
audience reactions (41.4%) frame the celebrity death as a story about what is socially
and ethically right and wrong, provoking debates about norms and values, that go well
beyond the positions of both celebrities and audiences in the capitalist system, thus
modifying the celebrity-as-mere-commodity thesis. Instead, it appears, in Butler
Breeses words (2010: 337), that celebrities are symbols by which we narrate,
negotiate, and interpret our collective experience, and establish moral boundaries.
In this, the celebritys private life and speculation about the real life provide
much of the food for thought. The press and its readership take the sad event as a lead to
discuss views on relationships, focusing on love, heartbreak, fidelity, and divorce.
Overall, relational bliss and marriage are presented as cornerstones of a happy life,
while media and audience fractions discuss different views on the nature of relational
break-ups (All Saints or Saint vs. Evil Ex) and on what is socially and morally acceptable
behavior both in and after a romantic relationship. Interestingly, the Soap of Life also
provokes discussion about depression and its (questioned) status as disease still a
taboo topic in contemporary society (Reynders et al., 2009). As such the celebrity event
seems to provide media and audience with an occasion to renegotiate prejudices and
socially sensitive topics.

104
 Similarly, and different from much of the literatures observations regarding the
difficult relationship between (famous) gays and the media (Gross, 1993), Yasmines
lesbianism is not criticized or brought into play by the media to explain her sudden
death. Instead, her status as lesbian icon is acknowledged and taken as an occasion to
discuss the position of gays and particularly lesbians in society. This seems to confirm a
growing acceptance of gays in Flemish society, suggested by the 2003 legalization of gay
weddings. Yet, counter-frames question this, as several reactions either point to the
difficulties for gays and lesbians (Lesbian Exception) or question homosexuality as such
(Gay Rejection). The interplay between media and audience views indicates that
homosexuality is still socially debated.
The issue of suicide most clearly illustrates that the media do not take the lead at
all times, since audiences fail to follow the medias moral guidance of acceptance. The
media did not explicitly frame Yasmines death as a case of (un)acceptable suicide. This
provoked a variety of reactions from readers, including a less tolerant Selfish Coward
counter-frame. Clearly, a part of the audience opposed the (implicit) media acceptance
of suicide. The medias avoidance of suicide as dominant topic indicates acceptance
following the secularization of society and/or conformity to the Recommendations for
Responsible Suicide Coverage. While the latter explanation seems confirmed by the
medias vocabulary (avoiding self-murder), both All Saints and Saint vs. Evil Ex frames
go against the guidelines. The latter ignores the advice to acknowledge the complexity of
suicide by framing it as the result of a single event (the break-up) (Werkgroep Verder,
2007). It dramatizes the story surrounding the suicide, thereby ignoring the privacy of
Yasmines relatives and friends. The All Saints and Celebrity Death frames go against the
advice by glorifying the deceased, addressing Yasmines public and private celebrity
persona in a sanctified manner, heightening imitation risks (Werkgroep Verder, 2007).
Although quality as much as popular media pay considerable attention to the
death and aftermath, suggesting that contemporary consumer culture is doing away
with class and cultural distinctions in favor of a no brow culture (Swirski, 2005), we
found considerable differences and even a rift between quality and popular media that
can be brought back to cultural distinctions (Bourdieu, 1979). The quality press fierce
attack on the popular medias handling of Yasmines death and aftermath, with quality
papers and their readership distancing themselves from the popular media by taking a
more critical stand, suggests that patterns of inclusion and exclusion remain and are

105
reflected and articulated in media and the reactions of the readership they cater to
(Abercrombie & Longhurst, 1998: 175). As such, the fact that the majority of audience
reactions of popular and quality papers reflect the respectively emotional and critical
type of coverage, is an indication of the social/class based nature of audiences media
preferences and of their decoding of media messages (cf. Hall, 1980).
At the same time, though, our study shows that the way in and extent to which
audiences follow media frames or produce counter-frames, i.e. decode the media
coverage in a hegemonic, negotiated or oppositional way, is mediated by personal
experiences, peers, and parasocial relationships, transgressing social/class based
distinctions. For one, peoples experiences can either strengthen the accordance with the
media frames or evoke counter-frames, depending on the nature of the personal
experiences. Counter-frames also arise within discussions among peers in which
personal views sometimes become polarized, as people try to convince the others of
ones own right. Parasocial relationships not only influence the moral judgment of the
celebrity (positive when there is a relationship), but also the emotional intensity of
audience reactions to even non-dramatic reports. This suggests that parasocial
bereavement is not just about the expression of frustrations with the capitalist system,
but, as Giles (2003: 119) observes: mimics the grief we experience from the death of a
close friend or relative. More research is needed to better understanding the role of
these intermediating factors in the way people deal with media frames, yet it is clear
that these decoding practices are about more than frustrations with the capitalist
system.
Obviously a case study focusing on a single celebrity in a relatively small
community such as the Flemish, can only say so much. Moreover, our results are
influenced by certain limitations, with the absence of wider audience reactions as the
most important one. People writing letters to the editor or posting online reactions may
not constitute a representative sample of the audience. Still, this study has allowed us to
move beyond the interpretation of the celebrity construct as the epitome of consumer
capitalism to see how celebrities provide media and audiences with material to
negotiate wider social, cultural, and moral issues, showing both media and audiences as
internally diverse with different and sometimes competing fractions.

106
References

Abercrombie, N., & Longhurst, B. (1998). Audiences: A sociological theory of performance

and imagination. London: Sage.
Adams, M., & Coltrane, S. (2007). Framing divorce reform: Media, morality, and the
politics of family. Family Process, 46(1), 17-34.
Alberoni, F. (1972). The powerless elite: Theory and sociological research on the
phenomenon of stars. In D. McQuail (Ed.), Sociology of Mass Communications (pp.
75-98). Harmondsworth: Penguin Books.
Beyers, H. (2004). Interactivity and online newspapers: A case study on discussion
boards. Convergence, 10(4), 11-20.
Beyers, I. (2009, June 30). Beste lezers [editorial]. Dag Allemaal, 27, 3.
Billiet, J., Maddens, B., & Frognier A. P. (2006). Does Belgium (still) exist? Differences in
political culture between Flemings and Walloons. West European Politics, 29(5),
912-932.
Bourdieu, P. (1979). La distinction: critique sociale du jugement. Paris: Editions de
minuit.
Broos, D., & Van den Bulck, H. (2008). Als wij en zij elkaar ontmoeten via
televisieframes. In H. Van den Bulck & A. Dhoest (Eds.), Media/Cultuur/Identiteit
(pp. 71-87). Gent: Academia Press.
Brown, W. J., Basil, M. D., & Bocarnea, M. C. (2003). Social influence of an international
celebrity: Responses to the death of Princess Diana. Journal of Communication,
53(4), 587-605.
Butler Breese, E. (2010). Meaning, celebrity and the underage pregnancy of Jamie Lynn
Spears. Cultural Sociology, 4(3), 337-355.
Cashmore, E. (2006). Celebrity culture. New York, NY: Routledge.
Chen, Y. Y., Tsai, P. C., Chen, P. H., Fan, C. C., Hung, G. C. L., & Cheng, A. T. A. (2010). Effect
of media reporting of the suicide of a singer in Taiwan: the case of Ivy Li. Social
Psychiatry and Psychiatric Epidemiology, 45, 363-369.
D. A., & L. B. (2009, August 4). Ouders Yasmine mogen hun kleindochter niet zien. Het
Laatste Nieuws. Retrieved from
http://www.hln.be/hln/nl/944/Celebrities/article/detail/956196/2009/08/04
/Ouders-Yasmine-mogen-hun-kleindochter-niet-zien.dhtml

107
DAngelo, P. (2002). News framing as a multi-paradigmatic research program : A
response to Entman. Journal of Communication, 52(4), 870-888.
De Bock, S. (2009a, June 27). Ze worstelen met geaardheid. De Standaard. Retrieved
from
http://www.standaard.be/artikel/detail.aspx?artikelid=512C206G&word=ze+w
orstelen+met+geaardheid
De Bock, S. (2009b, June 27). Yasmine was ons icoon. De Standaard. Retrieved from
http://www.standaard.be/artikel/detail.aspx?artikelid=512C206I&word=Yasmi
ne+was+ons+icoon
De Pourcq, E. (2009, June 25). Yasmine stapt uit het leven. Het Laatste Nieuws. Retrieved
from
http://www.hln.be/hln/nl/944/Celebrities/article/detail/904948/2009/06/25
/Yasmine-stapt-uit-het-leven.dhtml.
Deuze, M. (2005). Popular journalism and professional ideology: Tabloid reporters and
editors speak out. Media, Culture & Society, 27(6), 861-882.
Dow, B. J. (2001). Ellen, television and the politics of gay and lesbian visibility. Critical
Studies in Media Communication, 18, 123-140.
Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of
Communication, 43(4), 51-58.
Entman, R. M. (2007). Framing bias: Media in the distribution of power. Journal of
Communication, 57(1), 163-173.
Evans, J., & Hesmondhalgh, D. (2005). Understanding media: Inside celebrity.
Maidenhead: Open University Press.
Fejes, F., & Petrich, K. (1993). Invisibility, homophobia and heterosexism: Lesbians, gays
and the media. Critical Studies in Mass Communication, 10(4), 396-422.
Gamson, W. A., & Modigliani, A. (1989). Media discourse and public opinion on nuclear
power: A constructionist approach. The American Journal of Sociology, 95(1), 1-
37.
Giles, D. C. (2002). Parasocial interaction: A review of the literature and a model for
future research. Media Psychology, 4(3), 279-305.
Giles, D. C. (2003). Media psychology. London: Routledge.
Gross, L. P. (1993). Contested closets: The politics and ethics of outing. Columbia, MO:
University of Minnesota Press.

108
Hall, S. (1980). Encoding/decoding. In S. Hall, D. Hobson, D. Lowe & P. Willis (Eds.),
Culture, media, language: working papers in cultural studies (pp. 128-38). London:
Hutchinson.
Halman, L., Inglehart, R., Diez-Medrano, J., Lujix, R., Moreno, A., & Basnez, M. (2008).
Changing values and beliefs in 85 countries. Leiden: Brill.
Hoefkens, K. (2009, June 26). Zo was er maar n. De Standaard. Retrieved from
http://www.standaard.be/artikel/detail.aspx?artikelid=P72BVG36&word=Yasmi
ne
Holmes, S. (2005). Off-guard, unkempt, unready?: Deconstructing contemporary
celebrity in Heat magazine. Journal of Media and Cultural Studies, 19(1), 21-38.
Horton, D., & Wohl, R. (1956). Mass communication and parasocial interaction:
Observations on intimacy at a distance. Psychiatry, 19, 215-229.
Kerkhof, A., Arensman, E., Neelenman, J., & de Wilde, E.J. (2003). Sucide en
sucidepreventie in Nederland: Achtergrond document bij het Nationaal Actieplan
Sucidepreventie. Amsterdam: Ivonne van de Ven Stichting en Depressiestichting.
Kitch, C. L. (2005). Pages from the past: History and memory in American magazines.
Chapel Hill: University of North Carolina Press.
Lumby, C. (2006). Vanishing points. In P. D. Marshall (Ed.), The Celebrity Culture Reader
(pp. 530-546). London: Routledge.
Marshall, P. D. (1997). Celebrity and power: Fame in contemporary culture. Minneapolis:
University of Minnesota Press.
Martin, G. (1998). Media influence to suicide: The search for solutions. Archives of Suicide
Research, 4(1), 51-66.
Neuman, W. R., Just, M. R., & Crigler, A. N. (1992). Common knowledge: News and the
construction of political meaning. Chicago: The University of Chicago Press.
Pan, Z., & Kosicki, G. M. (1993). Framing analysis: An approach to news discourse.
Political Communication, 10(1), 55-75.
Reynders, A., Van Heeringen, C., De Maeseneer, J., & Van Audenhove, C. (2009,
September). Onderzoek naar verklarende factoren voor de verschillen in
sucidecijfers in Vlaanderen in vergelijking met Europese landen. Retrieved from
http://www.steunpuntwvg.be/swvg/_docs/Publicaties/200906_rapport09_Suici
de.pdf
Rojek, C. (2001). Celebrity. London: Reaktion Books.

109
Rothblum, E. D. (1990). Depression among lesbians: An invisible and unresearched
phenomenon. Journal of Gay & Lesbian Psychotherapy, 1(3), 67-87.
Russell, S. T., & Joyner, K. (2001). Adolescent sexual orientation and suicide risk:
Evidence from a national study. American Journal of Public Health, 91(8), 1276-
1281.
Sandvoss, C. (2005). Fans. Cambridge: Polity.
Schiappa, E., Gregg, P. B., & Hewes, D. E. (2006). Can one TV show make a difference?
Will & Grace and the parasocial contact hypothesis. Journal of Homosexuality,
51(4), 15-37.
SiV. (2011). Scheiding in Vlaanderen: Het rapport. Brussels: SVR.
Sonneck, G., Etzersdorfer, E., & Nagel-Kuess, S. (1994). Imitative suicide on the Viennese
subway. Social Science and Medicine, 38(3), 453-457.
Stack, S. (1987). Celebrities and suicide: a taxonomy and analysis, 1948-1983. American
Sociological Review, 52(3), 401-12.
Stack, S., & Kposowa, A. J. (2008). The association of suicide rates with individual-level
attitudes: A cross-national analysis. Social Science Quarterly, 89(1), 39-59.
Swirski, P. (2005). From lowbrow to nobrow. Montreal, QC: McGill-Queens University
Press.
Szasz, T. S. (2002). Fatal freedom: The ethics and politics of suicide. New York, NY:
Syracuse University Press.
Van den Bulck, H., & Panis, K. (2010). Michael as he is not remembered: Jacksons
forgotten celebrity activism. Celebrity Studies, 1(1), 242-244.
Van den Bulck, H., & Tambuyzer, S. (2008). De celebritysupermarkt. Berchem: EPO.
Van Gorp, B. (2006). Een constructivistische kijk op het concept framing. Tijdschrift voor
Communicatiewetenschap, 34(3), 246-256.
Van Poecke, L. (1996). Media culture and identity formation in the light of invisible
socialization: From modernity to postmodernity. Communications, 21(2), 183-
198.
Van Zoonen, L. (2007). Audience reactions to Hollywood politics. Media, Culture &
Society, 29(4), 531-547.
Werkgroep Verder (2007). Als journalist kan je levens redden: Aanbevelingen voor
verantwoorde berichtgeving over zelfdoding. Retrieved from

110
 http://www.iasp.info/pdf/task_forces/belgium_als_journalist_kan_je_levens_red
den.pdf
Wood, P. B., & Bartkowski, J. P. (2004). Attribution style and public policy attitudes
toward gay rights. Social Science Quarterly, 85(1), 58-74.
Zhou, Y., & Moy, P. (2007). Parsing framing processes: The interplay between online
public opinion and media coverage. Journal of Communication, 57(1), 97-98.

111
Appendix
1. Codebook
ARTICLES REACTIONS
V1. Number V1. Number
V2. Medium V2. Reaction to article number X
V3. Date V3. Medium
V4. Page number V4. Date
V5. Number of images V5. Page
V6. Title V6. Framing devices: Examples
V7. Number of reactions V7. Metaphors/stereotypes
V8. Number of voices V8. Word choice tone
V9. Who is talking? V9. What is the problem?
V10. Metaphors/stereotypes V10. Who is the problem?
V11. Word choice tone V11. Causes: Who?
V12. Examples V12. Causes: What?
V13. Description V13. Moral judgment
V14. What is the problem? V14. Solution
V15. Who is the problem?
V16. Causes: Who?
V17. Causes: What?
V18. Moral judgment
V19. Solution
V20. Who is talking? (2)
REPETITION OF V10-V19
V31. Who is talking? (3)
REPETITION OF V10-V19

112
113
 Chapter 3

The second case study in this PhD also combines the elements of text and reception in the
context of celebrity (news) and focuses on the theme of sex the ultimate real, behind-the-
scenes celebrity persona. This chapter provides more in-depth insight in the everyday celebrity
news coverage and accompanying audience reactions as analyzed in chapter 1 by limiting
itself to a segment of the data of chapter 1, namely the celebrity sex stories. The media coverage
and audience reactions to celebrity sex stories are subjected to a quantitative framing-based
analysis and a selection of the data covering the theme of adultery is subjected to a
qualitative framing analysis. This chapter thus bridges the domains of communication and
celebrity studies. The celebrity sex case study demonstrates that celebrity is not just a
commodity but fulfills important social roles in contemporary Western societies. Indeed,
celebrity sex stories evoke social and moral discussions among audience members, that are not
only guided by the media frames but also by audiences personal experiences, parasocial
relationships with the celebrities, and interaction with peers.

GUESS WHO TIGER IS HAVING SEX WITH NOW? CELEBRITY SEX AND THE FRAMING
OF THE MORAL HIGH GROUND.

Hilde Van Den Bulck and Nathalie Claessens

(Published in Celebrity Studies, 4(1),46-57)

For media and audiences fascinated with celebrities private lives, insight into their
sexual behavior is the ultimate pleasure as it deals with the most intimate details, the
real behind the celebrity construct. A celebritys sex life is not just food for gossip but
may facilitate the articulation of moral, ethical, and social views on sexual issues in
contemporary society. This is analysed by means of a framing analysis of celebrity sex
stories and audience reactions in a sample of three celebrity websites (Flemish HLN,
British Heat, and American People). Results reveal that media and readers focus on only
a few, highly scandalous stories of married heterosexual mens adultery. While media
in their framing of celebrity news refrain from explicit judgments, readers are eager to
openly criticize celebrities adulterous transgressions while taking the moral high
ground. Focusing on adultery, seven frames relating to different views in society were
found, presenting adultery in either sociological, biological, psychological, or religious
terms. Only the womens fault frame supports the sexual double standard, claiming
women are responsible for a mans adultery.

Keywords: Celebrity; sexual representations; framing analysis; online news; adultery

114
Introduction
Justin Timberlake cheats on Jessica Biel (HLN 2010d), Warren Beatty slept with over
13,000 women (HLN 2010a), Aniston and Butler have amazing sex (HLN 2010b),
Rape case dropped against David Copperfield (Breuer 2010). Every day, celebrities
sex lives capture the attention of media and audiences. Particularly since the growth of
online celebrity gossip websites, their sexual antics can be followed, discussed and
speculated about on a daily and sometimes even hourly basis. However, academic
interest so far has mainly focused on television or magazines (Ward 2003, Brown 2009)
or on - often spectacular - case studies (Butler Breese 2010). This contribution wishes to
fill a gap by analyzing the day-to-day coverage of and (online) reactions to celebrities
sex lives on three celebrity websites (HLN, Heat, and People). These websites were
selected for their popularity and the provisions for online audience reactions. The focus
on three different national settings allows a comparison between two large global (US,
UK) and one small local celebrity industry and different cultural settings with their own
sexual norms and values. By means of a study of general celebrity sex items and a
framing analysis of celebrities adultery, we wish to investigate two claims often made in
celebrity studies: the celebrity-as-commodity thesis (Marshall 1997, Rojek 2001,
Cashmore 2006), and the idea that celebrity provides a forum to discuss social and
moral issues (Dyer 1987, 1998, Marshall 2006, Butler Breese 2010, Gorin and Dubied
2011).

Celebrity, sex, and society

This contribution starts from the conception of celebrity as a construct, resulting from
negotiations between a person looking for (and sometimes fighting) fame, his/her
entourage, the media, and audiences (Turner 2004). Media are crucial to celebrities in
their communication with the outside world (Evans and Hesmondalgh 2005) and to
audiences for information about celebrities whom they never meet in person (Cashmore
2006). Despite the lack of personal contact, audience members develop parasocial
relationships (Horton and Wohl 1956), which can be described as the illusion of a long-
term friendship, analogous to social relationships and encompassing an emotional
connection, yet which are mediated and one-sided (Claessens and Van den Bulck 2011).
As a construct, a celebritys image results from the interplay between three
personae: the professional persona resulting from actions in public life, the official

115
private personae based on information about the celebritys private life, and the real
person behind the construction, off-guard, unkempt, unready (Holmes 2005: 21). In
celebrity culture and celebrity reporting, the private and the real are of central
importance. Attention to a public persons private life is considered as the main
indicator that (s)he has reached celebrity status. Dyer (1998: 45) further sees the focus
on the private in the medias denial of celebritys public performances as work. The
media construct narratives of stars living in a world without any material problems,
making only their private life and particularly relationships potentially problematic.
This focus on relationship problems in a further perfect world is ideal to attract media
and audiences interest. Moreover, while public performances can only attract attention
at irregular intervals a film release, a sport stars expensive transfer relationships
can be gossiped about on a daily basis, tying audiences to both celebrities and media
outlets (Cashmore 2006). It is thus not surprising that relationship issues form a large
part of celebrity culture.
If relationships are the staple diet of celebrity journalism, using a rhetoric of
authenticity to create a level of intimacy (Dyer 1987, Holmes 2005), then celebrity sex
is the ultimate pleasure. How much more intimate and private can media reporting get?
At its core, celebrity journalism wants to reveal the real celebrity behind the celebrity
faade and, surely, it is in the bedroom that this real celebrity is revealed? Following
wider media trends towards sensationalism and tabloidisation (Austin and Barker 2003,
Turner 2004), the combination of the search for the real celebrity and of sex as the
ultimate intimacy or truth (see Knee 2006), culminates in the heightened media search
for and audience interest in celebrity sex scandals (Lull and Hinerman 2000).

Framing celebrity sex

The aim of this article is to provide an empirical analysis of societal views on sex in
celebrity (online) news stories and in audience reactions to these stories, comparing the
celebrity websites HLN, Heat, and People.
This research is part of a wider content analysis (Van den Bulck and Claessens
2013a) of three websites devoted to celebrity news: HLNs showbizz page (Flanders, i.e.
the northern part of Belgium, www.hln.be/showbizz), Heat (UK, www.heatworld.com),
and People (US, http://www.people.com/). The overall sample consists of ten
constructed weeks from January to October 2010 (one week/month), and includes all

116
celebrity news and for reasons of feasibility the first one hundred audience reactions
to each article. Such reactions may not be representative of those of the general
audience as only a selection of people engage in online discussions, yet the online
reactions provide useful information on how people present themselves to others (van
Zoonen 2007).
From the overall sample of 1,289 articles and 16,000 audience reactions, all
articles covering sex and the accompanying audience reactions were selected, resulting
in 51 articles and 685 audience reactions. These are subjected to framing analysis, an
approach which, more than other methods, enables the study of both encoding (media
articles) and decoding (audience reactions) processes (see Hall, 1980). In addition,
framing provides direct insight into the essential elements of a message (problem, cause,
moral judgment, solution) and is, more than other methods like discourse analysis,
suited for large datasets. Finally, as frames relate to general ideas in society, framing
analysis enables us to uncover underlying cultural norms and values about sex.
Framing is a process whereby a frame determines which aspects of reality are
selected, rejected, emphasized, or modified in the production of a media text and, at the
same time, provides the audience with a context and suggested meaning (Van Gorp
2004: 16). Frames are present in a text through framing devices (word choice,
metaphors, symbols, and stereotypes (Pan and Kosicki 1993)) and reasoning devices
(problem definition, causal interpretation, moral evaluation, and treatment
recommendation (Entman 1993: 52)). Researchers have pointed to a list of framing
moderators which mediate audiences adoption of media frames, including interaction
with peers, personal experiences and cognitions, parasocial relationships, and a
successful fit between audience and media values, norms, and practices (Van den Bulck
and Claessens 2013b, Gamson and Modigliani 1989). The codebook, that is mainly based
on framings reasoning devices can be found in Appendix 1.

Whose sex life is discussed?

A variety of 59 celebrities is covered in the items on sex: 33 different celebrities are
found in the articles and 51 in the reactions. The top three most covered celebrities in
the sex sample articles consisted of US golfer Tiger Woods (19.6%), UK footballer Ashley
Cole (5.9%) and UK footballer David Beckham (5.9%). The reactions mostly deal with
Tiger Woods (20.1%), US reality TV host Jesse James (15.6%) and US actor David

117
Boreanaz (10.2%). Interestingly, these are all men who (allegedly) had an affair that
received public attention in 2010. Thus, while sex features in quite a large number of
items, media and audiences seem to focus mainly on just a few celebrity cases related to
sex which are explored exhaustively and almost all involve a sex scandal, confirming the
importance of (sensational) scandals in celebrity news.
Exploring further, 68.6% of the sex articles and 77.9% of the reactions cover male
celebrities. This differs from the overall celebrity sample where women dominate and
refutes Duran and Prusanks (1997) statement that sexuality is the domain of women.
The high percentage of male celebrities is related to the large amount of adultery
scandals in which the man is portrayed as the main actor. However, women are often
included in these items, as the mans sexual partner, mistress, wife, and so on. This
reflects the traditional sexual double standard which portrays men as sexually
aggressive the main actors and women as sexually passive the sexual objects or
even victims (Oliver and Hyde 1993, Duran and Prusank 1997, Gamson 2001a, Ward
2003). The articles focus most on single (43.1%) and then on married (37.3%)
celebrities, and to a lesser extent on celebrities in a relationship (17.6%). However,
audience reactions focus more on married (52.2%) than single (43.2%) celebrities and
very little on celebrities in a relationship (10.3%). The latter suggests that readers, even
more than the media, show an interest in scandal as most stories about married
celebrities cover adultery. It also implies that, for media and audience, (adulterous) sex
stories are more interesting (and scandalous) when covering married celebrities rather
than celebrities in non-marital relationships.
The celebrities in the articles and reactions are almost invariably heterosexual
(96.1% - 99.0%), rooting celebrities sexual escapades firmly within the heterosexual
paradigm that sex should only take place within a relationship and between a man and a
woman (Rissel et al. 2007). This result reflects the lack of homo- or bisexual sexual
relationships found in sexual magazine content (Ward 2003).

What sex topics are addressed?

Within the general theme of sex, several topics are discussed. Clearly dominant is
adultery, accounting for 39.2% of the articles and 71.1% of audience reactions. Adultery
comprises allegations of adultery, the adulterous act itself, and its consequences (wifes
reaction, public apology). Other topics include multiple sexual partners (e.g., Warren

118
Beatty sleeping with over 12,000 women), sex addiction (e.g., Tiger Woods in sex rehab),
sex abstinence (e.g., actor Matthew McConaughey after his wife gave birth), abuse (e.g.,
US magician David Copperfield rape case), and prostitution (e.g., UK TV
presenter/model Peaches Geldof being accused of prostitution).
Compared to sexual content on TV or in magazines (Duran and Prusank 1997,
Brown 2009), celebrity news appears less prudish and more scandalous. In a
comprehensive overview of the field, Ward (2003) found that although sexual content
on TV in Western societies has increased over the past decades, it remains mostly verbal
and precursory. Further, unlike celebrity news, TV focuses on sex in non-marital rather
than marital relationships (Ward 2003: 351-5). In magazines in Western societies,
sexual media content has multiplied in the last decades (e.g. for US magazines there was
an increase from a .17 ratio per page in 1950s to .88 in 1980s (Scott 1986)), is more
explicit and direct than TV content, and has become less conservative (Ward 2003).
Ward (2003: 355-8) found that magazines focus on the goal for women to be sexually
desirable and attract male attention and often describe relationships as the exclusive
domain of women. They tend to portray women as sexually passive and men as
aggressive, confirming the traditional sexual double standard, also by portraying women
as having little sexual desire and needing to be cautious and discrete, and men as in a
constant state of sexual desire and readiness (Ward 2003: 355-8). Our results suggest
that sexual content in celebrity news and the accompanying audience reactions is less
everyday or prudish than the sexual content on television or in magazines. This can be
explained by the desire to reveal the real celebrity behind the formal public and
private image, providing food for scandal that fits the growing dominance of news
values biased toward drama, entertainment, and spectacle (Austin and Barker 2003,
Turner 2004).

How are celebrities sex lives evaluated?

While the framing of the sexual topics includes implicit judgments (cf. infra), the articles
dealing with sexual topics are mostly without explicit judgment (66.7%) and, when
providing an explicit evaluation, are more negative (27.5%) than positive (5.9%). The
readers are far more judgmental (83.9%) in their discussion of celebrity sex scandals,
mostly negatively criticizing the celebrity (74.6%). The fact that readers are more
judgmental than the media can be explained by several factors. First, media want to (at

119
least appear to) adhere to journalistic standards by seemingly bringing factual
information rather than judgments, whereas readers are ultimately aimed at expressing
opinions. Second, whereas the media have to be careful in their articles to maintain good
working relations with celebrities and to avoid law suits, the (anonymous) readers have
no need to hold back in their reactions, because they do not depend on celebrities for
their income and cannot be sued.
The mostly negative character of the readers judgments can be explained by
situating it within the notions of gossip and the moral high ground. As Turner (2004: 24)
states: gossip [is] an important social process through which relationships, identity, and
social and cultural norms are debated, evaluated, modified and shared. Gossip is often
employed to crystallize and reinforce community values, thereby furthering the
coherence and unity of the social group (Wilson, 1974: 93). As online communities
rooted in American, British, or Flemish (Belgian) culture, celebrity news readers
reinforce their cultural values and coherence by being critical of celebrities sexual
antics that are outside of dominant sexual norms. For example, by criticizing celebrities
adultery, the readers confirm and even reinforce that it is an act which is considered
never justifiable according to 54% Belgian, 55% UK, and 70% US citizens (Halman et al.,
2008). Gossip further functions as a means to protect individual self-interest and
enhance ones moral status (Paine, 1976). This can be linked to Goffmans (1959)
impression management theory which states that an individual represents oneself in the
best possible way. In an online environment, people have greater control over their self-
presentation as they can act more reflexively and intentionally (Ellison et al., 2006), for
instance by taking the moral high ground in discussing celebrities sexual transgressions
to enhance their own status as morally good people.
Our analysis so far thus indicates that, at one level, celebrity sex news is mainly
about entertainment and providing food for gossip, which is similar to Hermes (1995)
findings on the pleasures of reading womens and gossip magazines. At the same time,
though, there are strong claims that it helps audiences make sense of the social world
(Gamson 2001b, Evans and Hesmondalgh 2005, Gorin and Dubied 2011). This confirms
Feaseys (2008: 693) findings that:

120
 Popular star coverage appeals to the reader precisely because it can be used to
engage in debates about fundamental moral issues, such as infidelity and the role
of violence in society, without passing judgments and making potentially
unpopular comments about friends, family and work colleagues.

Adultery: addiction, sin, or the ultimate sign of masculinity?

As adultery dominates both the media and (especially) audience attention to celebrity
sex and since the majority of the most-covered celebrities allegedly had adulterous
affairs, an in-depth framing analysis of articles and reactions about adultery was
conducted. To this end, 17 articles and 366 reactions covering adultery were subjected
to an inductive, qualitative framing analysis. This reveals seven frames, next to a few
purely factual account and some none of our business reports. The factual accounts
merely state that the adultery took place, whereas the none of our business reports
mention a cause or solution as well as the statement that one should not judge the
adultery: He doesn't owe anyone an apology except his family (reaction to Fleeman
2010b).
A first frame treats adultery as an addiction, which should be recognized as such
rather than judged and can be solved through treatment in rehab: Glad Tiger [Woods] is
now stepping up to the plate and is on his healing journey (reaction to Fleeman 2010b).
This frame echoes a growing debate on sex addiction in Western society. Although
controversial, there are claims that sexual addiction can be listed among other substance
addictions (alcohol, drugs...) (Carnes 1983), and 3 to 6% of the US population is believed
to be addicted to sex (Gold and Heffner 1998). The issue has surfaced in celebrity gossip
in the past with Hollywood actors Michael Douglas and David Duchovny framed as
(recovered) sex addicts.
A second frame describes adultery as the result of a flawed personality for which
only the cheater and his terrible personality are to blame: Ashley [Cole] is the kind of
bastard who would cheat and treat people like scum (Reaction to Heat 2010a). Links
between personality traits and adultery are often made in society and psychology
scholars (Schmitt 2004) relate infidelity to low agreeableness (i.e. no trust or empathy)
and low conscientiousness (i.e. disorganized and unreliable). The suggested solution in
this frame is to get a divorce or for the adulterer to change his personality and become a
better person, if at all possible (as psychological traits are considered stable rather than

121
changeable). This frame is similar to the addiction frame, as both link adultery to
psychological causes, but the frames differ in terms of judgment (neutral vs. negative)
and solution. While the addiction frame sees it as a disease that can be cured in rehab, a
flawed personality cannot be cured and only (rarely) changed.
The third frame describes adultery as a sin, which is caused by lust and cannot be
justified. This frame is often promoted by references to Christian religion, which
considers adultery as one of ten mortal sins: Tiger [] have you not read your Bible
where it says Thou shalt not commit adultery? (reaction to Fleeman 2010a). Although
confidence in churches is found to be decreasing and religion is important in the lives of
18% of Belgian, 13% of UK, and 57% of US residents, the frame illustrates that religion
still matters to people, confirmed by 2008 data indicating that 49% Belgian, 37% UK,
and 80% of the US residents get comfort and strength from religion (Halman et al.
2008). Further, 43% Belgian, 67% UK, and 90% US respondents believe in the concept
of sin (Halman et al. 2008). Given these figures, it is not surprising that most
references to this religion-inspired frame are posted to the American People website.
The solution presented in this frame is divorce, which is surprising as religion strongly
promotes marriage until death do us part. The promotion of divorce as a solution in
this religious frame may reflect the increasingly tolerant social attitude toward divorce
in contemporary (Flemish, UK, and US) society (Halman et al. 2008).
Fourth, adultery is also framed as a societal phenomenon caused by a loss of
(sexual) norms. This relates to Durkheims (1951[1897]) notion of anomie: a
breakdown of social norms regulating individual behavior and social interaction which
is a property, not of the individual, but of the social system (Zhao and Cao 2010: 1212).
In this frame, processes of (sexual) value liberalization are evaluated negatively and
related to popular claims about the moral decay of contemporary Western society, as
secularization leads traditional restrictions to lose their power (Petersen and
Donnenwerth 1997). This can be contrasted to the sin frame where secularization
processes are not yet dominant and religion remains an important bearer of norms and
values. The societal phenomenon frame variably assigns the loss of norms (and thus the
adultery) to all mankind or only to Hollywood (wo)men: WTF is wrong with people?
God, it's like a rash on the ass of Hollywood!! (reaction to Garcia 2010).

122
 The fifth frame describes adultery as part of human nature, either in the sense
that monogamy is not natural or that making mistakes is human. In both variations
adultery is considered the natural consequence of human nature, with no evaluative
judgment being passed. Fitting the first option, readers state: monogamy is not natural
anyway. The man is a hunter by nature and driven by the power to mate (reaction to
HLN 2010c). The proposed solution is not to get married as human beings are not
programmed to stay with one partner their whole lives. These statements can be related
to socio-biology (Oliver and Hyde 1993) and evolutionary psychology (Daly and Wilson
2000) which state that men are not naturally fit for a monogamous life but are
programmed to spread their genes and have as many sexual partners as possible. In the
other variation, readers echo the popular saying nobody is perfect and state: I think
he's just human. We all make mistakes, some worse than others (reaction to Garcia
2010). The suggested solution here is forgiveness as all humans are flawed and thus
destined by nature to make mistakes.
The sixth frame, the ultimate sign of masculinity, is the only frame in our sample
that frames adultery positively. Here, the adultery is implicitly caused by the nature of
men which is predominantly driven by lust and stimulates them to have affairs. Like the
human nature frame, this is linked to ideas rooted in socio-biology or evolutionary
psychology. However, this frame passes a positive judgment: (real) men should fulfill
their task of passing on their genes: Tiger=The Man. Anyone saying he would react
differently in his position, denies himself (reaction to HLN 2010c). As adultery is not
considered problematic, no solution is provided, except the continuation of the observed
behavior.
A final frame covers adultery as the womens fault. Here, the women as wife
and/or mistress are responsible for the mans adulterous behavior. The man is
evaluated as a victim of circumstances beyond his control, while the women are judged
negatively and it is suggested that they should change. In some cases blame is cast onto
the wife, accused of being absent too often or not providing her man with enough
marital sex: Obviously, Tiger didnt have enough sex at home (reaction to HLN 2010c).
At other times, the mistress is seen to have seduced the (seemingly powerless) man:
Whats really disgusting though are the women who throw themselves at stars when
they know they are married (reaction to Garcia 2010). This frame reaffirms sexual
double standards by confirming that (sexual) relationships are the domain and

123
responsibility of women and that men can be manipulated into having sex because of
their natural appetite (Duran and Prusank 1997). The contrast between the sexually
failing wife and the seductive mistress reflects the virgin-whore dichotomy that is often
used as a narrative guide in sex scandals, describing women either as young, naive
victims (virgin) or publicity seeking seductresses (whore) (Gamson 2001a: 164). While
the image of the mistress as a seductress fits this frame perfectly, the virgin wife image
is mitigated in this frame to the extent that she is not described as a victim but seen
negatively, as the cause for her husbands escapades.

Adultery frames: media vs. audience

The different frames are not spread evenly across the sample. In media articles, only the
neutral, addiction, flawed personality, and womens fault frames are used and the latter
two dominate. The three websites differ to some extent as People is the least judgmental
(using the neutral and addiction frame) and Heat the most evaluative (employing the
flawed personality and womens fault frames), with HLN coverage of adultery in
between (framing it as the womens fault, addiction, or the result of a flawed
personality). Audiences do not always follow the medias lead but can be seen to provide
counter-frames in negotiated or oppositional readings of the media frames (Hall 1980).
As such, audience reactions present the whole variety of frames and the flawed
personality and the womens fault frame dominate. The addiction and the sin frame
hardly ever occur, except with some People readers. Interestingly, only the HLN readers
employ the positive masculinity frame. HLN readers further promote the neutral, the
womens fault, and the societal phenomenon frame, hereby combining judgment-free
and negative approaches. Heat readers are the most critical, either toward the adulterer
or the women, as they mostly employ the flawed personality and the womens fault
frame. The human nature and the none of our business frame are mostly used by the
People readers who alternate between frames without explicit judgment and the
negative flawed personality frame.
The readers are thus found to either adopt the frame presented in the media
article or to develop a (counter-)frame. Analyzing further, it appears audience reactions
result from framing moderators, including personal experiences, interaction with peers,
and parasocial relationships. When readers refer to personal experiences (e.g., being
cheated on), they mostly frame adultery as a sin or as the result of a flawed personality:

124
They're always open and honest when they're caught - and not before. Well, except for
my ex. He wasn't even open and honest after he was caught. LOL (reaction to Garcia
2010).
In interactions with peers, readers are often found to defend or strengthen their
opinion, as can be seen in the following discussion among HLN readers on the sexual
double standard in response to singer and actor Justin Timberlakes (alleged) betrayal:

What is wrong with women nowadays? You hardly know the person and sleep
with him. Just for a one-night-stand. Wow. How do they feel afterwards? That is
what I always wonder? Used? Dirty? Id rather have sex out of love than out of
lust.

What do you mean with women nowadays? Who do you think she had a one-
night-stand with? This is another perfect example of a double standard. The
woman should feel used and dirty and have sex out of love and about the man
not a word. If she wants to have sex with someone she barely knows, it is her
right. And what to say oft Justin Timberlake? He is cheating on his girlfriend and
lying to get a woman in his bed. But, of course, that is different, hes a man.
(reactions to HLN 2010d)

Finally, parasocial relationships with celebrities can encourage the construction

of counter-frames. Some readers maintain a negative parasocial relationship with the
cheater or mistress and frame the adultery as a sin, resulting from the frames relating to
a flawed personality, or the womens fault: Wayne [Rooney], u need your fat hairy balls
cut off for what youve put the MOTHER OF YOUR CHILD through! (reaction to Heat
2010b). These reactions are often very sexualized and aggressive, suggesting strong
negative feelings toward both the topic of cheating and the celebrity. Others maintain a
positive relationship with the cheater and frame it neutrally as an addiction or part of
human nature, or defend it as the womens fault or a sign of masculinity: To David and
Jaime, you have all of my support and don't worry what others think, it's your life and
marriage and no matter what, we all make mistakes (reaction to Garcia 2010). In the
frame of adultery as a sign of masculinity, the reactions are gender-based as it is mainly
articulated by male readers: GO TIGER !!! FERTILISE MORE!!! BE A REAL MAN
(reaction to Fleeman 2010a), or I know what a lot of men think while reading this
article, never mind guys, it is all natural, we are born and will die this way, women try to
change us, but they wont succeed (reaction to HLN 2010c). We should note that this
finding is derived from the readers online or (in the case of People) Facebook names,

125
which are not necessarily truthful representations of their real-life identities, but still
are valuable as (online) self-presentations. These findings confirms Oliver and Hydes
(1993) meta-analysis of gender differences and sexual attitudes/behavior that shows
men hold more permissive attitudes toward extramarital sex. Finally, several readers
refer to their positive parasocial relationship with the victim (the wife) and employ a
negative frame to describe the adultery: Colleen, just LEAVE HIM!!!!!! [] Believe it or
not, YOU DESERVE MORE (reaction to Heat 2010b).

Celebrity news and societal views on adultery

The in-depth framing analysis of adultery reveals several frames relating to different
viewpoints in contemporary Western society, confirming Dyers (1987) claim that (news
about) celebrities relate(s) to general societal ideas. Adultery is either framed in
sociological (societal phenomenon), psychological (addiction, flawed personality),
biological (human nature, sign of masculinity), or religious (sin) terms. The media only
use approaches without explicit judgment and psychological frames, whereas audiences
employs a variety of frames to make meaning of celebrities adultery.
Interestingly, most of these frames stick to conservative rather than liberal sexual
norms and values. With the exception of the ultimate sign of masculinity, all frames
implicitly or explicitly support monogamous marriages as a (sacred) institution. This
seems reflective of societys dominant attitudes toward marriage, sexuality, and adultery
(Halman et al. 2008) and of the heterosexual paradigm (Rissel et al. 2007). While figures
indicate that one-third of Belgian, UK and US residents believe people should be able to
enjoy sexual freedom without restrictions, this more liberal ideology is absent from our
sample and only surfaces indirectly as an inevitable consequence of biological
characteristics rather than social norms and values. With this exception, frames do not
break with or reshape existing norms and values concerning sex and marriage. Adultery
scandals are considered a transgression and evoke very emotional and sometimes
coarse comments. At the same time, one of the most conservative stereotypes, the sexual
double standard (Duran and Prusank 1997, Ward 2003), was only supported in one
frame the womens fault in which women (wife/mistress) are considered responsible
for a mans sexual behavior including adultery (Duran and Prusanks 1997) and in which
aspects of the virgin-whore dichotomy are found (Gamson 2001a, 2001b).

126
Conclusion
This study shows that celebrity sex news is not only about entertainment and providing
food for gossip, as it also helps audiences make sense of the social world. One dominant
thesis interprets this meaning creation within the notion of celebrity-as-commodity,
there to embody capitalist society as both objects and vehicles of consumption,
encouraging and validating consumer culture (Cashmore, 2006: 269). Celebrity sex
stories not only contribute to the media and entertainment industry but help keep the
capitalist system alive by trying to divert audiences attention from real problems to
the trivialities of celebrities private lives. Audiences willingness to engage with these
stories and celebrities is seen as compensating for anonymity, alienation, and
frustrations in capitalist society. In our findings, the readers almost exclusively focus on
the scandalous adultery stories. This seems to support - at least partly - the hegemonic
capitalist and celebrity-as-commodity claim that celebrities sex lives are used as
consumer products to attract sensation-loving audiences and distract them from
potential criticism of the celebrity and hyper-consumerist systems itself. Our results
suggest that most audience members are eager to participate and comment on the
scandalous lives of the famous elite.
However, discarding the results as a mere case of capitalist hegemony, provides
too narrow a view on the role of media and celebrity in processes of meaning-making
and understanding of the social world. People may read and engage with such celebrity
gossip and at the same time are aware of and concerned about wider political and
societal issues. What is more, most authors maintain that celebrity coverage - even its
most trivial topics - can provide media and society with a forum to discuss social,
cultural, and moral issues, values, and norms, also regarding acceptable sexual behavior
(Dyer 1987, 1998, Marshall 2006, Austin and Barker 2003, Feasey 2008, Butler Breese
2010, Gorin and Dubied 2011). Indeed, dominant societal attitudes toward sex both
liberal views following the sexual revolution of the 1960s and the remaining
importance of the sexual double standard and the heterosexual paradigm are present
in our findings for celebrity sex news. In the case of adultery, seven frames are found
that relate to a variety of societal viewpoints (religious, sociological, psychological,
biological). Media coverage of celebrities sexual antics not only represents societys
norms and values but also provides readers with food for thought. Those readers do not
simply adopt the media stories but actively discuss sexual issues (mainly adultery) and

127
make strong evaluations (mostly negative) of the sexual actions of celebrities. In doing
so, they discuss different views, social norms and values regarding sex in society.
However, it should be noted that these discussions mostly reproduce normative values
around morality and sexual conduct and only rarely substantively criticize hegemonic
ideas of contemporary society.

128
References
Austin, T., & Barker, M. (2003). Contemporary Hollywood stardom. London: Arnold.
Breuer, H. (2010). Rape case dropped against David Copperfield. People [online], 12
January. Available from:
http://www.people.com/people/article/0,,20336452,00.html.
Brown, J. D. (2009). Media and sexuality. In R.L. Nabi, & M.B. Oliver (Eds.), The SAGE
Handbook of Media Processes and Effects (pp. 409-422). London: Sage.
Butler Breese, E. (2010). Meaning, celebrity and the underage pregnancy of Jamie Lynn
Spears. Cultural Sociology, 4, 337-355.
Carnes, P. (1983). Out of the shadows: Understanding sexual addiction. Minneapolis, MN:
CompCare Publishers.
Cashmore, E. (2006). Celebrity Culture. New York: Routledge.
Claessens, N., & Van den Bulck, H. (2011). Personal Bonding with Celebrities Through
Media Reception: The PSR Scale. Paper presented at the 7th Media Psychology
Conference. Bremen, 10 August.
Daly, M., & Wilson, M. I. (2000). The evolutionary psychology of marriage and divorce. In
L. J. Waite, & C. Bacharach (Eds.), The ties that bind: Perspectives on marriage and
cohabitation (pp. 91-110). New Jersey: Transaction Publishers.
Duran, R. L., & Prusank, D. T. (1997). Relational themes in mens and womens popular
nonfiction magazine articles. Journal of Social and Personal Relationships, 14(2),
165-189.
Durkheim, E. (1951 [1897]). Suicide: A study in sociology. New York: The Free Press.
Dyer, R. (1987). Heavenly bodies: Film stars and society. London: MacMillan.
Dyer, R. (1998). Stars. London: British Film Institute.
Ellison, N., et al. (2006). Managing impressions online : Self-presentation processes in
the online dating environment. Journal of Computer-Mediated Communication, 11,
415-441.
Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of
Communication, 43(4), 51-58.
Evans, J., & Hesmondhalgh, D. (2005). Understanding media: Inside celebrity. Maidenheid:
Open University Press.
Feasey, R. (2008). Reading heat: The meanings and pleasures of star fashions and
celebrity gossip. Continuum, 22(5), 687-699.

129
Fleeman, M. (2010a). Tiger Woods to apologize, discuss his past and future. People
[online], 17 February. Available from:
http://www.people.com/people/article/0,,20345058,00.html.
Fleeman, M. (2010b). PHOTO: Tiger Woods surfaces in Florida!. People [online], 17
February. Available from:
http://www.people.com/people/article/0,,20345137,00.html.
Gamson, J. (2001a). Jessica Hahn, media whore: sex scandals and female publicity.
Critical Studies in Media Communication, 18(2), 157-173.
Gamson, J. (2001b). Normal sins: Sex scandal narratives as institutional morality tales.
Social Problems, 48(2), 185-205.
Gamson, W. A., & Modigliani, A. (1989). Media discourse and public opinion on nuclear
power: A constructionist approach. The American Journal of Sociology, 95, 1-37.
Garcia, J. (2010). David Boreanaz admits: I cheated on my wife. People [online], 3 May.
Available from: http://www.people.com/people/article/0,,20366077,00.html.
Goffman, E. (1959). The presentation of self in everyday life. New York: Anchor.
Gold, S. N., & Heffner, C. L. (1998). Sexual addiction: Many conceptions, minimal data.
Clinical Psychology Review, 18(3), 367-381.
Gorin, V., & Dubied, A. (2011). Desirable people: Identifying social values through
celebrity news. Media, Culture and Society, 33(4), 599-618.
Hall, S. (1980). Encoding/decoding. In S. Hall, D. Hobson, D. Lowe, & P. Willis (Eds.),
Culture, media, language: working papers in cultural studies (pp. 128-138).
London: Hutchinson.
Halman, L., et al. (2008). Changing values and beliefs in 85 countries. Leiden: Brill.
Heat (2010a). The 9 OClock News Thu 25 Feb: Ashley Cole in another world since
separation. Heat [online], 25 Feb. Available from:
http://www.heatworld.com/Celeb-News/2010/02/The-9-0Clock-News--Thu-
25-Feb/.
Heat (2010b). Wayne Rooneys own family think hes an utter disgrace. Heat [online], 6
Sept. Available from: http://www.heatworld.com/Celeb-News/2010/09/Wayne-
Rooneys-own-family-think-hes-an-utter-disgrace/.
Hermes, J. (1995). Reading womens magazines: an analysis of everyday media use.
Cambridge: Polity Press.

130
HLN (2010a). Warren Beatty sliep met bijna 13.000 vrouwen. HLN [online], 4 January.
Available from:
http://www.hln.be/hln/nl/944/Celebrities/article/detail/1049218/2010/01/0
4/Warren-Beatty-sliep-met-bijna-13-000-vrouwen.dhtml.
HLN (2010b). Aniston en Butler hebben fantastische seks. HLN [online], 17 February.
Available from:
http://www.hln.be/hln/nl/944/Celebrities/article/detail/1068563/2010/02/1
7/Aniston-en-Butler-hebben-fantastische-seks.dhtml.
HLN (2010c). Tiger bekent 120 affaires, buurmeisje is er te veel aan. HLN [online], 29
April. Available from: http://www.hln.be/hln/nl/923/Kanaal-
Sport/article/detail/1099453/2010/04/29/Tiger-bekent-120-affaires-
buurmeisje-is-er-te-veel-aan.dhtml.
HLN (2010d). Justin Timberlake bedriegt Jessica Biel. HLN [online], 28 Oct. Available
from:
http://www.hln.be/hln/nl/944/Celebrities/article/detail/1175718/2010/10/2
8/Justin-Timberlake-bedriegt-Jessica-Biel.dhtml.
Holmes, S. (2005). Off-guard, unkempt, unready?: Deconstructing contemporary
celebrity in Heat magazine. Journal of Media & Cultural Studies, 19(1), 21-38.
Horton, D., & Wohl, R. (1956). Mass communication and parasocial interaction:
Observations on intimacy at a distance. Psychiatry, 19, 215-229.
Knee, A. (2006). Celebrity skins: the illicit textuality of the celebrity nude magazine. In S.
Holmes, & S. Redmond (Eds.), Framing celebrity: new directions in celebrity
culture (pp. 161-176). London: Routledge.
Lull, J., & Hinerman, S. (2000). Media scandals. New York: Columbia University Press.
Marshall, P. D. (1997). Celebrity and power: Fame in contemporary culture. Minneapolis:
University of Minnesota Press.
Marshall, P. D. (2006). The celebrity culture reader. London: Routledge.
Oliver, M. B., & Hyde, J. S. (1993). Gender differences in sexuality: A meta-analysis.
Psychological Bulletin, 114(1), 29-51.
Paine, R. (1967). What is gossip about? An alternative hypothesis. Man, 2(2), 278-85.
Pan, Z., & Kosicki, G. M. (1993). Framing analysis: An approach to news discourse.
Political Communication, 10(1), 55-75.

131
Petersen, L. R., & Donnenwerth, G. V. (1997). Secularization and the influence of religion
on beliefs about premarital sex. Social Forces, 75(3), 1071-1089.
Rissel, C. E., et al. (2007). Attitudes towards sex in a representative sample of adults.
Australian and New Zealand Journal of Public Health, 27(2), 118-123.
Rojek, C. (2001). Celebrity. London: Reaktion Books.
Schmitt, D. (2004). The big five related to risky sexual behavior across 10 world regions:
Differential personality associations of sexual promiscuity and relationship
infidelity. European Journal of Personality, 18(4), 301-319.
Scott, J. E. (1986). An updated longitudinal content analysis of sex references in mass
circulation magazines. Journal of Sex Research, 22(3), 385-392.
Turner, G. (2004). Understanding celebrity. London: Sage.
Van den Bulck, H., & Claessens, N. (2013a). Of Local and Global Fame: A Comparative
Analysis of News Items and Audience Reactions on Celebrity News Websites
People, Heat, and HLN. Journalism [online]. Available at:
http://jou.sagepub.com/content/early/2013/05/28/1464884913488725
Van den Bulck, H., & Claessens, N. (2013b). Celebrity Suicide and the Search for the
Moral High Ground: Comparing Frames in Media and Audience Discussions of the
Death of a Flemish Celebrity. Critical Studies in Media Communication, 30(1), 69-
84.
Van Gorp, B. (2004). Frames in de Nieuwsmedia (Doctoral dissertation). Antwerp:
University of Antwerp.
van Zoonen, L. (2007). Audience reactions to hollywood politics. Media, Culture &
Society, 29, 531-547.
Ward, M. (2003). Understanding the role of entertainment media in the sexual
socialization of American youth: a review of empirical research. Developmental
Review, 23, 347-388.
Wilson, P. J. (1974). Filcher of good names: an enquiry into anthropology and gossip.
Man, 9(1), 93-102.
Zhao, R., & Cao, L. (2010). Social change and anomie: A cross-national study. Social
Forces, 88(3), 1209-1229.

132
Appendix
1. Codebook
ARTICLES
V1. Number
V2. Medium
V3. Date
V4. Title
V5. Number of reactions
V6. Which celebrity is discussed?
V7. Celebrity: Gender
V8. Celebrity: Nationality
V9. Celebrity: Relationship status
V10. Celebrity: Ethnicity
V11. Celebrity: Sexual orientation
V12. Celebrity: Domain
V13. Who is talking?
V14. Framing devices
V15. Problem (who, what)
V16. Cause (who, what)
V17. Moral judgment
V18. Solution.
V19. Who is talking? (2)
REPETITION OF V14-V19
V20. Who is talking (3)
REPETITION OF V14-V19
REACTIONS
V1. Number
V2. Reaction to article number X
V3. Medium
V4. Date
V5. Name reader
V6. Gender reader
V7. Which celebrity is discussed?
V8. Celebrity: Gender
V9. Celebrity: Nationality
V10. Celebrity: Relationship status
V11. Celebrity: Ethnicity
V12. Celebrity: Sexual orientation
V13. Celebrity: Domain
V14. Framing devices
V15. Problem (who, what)
V16. Cause (who, what)
V17. Moral judgment
V18. Solution.
V19. Parasocial interaction
V20. Interaction with peers
V21. Personal experiences
133
 Chapter 4

Based on the previous chapters findings that parasocial relationships (PSR) with celebrities are
referred to by a segment of the audience and have the potential to mediate the adoption of
media frames, this concept is further explored here. Indeed, chapter 4 focuses on celebrity
audiences and presents a large-scale quantitative online survey study that examines the
selection of favorite celebrities and the PSR with them. This study wishes to provide more
systematic knowledge of the concept by addressing several gaps in the current body of PSR
research (focus on TV, confusion between PSI and PSR, student samples, neglect of socio-
demographic variables). This chapter examines the PSR concept in the context of celebrity
studies, hereby bridging psychology and celebrity studies. The consideration of the socio-
demographic variables can be considered a more cultural studies approach to PSR. This chapter
thus combines different domains while respecting their specificities and the accompanying
methodologies. Indeed, the rigorous statistical procedures that are prescribed in this type of
survey research and the development of scales are followed meticulously in this study.

PARASOCIAL RELATIONSHIPS WITH AUDIENCES FAVORITE CELEBRITIES: THE

ROLE OF AUDIENCE AND CELEBRITY CHARACTERISTICS IN A REPRESENTATIVE
FLEMISH SAMPLE

Nathalie Claessens and Hilde Van den Bulck

(Submitted to Communications: The European Journal of Communication Research)

This article provides insight into one form of audience involvement with celebrities:
parasocial relationships (PSR). To address several shortcomings in PSR research focus
on TV, confusion between PSI and PSR, use of student samples, neglect of socio-
demographic variables a representative online survey is conducted with 1000 Flemish
adults who indicate 382 celebrities as favorites. The construction of a new PSR scale
reveals that the concept contains two important elements: emotional connections and
analogy with social relationships. Confirming previous research, most favorite celebrities
are male and cultural proximity is especially important for the celebrity preferences of
older and lower-educated respondents. In one combined model, respondents and
celebrities (socio-demographic) characteristics are included as potential PSR predictors.
This model nuances previous research and reveals that people who are male, older than
50, lower-educated, and interested in celebrity news have stronger PSR. Further,
stronger PSR are found for local and religious and political celebrities.

Keywords: parasocial relationships, favorite celebrity, analogy to social relationships,

emotional connection, cultural proximity

134
Contemporary Western societies are characterized by the omnipresence of celebrities,
which is widely acknowledged in celebrity studies (Marshall, 2006; Rojek, 2001; Turner,
2004). Audience research shows that celebrities not just provide food for entertainment
or gossip, but fulfill various social roles, among which social companionship. Audience
members feel involved with celebrities, an involvement ranging from fandom (Sandvoss,
2005) and identification (Cohen, 2009) to parasocial interaction (PSI) or parasocial
relationships (PSR) (Giles, 2002; Horton & Wohl, 1956). The latter PSI and PSR are
the most widespread and influential forms of audience-celebrity involvement (Giles,
2002; Klimmt, Hartmann, & Schramm, 2006; Tian & Hoffner, 2010) and can have
considerable impact on audiences information processing, attitudes, and even behavior
(Boon & Lomore, 2001; Schiappa, Gregg, & Hewes, 2005). Despite the considerable
amount of PSI/PSR studies, the research still shows important gaps or shortcomings:
first, television celebrities are often the only type considered whereas contemporary
celebrity culture includes a large variety of celebrity domains; second, although
distinctive concepts, PSI and PSR are often used interchangeably; third, most studies use
college student samples; and, fourth, socio-demographic variables are rarely properly
addressed in PSI/PSR studies.
This study wishes to contribute, first, by looking at a wide range of celebrities,
resulting from asking respondents in a survey who their favorite celebrities are without
any domain-related limitations. This allows for an analysis of favored celebrities in
terms of gender, domain, and nationality. In addition, the celebrity preferences will be
analyzed with regard to respondents gender, age, and education, addressing the fourth
shortcoming described above. Second, the study wishes to understand the nature of
audience members long-term PSR rather than short-term PSI and the way it relates
to both audience members gender, age, education, and celebrity news interest and to
celebrities gender, domain, and nationality for its strength. To this end, an online
questionnaire is completed by a representative sample of the Flemish (the northern part
of Belgium) adult population (N = 1000)5, hereby addressing the limitation of college
student samples. The data from these surveys help to investigate the respondents
favorite celebrities as well as the strength and nature of PSR. In addition, the celebritys

55
This study is part of a wider project funded by an internal grant of the University of Antwerp
(BOF-NOI 2008-2011) and a grant of the Flemish Research Council (FWO 2008-2011), both
obtained competitively after peer review

135
and respondents characteristics are combined in one model predicting the strength of
PSR.
Further, the aim of this study is to meet the need for a commonly used, theory-
driven, empirically-tested, coherent measure of PSR (Hartmann, Stuke, & Daschmann,
2008: 27). While there are scales (Auter & Palmgreen, 2000; Rubin & Perse, 1987) to
measure PSI asymmetrical interactions that take place as situational processes of
character perception and elaboration during media exposure (Hartmann et al., 2008:
25) these are not ideal for PSR one-sided interpersonal relationships that television
audiences establish with media characters (Hartmann et al., 2008: 25). Additionally,
there are several scales to measure PSR, but these are limited to a specific segment of
celebrities: TV personalities (Bocarnea & Brown, 2007) and sportsmen (Hartmann et al.,
2008). This contribution will therefore compose a new scale fit to measure PSR with any
type of celebrity by building on and combining items of the existing scales. By
investigating the general audiences favorite celebrities and PSR, this article wishes to
contribute to the knowledge on the relevance of celebrity culture in audiences everyday
lives.

Celebrities and Audiences

Following Turners definition of celebritization (2004: 8; see Evans & Hesmondalgh,
2005), public figures become celebrities when media interest in their activities is
transferred from reporting on their public role [] to investigating the details of their
private lives. In this article, celebrities are similarly defined as famous people whose
private lives are at least as important to media and audiences as their professional
activities. Celebrity is further conceptualized as constructed within the celebrity
apparatus, consisting of the famous person, the entourage, the media, and audiences.
Within the celebrity apparatus, there are multiple relationships between these actors.
For this study, the famous personaudience axis is crucial as celebrities need audiences
to achieve/maintain their status. In turn, celebrities fulfill social roles for audiences (Van
den Bulck & Claessens, 2013a, 2013b; Dyer, 1998; Marshall, 2006; Rojek, 2001; Turner,
2004). First, celebrities provide entertainment, relaxation, and a pleasant pastime.
Second, celebrities promote social interaction by providing food for gossip, in which the
celebrity functions as a shared, but distant, acquaintance. Further, celebrities help
people make sense of the social world and their own identities, by representing what it

136
is to be human (Dyer, 1998: 99; see Van den Bulck & Claessens, 2013a, 2013b). Finally,
audience members can develop personal connections with celebrities (PSR) in which
celebrities are social companions.

Parasocial Interaction versus Parasocial Relationships

One form of involvement between audiences and celebrities is that of PSI or PSR. The
term PSI, developed by Horton and Wohl (1956), describes the illusion of a face-to-face
friendship between an audience member and a celebrity (see Cohen, 2009; Giles, 2002,
2003). PSI is decidedly one-sided, as a large group of people, the audience, knows a lot
about a celebrity who hardly knows anything about them (Giles, 2002). This parasocial
contact almost exclusively takes place through media as people rarely encounter
celebrities in real life (Giles, 2002).
Although there is a vast body of PSI/PSR studies, there are still several
shortcomings. First, most PSI/PSR research focuses on television characters or
personalities (e.g., Cohen, 1997; Eyal & Dailey, 2012; Eyal & Rubin, 2003; Lather &
Moyer-Guse, 2011; Levy, 1979; Rubin & McHugh, 1987; Rubin & Perse, 1987; Schramm
& Hartmann, 2008; Tian & Hoffner, 2010; Turner, 1993; Wang, Fink, & Cai, 2008).
However, current celebrity culture comprises a wide range of celebrity domains (film,
television, music, sports, fashion, politics, religion, business, and so on). PSI/PSR
research would thus benefit from including this variety which is why this study includes
non-television celebrities, following the example of other studies (Bocarnea & Brown,
2007; Cohen, 2010).
Second, the concepts of PSI and PSR are often used interchangeably, while they
actually point to two distinct processes (Klimmt et al., 2006; Schramm & Hartmann,
2008). Whereas PSI refers to short-term encounters in which audience members have
the illusion of interaction with celebrities, PSR imply long-term connections between
audience members and their favorite celebrities, that are developed over time and
through a variety of mediated encounters or PSI (Bocarnea & Brown, 2007; Giles, 2003;
Hartmann et al., 2008; Klimmt et al., 2006). Indeed, audience members come in contact
with celebrities through a variety of celebrities performances, popular and quality
media outlets, and in social conversations. The conflation of these two concepts is
problematic as several scales measuring PSI or PSR include items tapping into both
processes, rather than focusing on one (Auter & Palmgreen, 2000; Rubin & Perse, 1987).

137
This is also noted by Schramm and Hartmann (2008) who, in response, have developed
the PSI-Process scales to clarify the distinctions between PSI and PSR. For PSR, several
scales have been introduced (Bocarnea & Brown, 207; Hartmann et al., 2008), but these
are limited to one type of celebrity. Therefore, this study wishes to develop a new scale,
building on the existing scales, that focuses on long-term PSR and applies to a wide
range of celebrities.
This study further investigates the nature of PSR by distinguishing two main
elements described in previous research. First, emotional connections are developed in
PSR that encompass feelings of empathy, intimacy, and comprehension (Boon & Lomore,
2001; Horton & Wohl, 1956; Klimmt et al., 2006; Rubin & Step, 2000). Second, PSR are
similar to social relationships (Cohen, 2010; Eyal & Dailey, 2012; Giles, 2003; Lather &
Moyer-Guse, 2011; Rubin & Perse, 1987; Schramm & Hartmann, 2008; Turner, 1993;
Wang et al., 2008) in terms of interpersonal expectations and processes of relationship
development (Rubin & McHugh, 1987; Turner, 1993) and break-ups (Cohen, 2003), and
knowledge gathering and storage (Schiappa et al., 2005). Indeed, celebrities are part of
peoples social worlds and the processes of interaction with them are similar to those
with friends or family members (Boon & Lomore, 2001). Therefore, this study
hypothesizes that:

H1: PSR consist of two main elements:

1) an emotional connection
2) an analogy with social relationships

Third, most PSR/PSI studies employ college student samples (e.g., Auter &
Palmgreen, 2000; Cohen, 1997; Cohen, 2010; Eyal & Dailey, 2012; Eyal & Rubin, 2003;
Lather & Moyer-Guse, 2011; Rubin & McHugh, 1987; Rubin & Perse, 1987; Schiappa et
al., 2005; Turner, 1993; Wang et al., 2008). Most scholars acknowledge this limitation
but only a few studies actually address this by obtaining a more representative sample
(e.g., Cohen, 2003; Levy, 1979; Rubin & Step, 2000; Schramm & Hartmann, 2008). The
current academic knowledge on PSI and PSR should thus be contextualized within the
college student population, a very specific, and in many ways a-typical group in society.
This study wishes to address this by working with a large (N = 1000) representative
sample of the adult population in Flanders.

138
 Finally, the relevance of audiences socio-demographic variables age, gender,
education has not been addressed properly with regard to celebrity preferences and
PSI/PSR. The variables of gender, age, and education have hardly been addressed
separately and to our knowledge never combined within one model, which is where
this study wishes to contribute. With regard to the gender of favorite celebrities, Boon
and Lomore (2001) and Cohen (1997, 2003) found that both men and women prefer
male celebrities. The hypothesis is thus as follows:

H2: Male celebrities are selected more often as favorite celebrities

Celebrity preferences can further be related to age, especially in terms of cultural

proximity (Tian & Hoffner, 2010), which can be defined as the tendency to prefer media
products from one's own culture or the most similar possible culture (Straubhaar,
2003: 85). This concept is relevant in a celebrity culture where audiences are
increasingly faced with both global and local celebrities as a result of media
globalization (Ferris, 2010; Turner, 2004). With regard to age, De Backer, Nelissen,
Vyncke, Braeckman, and McAndrew (2007) found that younger people prefer American
celebrities because they admire the glamour, while older people want to feel connected
to local celebrities. Age is thus expected to be positively related to cultural proximity.

H3: Younger people prefer global (American) celebrities and older people prefer
local celebrities

Further, Straubhaar (1991, 2003) found a preference for local media (television)
content among lower-educated people while higher-educated prefer global (US) content.
This can be linked to the greater access that higher-educated people have to new media,
promoting internationalization (Straubhaar, 1991, 2003). This can be translated to
celebrities as follows:

H4: Lower-educated people prefer local celebrities and higher-educated people

prefer global (US) celebrities

In PSI/PSR research, only a few studies have explicitly looked at the role of
gender (instead of using it as a control variable, as in Cohen, 2010; Wang et al., 2008)
but the findings are contradictory. Cohen (1997, 2003) and Lather and Moyer-Guse
(2011) found that women have stronger PSR, whereas Eyal and Dailey (2012) and Eyal

139
and Rubin (2003) show that men have stronger PSR. Therefore, this study examines the
following:

RQ1: How is gender related to PSR?

In PSI/PSR research, age and education have been considered even less
frequently than gender. Two exceptions are Cohen (2003) and Levy (1979). Cohen
(2003) looked at the difference between teens and adults with regard to PSR but found
no significant distinctions. Levy (1979), however, did find a significant, positive
relationship between age and PSI, implying that as people grow older, they have more
PSI. This was also found in a qualitative study by De Backer et al. (2007).

H5: There is a positive relationship between age and PSR

To our knowledge, Levys (1979) study is the only one to investigate education
and PSI and he found that lower-educated audience members maintain more PSI. Levy
(1979) links this to the higher level of social interactions for the higher-educated, which
may limit their need for parasocial replacements of social interactions. Another possible
reason is that higher-educated people may be less inclined to discuss their PSR with
celebrities, as this is considered a lower form of culture (Levy, 1979).

H6: There is a negative relationship between education and PSR

With the exception of Levys study (1979) which combines age and education, the
socio-demographic variables described above have not been combined in one model,
which is where this study wishes to contribute. In this study, PSR are further predicted
to be positively related to peoples celebrity news interest. This can be linked to the
degree to which audience members have mediated encounters with (favorite)
celebrities. Similarly, television affinity (e.g., Auter & Palmgreen, 2000; Lather & Moyer-
Guse, 2011) or TV news exposure (e.g., Levy 1979; Rubin & McHugh, 1987; Turner,
1993) are found to positively predict PSR. When transferred to celebrity culture,
television affinity or exposure could be translated into celebrity news interest.

H7: There is a positive relationship between celebrity news interest and PSR

140
 The relevance of cultural proximity for PSI/PSR is confirmed by work into
interpersonal liking which indicates that homophily or perceived similarity predicts a
strong association between attitude/background similarity and interpersonal liking
(Klimmt et al., 2006; Rubin & Step, 2000; Tian & Hoffner, 2010; Turner, 1993). Taking
into account this homophily concept, the celebritys cultural proximity (operationalized
as nationality or country of origin) can be an interesting predictor of PSR. The
hypothesis is therefore that:

H8: PSR are stronger with local than global celebrities

Method
Participants
To obtain information on audience members favorite celebrities and the nature and
predictors of PSR, data are gathered in a large-scale representative online survey
(N=1000) in Flanders, administered by iVOX (a Belgian research facilitator). The
respondents are 501 men and 499 women. Age is measured by means of three groups:
under 29 (20,1%), between 30 and 49 (40,9%), and over 50 (39,0%). Education is also
measured by means of three groups: lower secondary education (i.e., respondents have
finished the third year of secondary school or less; 37,1%), higher secondary education
(i.e., respondents have finished all six years of secondary school; 35,8%), and higher
education (i.e., respondents have finished college or university studies; 27,0%). This
sample is representative of the Flemish population in terms of gender (FOD Economie,
2010a), age (with a 12,2% overrepresentation of the 30-49 group, FOD Economie,
2010b), and education (FOD Economie, 2010c).

Procedure
The survey first asks respondents to name their favorite celebrity, with no restrictions in
terms of nationality, gender, or domain. This allows to grasp the range of celebrity types
in contemporary society. The selection of the favorite celebrity is followed by 19 five-
point Likert-type statements (1 = strongly disagree; 5 = strongly agree) on the PSR with
respondents favorite celebrities. Subsequently, respondents are asked to indicate their
gender, age, education level, and interest in celebrity news. The latter is done by means
of five five-point Likert-type items ( = .86) (see Table 1).

141
Table 1 Celebrity News Interest
Celebrity News
Nr. Item Mean SD
Interest
1. By reading news about celebrities, I often understand things
2.4 1.1 .819
better
2. Celebrities are important to me 2.1 1.0 .811
3. I often use information from celebrity news in my own life 2.3 1.1 .809
4. I actively search for celebrity news 2.1 1.0 .805
5. Celebrity news often urges me to think 2.8 1.2 .760
Note: Principal Components Analysis (VARIMAX); N = 1000

To enquire after PSR, the survey combines the relevant items from existing scales
focusing on the relational aspect with new items. From Auter and Palmgreens
Audience-Persona Interaction (API) scale (2000), Rubin and Perses 10-item PSI scale
(1987), Bocarnea and Browns CPPI (Celebrity-Persona Parasocial Interaction) scale
(2007), and Hartmann, Stuke, and Daschmanns Positive PSR scale (2008), those
relational items are selected that compare the celebrity to a friend, that investigate
empathy, the wish to read more about and meet the celebrity, and the latters
attractiveness. The remaining items are not selected as they examine other concepts
(e.g., identification), short-term PSI, or one specific type of celebrity. Additional items
are included on the analogy with social relationships (family, friendship [Adams &
Blieszner, 1994]) as well as the use of new social media in celebrity-audience
communication (see Baym, 2011). This results in a new scale of nineteen statements
that investigates long-term audience-celebrity connections and is applicable to celebrity
in its current diversity.

142
Table 2 Initial 19 items
1 I feel like I know MFC well
2 MFC could be my friend
3 I think MFC is attractive
4 I can empathize with the emotions of MFC
5 I learn from the acts of MFC
6 I often have the same point of view as MFC
7 When something bad happens to MFC, I feel bad
8 When something bad about MFC appears in the media, I feel hurt
9 When my friends laugh at MFC, I feel hurt
10 I like to talk about MFC with other people
11 I talk about MFC like I talk about my friends
12 I feel connected to MFC as I do to my friends
13 MFC is like a family member to me
14 I believe it is important to know everything about MFC
15 Sometimes I actively search for information on MFC
16 I would like to meet MFC
17 I have tried to get in contact with MFC
18 I dont want to meet MFC, because I believe the magic will disappear if I do
19 Being able to follow MFC on Facebook or Twitter, makes me feel close to him/her
Note: MFC = My Favorite Celebrity

Analysis
All analyses are conducted in SPSS and all respondents that indicated a favorite celebrity
are included in the analyses (N = 895). To test H1, a new PSR scale is constructed in
which the 19 initial items described above are subjected to a Principal Factor analysis.
H2, H3, and H4 are tested by means of the values from crosstab calculations. The
research question (RQ1) and hypotheses (H5, H6, H7, H8) on the predictors of the
strength of PSR are tested in preparatory t-tests, ANOVAs, and Pearsons correlation
matrix and, further, in a hierarchical regression analysis with the new PSR scale as
dependent variable and the respondents gender, age, education, and celebrity news
interest (step 1) and the celebrities gender, domain, and nationality (step 2) as
independent variables.

The nature of PSR

As stated above, a new PSR scale is developed by means of a Principal Factor analysis
using oblique Direct OBLIMIN rotation (as the factors are highly correlated [.69]) to
examine the nature of PSR. The choice for a factor analysis instead of a components
analysis is motivated by the fact that component analyses are principally a-theoretical
and aimed at data reduction instead of looking for underlying constructs (Van den
Bosch, 2008).

143
 Initially, four factors are retained in the analysis, but after selecting the items
loading higher than .50 on one factor and less than .50 on the others, only two factors
are retained. Thus, thirteen items remain, representing two factors that can be defined
as Emotional Connection (M = 3.04, SD = 0.82) and Analogy to Social Relationships (M =
2.21, SD = 0.86) and that, together, constitute the PSR scale (M = 2.59, SD = 0.76) (see
Table 3 for items and factor loadings). The variance explained by these factors is 52
percent and the reliability results in a Cronbachs alpha of .91 for the entire scale, .85 for
the first factor (Emotional Connection; mean inter-item correlation = .49), and .89 for
the second factor (Analogy; mean inter-item correlation = .54) which indicates high
internal consistency and reliability. The first factor contains six items on empathy with
and comprehension of the celebrity, which confirms that an emotional connection with a
celebrity is an essential aspect of PSR. The second factor consists of seven items that
explore the comparison between the celebrity and friends or family, reaffirming the
expected importance of this analogy for the definition of PSR.

Table 3 The PSR scale

Analogy to
Emotional
Nr. Item Mean SD social
connection
relationships
1. When something bad happens to MFC, I feel bad 3.09 1.13 .793 -.031
2. When something bad about MFC appears in the media, I
2.85 1.13 .723 .069
feel hurt
3. When my friends laugh at MFC, I feel hurt 2.74 1.17 .682 .049
4. I learn from the acts of MFC 3.06 1.06 .660 .016
5. I often have the same point of view as MFC 3.16 .93 .654 -.042
6. I can empathize with the emotions of MFC 3.37 1.05 .638 .021
7. MFC is like a family member to me 1.96 1.00 .007 .837
8. I talk about MFC like I talk about my friends 2.16 1.09 -.005 .824
9. I feel connected to MFC as I do to my friends 2.17 1.09 .084 .786
10. Being able to follow MFC on Facebook or Twitter, makes
2.10 1.06 -.081 .727
me feel close to him/her
11. I have tried to get in contact with MFC 1.77 1.07 -.053 .647
12. I believe it is important to know everything about MFC 2.52 1.17 .145 .641
13. Sometimes I actively search for information on MFC 2.75 1.28 .131 .505
Note: Principal Factor Analysis (Direct OBLIMIN); MFC: My Favorite Celebrity

H1 is thus supported and PSR can be defined as illusions of long-term friendships

between audience members and celebrities, which are one-sided and created by the
media, but encompass emotional connections and are similar to social relationships.

144
Who are the Favorite Celebrities?
Ninety percent of the respondents indicate having a favorite celebrity and a range of 382
different celebrities is named. The top 10 contains five local Flemish (TV presenter Koen
Wauters, singers Helmut Lotti & Will Tura, retired tennis player Kim Clijsters, and
retired cyclist Eddy Merckx), four American (actors Johnny Depp, Angelina Jolie, and
George Clooney and president Barack Obama), and one Irish celebrity (singer Bono). In
the sample of 382 celebrities, the respondents mainly choose male celebrities (72%) and
celebrities from music (40%), film (26%), TV/radio (13%), sports (11%), and politics
(6%). Boon and Lomore (2001) found a similar dominance of music and film celebrities
in their study. Most respondents choose local Flemish celebrities as their favorite (44%),
followed by Americans (35%), implying that, while cultural proximity is important,
international American celebrities are also appealing, confirming a lasting dominance of
the US celebrity culture as a result of globalization processes (cf. Van den Bulck &
Claessens, 2013c).
In terms of respondents gender, both men (71%) and women (73%) prefer male
over female celebrities ([1, N = 895] = 0.47, p = .49), supporting H2 and confirming
earlier findings (Boon & Lomore, 2001; Cohen, 1997, 2003). Both men and women
almost equally prefer Flemish celebrities (men = 42%, women = 46%), followed by
Americans (men = 36%, women = 34%) ([6, N = 894] = 13.31, p = .038).
All three respondents age groups have a preference for male celebrities ([2, N
= 895] = 0.31, p = .857). Interestingly, respondents over the age of 50 name fewer
international celebrities and more local celebrities (59%), whereas respondents
younger than 30 prefer international, especially American (47%), celebrities ([12, N =
895] = 70.32, p < .001). The 30-49 age group lies in between, slightly preferring
American celebrities (40%) over the Flemish (37%). This supports H3 as well as De
Backer et al.s (2007) findings that younger people prefer global, American celebrities
because they admire the glamour, while older people want to feel connected to local
celebrities.
In terms of education, all groups prefer male over female celebrities ([2, N =
895] = 2.51, p = .285). Interestingly, respondents with a lower secondary education
prefer Flemish (56%) over American celebrities (27%), contrasting the group with a
higher secondary (40% American; 39% Flemish) and a higher education (39%

145
American, 34% Flemish) ([12, N = 895] = 40.14, p < .001). Cultural proximity thus
seems to play an especially important role for lower-educated groups, supporting H5.

Predictors of PSR
To examine the relations between the PSR scale and the characteristics of the
respondents (gender, age, education, and celebrity news interest) and the celebrities
(gender, domain, and nationality), preparatory t-tests (for gender), one-way ANOVA
tests (for age, education, nationality, and domain), and a Pearson correlations matrix
(for celebrity news interest) are computed. These preparatory tests indicate which
variables should be included in the regression analyses and give a first impression of
their relation to PSR. The t-tests show that the respondents gender is significantly
related to the full PSR scale, t(893) = 2.19, p = .029, and the Analogy factor ,t(893) = 2.55,
p = .011, but not for the Emotional Connection factor, t(893) = 1.30, p = .196. The male
respondents (M = 2.65, SD = 0.76, N = 454) maintain slightly stronger PSR than the
women (M = 2.54, SD = 0.75, N = 441) and the men (M = 2.28, SD = 0.86) consider PSR to
be slightly more analogous to social relationships than the women (M = 2.13, SD = 0.85).
The celebrities gender is not significantly correlated to the PSR scale, t(893) = 1.48, p =
.138, or the factors (Emotional Connection t(893) = 1.32, p = .187, and Analogy t(893) =
1.36, p = .175) and will not be included in subsequent analyses.
The one-way ANOVA tests demonstrate that the respondents age is significantly
related to the PSR scale, F(2, 892) = 18,80, p < .001, and the two factors (Emotional
Connection, F(2, 892) = 21.57, p < .001 and Analogy, F(2, 892) = 10.99, p < .001). Post-
hoc Scheffe tests show that not all three age groups differ significantly for the PSR scale
and the two factors: only the youngest (<29) and oldest group (>50). These two age
groups are dummified and included in the regression analyses with the middle age
group (30-49) as reference category. Respondents educational level also relates
significantly to the PSR scale, F(2, 892) = 28.07, p < .001, and Emotional Connection, F(2,
892) = 25.75, p < .001, and Analogy, F(2, 892) = 21.54, p < .001. The educational levels
are dummified and included in the analyses, with lower secondary education as the
reference category. With regard to the nationality of the celebrities (with seven
categories), the ANOVA shows a significant relation with the PSR scale, F(6, 888) = 8.82,
p < .001, and the two factors (Emotional Connection, F(6, 888) = 10.74, p < .001, and
Analogy, F(6, 888) = 5.23, p < .001). Post-hoc Scheffe tests show that only the American

146
and Flemish nationalities differ significantly for the full scale and the factors. Therefore,
only the local, Flemish nationality is dummified and included in the analyses, with the
global, American nationality as reference category. Finally, the celebrity domain is
significantly related to the PSR scale, F(10, 884) = 6.45, p < .001, and the Emotional
Connection, F(10, 884) = 9.74, p < .001 and Analogy factors, F(10, 884) = 3.86, p < .001.
In post-hoc Scheffe tests, only particular domains differ significantly for the PSR scale
and the factors, namely film, music, sports, politics, and religion. Only these domains are
dummified with film as the reference category and all but film are included in
subsequent analyses. The Pearson correlation matrix shows that the celebrity news
interest is significantly related to the PSR scale, r(891) = .54, p < .001, and the factors of
Emotional Connection, r(891) = .42, p < .001, and Analogy, r(891) = .54, p < .001.
To examine the relative importance of these variables as predictors for the
strength of PSR within one combined model, a hierarchical regression analysis is
performed.

PSR
As observed in Table 4, 37 percent of the strength of PSR can be predicted by the
respondents gender, age (>50), education (higher secondary and higher), celebrity
news interest and the celebrities domain (politics and religion) and nationality
(Flemish). Interestingly, the respondents characteristics explain most of the variance
for PSR (37%), whereas the celebrities characteristics only add two percent (p < .001).
The strongest predictors (p < .001) are the respondents higher education and celebrity
news interest as well as the celebrities domain of religion. The respondents with a
higher education have weaker PSR than the lower-educated respondents, confirming
H6. Respondents with a high interest in celebrity news have strong PSR, supporting H7.
Further, PSR with religious celebrities tend to be stronger than those with film
celebrities (the reference category). The respondents gender is negatively related to
PSR, indicating that men have stronger PSR than women (p < .05). RQ1 can thus be
answered by stating that men have stronger PSR than women, confirming the findings of
Eyal and Dailey (2012) and Eyal and Rubin (2003) but contrasting those of Cohen (1997,
2003) and Lather and Moyer-Guse (2011). Respondents in the age group of 50 and older
tend to have stronger PSR than those in the younger age group (30-49 is the reference
category) (p < .05), confirming H5, and respondents with a higher secondary education

147
have weaker PSR than those with a lower secondary education or less (p < .05), once
again confirming H6. Further, PSR tends to be stronger for politicians than for film
celebrities and for local Flemish than for global (US) celebrities (both p < .05). The latter
finding confirms H8 as cultural proximity is a positive predictor of PSR strength.

Table 4 Hierarchical regression analysis for PSR

PSR B (SE) R R change F
1 Gender (R) -.13 (.04) -.09* .35 .35** F(6,889) = 78.47**
Age (R):
- < 29 -.08 (.06) -.05
- >50 .21 (.05) .13**
Education (R):
- Higher secondary -.14 (.05) -.09*
- Higher -.25 (.06) -.15**
Celebrity news interest .40 (.02) -.52**
(R)

2 Gender (R) -.13 (.04) -.08* .37 .02** F(11,884) = 47.00**

Age (R):
- < 29 -.07 (.06) -.04
- >50 .15 (.05) .10*
Education (R):
- Higher secondary -.12 (.05) -.08*
- Higher -.25 (.06) -.14**
Celebrity news interest .39 (.02) .50**
(R)
Domain (C):
- Music .09 (.05) .06
- Sports .14 (.07) .06
- Politics .21 (.10) .06*
- Religion .81 (.21) .11**
Nationality (C):
- Flemish (local) .12 (.04) .08*
Note: *p < .05, **p < .001 [R=Respondent; C=Celebrity], N = 895

To gain a more profound understanding of the relative importance of these

characteristics of respondents and celebrities for the two main elements of PSR, a
hierarchical regression analysis is conducted for the two factors.

Emotional Connection
The total variance explained for the emotional connection factor is 28 percent, of which
the respondents characteristics explain 23 percent and the celebrities variables four
percent (p < .001) (Table 5). The respondents higher education and celebrity news
interest are strongly significant (p < .001) predictors, indicating that lower-educated
people with a high celebrity news interest have stronger emotional connections with
their favorite celebrities. Further, emotional connections are stronger for politicians and
148
religious celebrities (compared to film celebrities, p < .001). The respondents in the age
group of 50 and older further have stronger emotional connections than the middle
group (30-49) (p < .05) and those with a higher secondary education have weaker
emotional connections than those with a lower secondary education or less (p < .05).
Emotional connections are further stronger for sports than film celebrities (p < .05) and
for local Flemish than global (US) celebrities (p < .05).

Table 5 Hierarchical regression analysis for Emotional Connection

Emotional Connection B (SE) R R change F
1 Gender (R) -.09 (.05) -.06 .24 .24** F(6,889) = 45.21**
Age (R):
- < 29 -.09 (.07) -.04
- >50 .25 (.06) .15**
Education (R):
- Higher secondary -.20 (.06) -.12*
- Higher -.26 (.07) -.14**
Celebrity news interest .33 (.03) .40**
(R)

2 Gender (R) -.07 (.05) -.04 .28 .04** F(11,884) = 30.57**

Age (R):
- < 29 -.08 (.06) -.04
- >50 .16 (.06) .09*
Education (R):
- Higher secondary -.19 (.06) -.11*
- Higher -.26 (.06) -.14**
Celebrity news interest .32 (.02) .38**
(R)
Domain (C):
- Music .02 (.08) .01
- Sports .20 (.08) .08*
- Politics .43 (.11) .12**
- Religion .99 (.24) .12**
Nationality (C):
- Flemish (local) .17 (.05) .10*
Note: *p < .05, **p < .001 [R=Respondent; C=Celebrity], N = 895

Analogy to Social Relationships

The variables in Table 6 explain 35 percent of the variance of the analogy to social
relationships. The respondents variables explain 34 percent and the celebrities
characteristics only one percent (p < .001). The strongest predictors are respondents
higher education and celebrity news interest (p < .001), followed by gender and the age
group of 50 and older (p < .05), indicating that analogy to social relationships is higher
for older (>50), lower-educated men who are highly interested in celebrity news. For the
celebrity characteristics, only the music and religious domains are significant predictors

149
(p<.05), implying that the analogy to social relationships is stronger for music and
religious than for film celebrities.

Table 6 Hierarchical regression analysis for Analogy to Social Relationships

Analogy to Social B (SE) R R change F
1 Gender (R) -.16 (.05) -.10* .34 .34** F(6,889) = 74.87**
Age (R):
- < 29 -.08 (.06) -.04
- >50 .18 (.06) .10*
Education (R):
- Higher secondary -.08 (.06) -.05
- Higher -.25 (.06) -.13**
Celebrity news interest .46 (.02) .53**
(R)

2 Gender (R) -.16 (.05) -.10* .35 .01* F(11,884) = 42.97**

Age (R):
- < 29 -.07 (.06) -.03
- >50 .14 (.06) .08*
Education (R):
- Higher secondary -.07 (.06) -.04
- Higher -.23 (.06) -.12**
Celebrity news interest .45 (.02) .52**
(R)
Domain (C):
- Music .14 (.05) .08*
- Sports .09 (.08) .03
- Politics .01 (.11) .00
- Religion .66 (.24) .08*
Nationality (C):
- Flemish (local) .07 (.05) .04
Note: *p < .05, **p < .001 [R=Respondent; C=Celebrity], N = 895

Discussion
This article wants to help fill the gaps in PSR research, namely its focus on television
celebrities, the confusion between PSI and PSR, the use of student samples, and the lack
of consideration of socio-demographic variables as predictors of PSR. The current study
therefore comprises a wide range of celebrities, both in the respondents selection of
favorite celebrities and in the construction of the new PSR scale. To address the
confusion between short-term PSI and long-term PSR, the newly developed scale
explicitly focuses on PSR by combining the relational items from previous scales with
several new items comparing PSR to friendships/family bonds and one item on new
media. Further, the representative sample provides insight into celebrity preferences
and PSR in the broad adult population of Flanders. Finally, the relationship between, on
the one hand, the celebrity preferences and the strength of PSR, and on the other hand,
the respondents gender, age, and education is analyzed in one combined model which
150
also includes the respondents interest in celebrity news and the celebrities gender,
domain, and nationality.
In this study, eight hypotheses and one research question are addressed. In a
factor analysis, the two main elements of PSR are identified in accordance with the
literature: emotional connection and analogy to social relationships. This confirms the
definition of PSR as illusions of long-term friendships between audience members and
celebrities, which are one-sided and created by the media, but encompass an emotional
connection and are similar to social relationships.
With regard to the respondents celebrity preferences, H2, H3, and H4 are
supported, which means that male celebrities are preferred by both men and women
(H2) and that cultural proximity is more important for older than for younger
respondents (H3) and more important for lower-educated than for higher-educated
respondents (H4). This confirms previous research for the gender preferences (Boon &
Lomore, 2001), the preference for local celebrities among older people and for global
(US) celebrities among younger people (De Backer et al., 2007), and for the preference
for local celebrities among lower-educated people and for global (US) celebrities among
higher-educated people (Straubhaar, 1991). The preference of younger adults for global
celebrities can be linked to the globalization processes of the past decades, bringing
small countries in closer contact with international celebrities, while, in the past,
audiences were more restricted to local celebrities (Turner, 2004). The educational
differences can be related to Straubhaars (1991: 51) finding that a greater
traditionalism and loyalty to national and local cultures by lower or popular classes,
who show the strongest tendency to seek greater cultural proximity in television
programs and other cultural products.
The strength of PSR and the two factors is predicted most strongly and
significantly by the respondents characteristics and, in particular, their celebrity news
interest confirming H7 and education confirming H6 especially for the highest-
educated group. The former confirms the comparability of television affinity (Lather &
Moyer-Guse, 2011) and celebrity news interest as a factor that strongly increases PSR.
The latter is interesting because the highly significant differences with regard to
education only come up when the highest-educated group is compared with the lowest-
educated respondents, which nuances Levys (1979) findings that PSR increase as
education decreases. Indeed, the difference between the lower secondary and higher

151
secondary education groups in the sample are less strong and significant. Education
remains an underexplored topic in audience research in the realm of PSR and the wider
domain of celebrity studies and should be examined in future research.
Further, with regard to RQ1, respondents gender is negatively linked to PSR and
the analogy factor which implies that men maintain stronger PSR with their favorite
celebrities than women and that they consider PSR to be more like social relationships
than women. As stated above, this confirms the findings by Eyal and Dailey (2012) and
Eyal and Rubin (2003) but contrasts those of Cohen (1997, 2003) and Lather and
Moyer-Guse. As most studies start from the assumption that women maintain stronger
PSR than men, because women are better adapted to maintain social relationships and
have larger social networks (Cohen, 1997, 2003), this finding is important and should be
explored further in future research. One potential explanation is that men are less able
to maintain their networks and therefore turn to PSR to fulfill their social needs, as they
consider them more similar to social relationships than women. These findings are
somewhat surprising considering that celebrities usually attract a more female following
(Sandvoss, 2005: 16) and demonstrate the need to explore gender in future research.
Age is a significant predictor of PSR and the two factors but only for the
distinction between the oldest (50 or older) and middle (30-49) group. This finding is
interesting as it nuances previous findings on age and PSR (De Backer et al., 2007; Levy,
1979), namely that PSR with favorite celebrities become significantly stronger starting
from the age of 50. This can be linked to the specific characteristics of older adults as
they have more free time (after retirement), are often restricted in terms of mobility,
and have decreasing social networks (Drageset, 2004), all factors that may increase
(celebrity) media use and thus PSR with celebrities. These findings stress the need for
more research into the role of celebrities in the lives of older adults.
On the side of the respondents, PSR are thus stronger among older, lower-
educated men who are highly interested in celebrity news. On the side of the celebrities,
both domain and nationality are significant PSR predictors. With regard to domain,
stronger PSR are found with religious celebrities and politicians. Further, emotional
connections with celebrities are stronger with regard to religious celebrities, politicians,
and sports celebrities as compared to film celebrities. This may be linked to the central
role of emotion or affect in these domains: religion, politics, and sports (Marshall, 1997;
Rojek, 2001; Van Zoonen, 2005). Finally, the celebrities nationality or cultural proximity

152
is a significant predictor, with stronger PSR for local Flemish celebrities as compared to
global American celebrities. With regard to the two elements, only the emotional
connection with the favorite celebrities is significantly related to the celebrities cultural
proximity, with stronger emotional connections with local celebrities than with global
celebrities. Once again, this nuances previous research into cultural proximity and
interpersonal liking (e.g., Tian & Hoffner, 2010).
This study has several limitations as, first, the sample does not include
respondents younger than 18, while previous research shows teenagers have strong PSR
(De Backer et al., 2007). Second, the PSR scale only measures positive PSR, whereas
Hartmann et al. (2008) and Tian and Hoffner (2010) point to the relevance of negative
PSR, which should be included in future research. Third, new applications of the PSR
scale should be conducted to test the reliability in other contexts. Finally, qualitative
research is needed to analyze the specific processes involved in the development and
maintenance of PSR. Despite these shortcomings, this study demonstrates the relevance
of the role of celebrities in audience members everyday lives. It reveals that audiences
gender, age, education, cultural proximity, and celebrity news interest as well as
celebrities domains and nationalities are relevant factors for PSR which should be
systematically included in future research in order to gain more profound insight in the
workings of celebrity culture from an audience perspective.

153
References
Adams, R. G., & Blieszner, R. (1994). An integrative conceptual framework for friendship
research. Journal of Social and Personal Relationships, 11(2), 163-184.
doi:10.1177/0265407594112001
Auter, P. J., & Palmgreen, P. (2000). Development and validation of a parasocial
interaction measure: The audience-persona interaction scale. Communication
Research Reports, 17(1), 79-89.
Baym, N. (2011, May). Fans or friends? Paper presented at the annual meeting of the
International Communication Association, Boston.
Bocarnea, M. C., & Brown, W. J. (2007). Celebrity-persona parasocial interaction scale. In
R. A. Reynolds, R. Woods, & J. D. Baker (Eds.), Handbook of research on electronic
surveys and measurements (pp. 309-312). London: Idea Group.
Boon, S. D., & Lomore, C. D. (2001). Admirer-celebrity relationships among young adults:
explaining perceptions of celebrity influence on identity. Human Communication
Research, 27(3), 432-465. doi: 10.1111/j.1468-2958.2001.tb00788.x
Cohen, E.L. (2010). Expectancy violations in relationships with friends and media
figures. Communication Research Reports, 27(2), 97-111.
Cohen, J. (1997). Parasocial relations and romantic attraction: gender and dating status
differences. Journal of Broadcasting & Electronic Media, 41, 516-529. doi:
10.1080/08838159709364424
Cohen, J. (2003). Parasocial breakups: Measuring individual differences in responses to
the dissolution of parasocial relationships. Mass Communication & Society, 6(2),
191-202. doi: 10.1207/S15327825MCS0602_5
Cohen, J. (2009). Mediated relationships and media effects: Parasocial interaction and
identification. In R. L. Nabi, & M. B. Oliver (Eds.), The SAGE handbook of media
processes and effects (pp. 223-236). London: Sage.
De Backer, C. J. S., Nelissen, M., Vyncke, P., Braeckman, J., & McAndrew, F. T. (2007).
Celebrities: From teachers to friends. Human Nature, 18(4), 334-354. doi:
10.1007/s12110-007-9023-z
Drageset, J. (2004). The importance of activities of daily living and social contact for
loneliness: a survey among residents in nursing homes. Scandinavian Journal of
Caring Sciences, 18, 65-71.
Dyer, R. (1998). Stars. London: British Film Institute.

154
Evans, J., & Hesmondalgh, D. (2005). Understanding media: Inside celebrity. Maidenheid:
Open University Press.
Eyal, K., & Dailey, R.M. (2012). Examining relational maintenance in parasocial
relationships. Mass Communication & Society, 15(5), 758-781.
Eyal, K., & Rubin, A.M. (2003). Viewer Aggression and Homophily, Identification, and
Parasocial Relationships with Television Characters. Journal of Broadcasting &
Electronic Media, 47(1), 77-98.
FOD Economie (2010a). Bevolking van Belgische en vreemde nationaliteit naar geslacht.
Retrieved August 24, 2011 at
http://statbel.fgov.be/nl/binaries/N%2000.01%20y-2009%20AAE_tcm325-
128579.xls.
FOD Economie (2010b). Bevolking per geslacht, leeftijdsgroep en klasse 1990-2008.
Vlaams Gewest. Retrieved August 24, 2011 at
http://statbel.fgov.be/nl/binaries/Structuur%20v.%20d.%20bevolking%20per
%20leeftijd%20en%20geslacht%20-%20Vlaams%20Gewest%20-
%20Per%20jaar_finale_tcm325-114594.xls.
FOD Economie (2010c). Onderwijsniveau van de bevolking van 15 jaar en ouder in
procenten (1987-2010). Retrieved August 24, 2011 at
http://statbel.fgov.be/nl/binaries/Onderwijsniveau1987-2010_tcm325-
44615.xls.
Giles, D. (2002). Parasocial interaction : a review of the literature and a model for future
research. Media Psychology, 4, 279-305. doi: 10.1207/S1532785XMEP0403_04
Giles, D. (2003). Media psychology. London : Routledge.
Hartmann, T., Stuke, D., & Daschmann, G. (2008). Positive parasocial relationships with
drivers affect suspense in racing sports spectators. Journal of Media Psychology,
20(1), 24-34. doi: 10.1027/1864-1105.20.1.24
Horton, D., & Wohl, R. (1956). Mass communication and parasocial interaction:
Observations on intimacy at a distance. Psychiatry, 19, 215-229.
Klimmt, C., Hartmann, T., & Schramm, H. (2006). Parasocial interactions and
relationships. In J. Bryant, & P. Vorderer (Eds.), Psychology of entertainment (pp.
291-313). London: Lawrence Erlbaum Associates.

155
Lather, J., & Moyer-Guse, E. (2011). How do we react when our favorite characters are
taken away? An examination of a temporary parasocial breakup. Mass
Communication and Society, 14, 196-215.
Levy, M. R. (1979). Watching TV news as para-social interaction. Journal of Broadcasting,
23, 69-80. doi: 10.1080/08838157909363919
Marshall, P. D. (2006). The celebrity culture reader. London: Routledge.
Rojek, C. (2001). Celebrity. London: Reaktion Books.
Rubin, A. M., & Perse, E. M. (1987). Audience activity and soap opera involvement: A uses
and effects investigation. Human Communication Research, 14, 246-268. doi:
10.1111/j.1468-2958.1987.tb00129.x
Rubin, A. M., & Step, M. M. (2000). Impact of motivation, attraction, and parasocial
interaction on talk radio listening. Journal of Broadcasting & Electronic Media,
44(4), 635-654. doi: 10.1207/s15506878jobem4404_7
Rubin, R. B., & McHugh, M. P. (1987). Development of parasocial interaction
relationships. Journal of Broadcasting & Electronic Media, 31(3), 279-292. doi:
10.1080/08838158709386664
Sandvoss, C. (2005). Fans. Cambridge: Polity.
Schiappa, E., Gregg, P. B., & Hewes, D. E. (2005). The parasocial contact hypothesis.
Communication Monographs, 72(1), 92-115. doi:
10.1080/0363775052000342544
Schramm, H., & Hartmann, T. (2008). The PSI-Process Scales. A measure to assess the
intensity and breadty of parasocial processes. Communications, 33, 385-401.
Straubhaar, J. D. (1991). Beyond media imperialism: Assymetrical interdependence and
cultural proximity. Critical Studies in Mass Communication, 8, 39-59.
Straubhaar, J. D. (2003). Choosing national TV: Cultural capital, language, and cultural
proximity in Brazil. In M.G. Elasmar (Ed.), The impact of international television: A
paradigm shift (pp. 77-110). Mahwah, NJ: Lawrence Erlbaum Associates.
Tian, Q., & Hoffner, C. A. (2010). Parasocial interaction with liked, neutral, and disliked
characters on a popular tv series. Mass Communication and Society, 13(3), 250-
269. doi: 10.1080/15205430903296051
Turner, G. (2004). Understanding celebrity. London : Sage.

156
Turner, J.R. (1993). Interpersonal and psychological predictors of parasocial interaction
with different television performers. Communication Quarterly, 41(4), 443-453.
doi: 10.1080/01463379309369904
Van den Bosch, K. (2008). Syllabus factoranalyse. Antwerp: University of Antwerp.
Van den Bulck, H., & Claessens, N. (2013a). Celebrity Suicide and the Search for the
Moral High Ground: Comparing Frames in Media and Audience Discussions of the
Death of a Flemish Celebrity. Critical Studies in Media Communication, 30(1), 69-
84.
Van den Bulck, H., & Claessens, N. (2013b). Guess who Tiger is having sex with now?
Celebrity sex and the framing of the moral high ground. Celebrity Studies, 4(1),46-
57.
Van den Bulck, H., & Claessens, N. (2013c). Of Local and Global Fame: A Comparative
Analysis of News Items and Audience Reactions on Celebrity News Websites
People, Heat, and HLN. Journalism [online]. Available at:
http://jou.sagepub.com/content/early/2013/05/28/1464884913488725
Van Zoonen, L. (2005). Entertaining the citizen: When politics and popular culture
converge. Oxford: Rowman & Littlefield.
Wang, Q., Fink, E.L., & Cai, D.A. (2008). Loneliness, gender, and parasocial interaction: A
uses and gratifications approach. Communication Quarterly, 56(1), 87-109.

157
 Chapter 5

The last two empirical chapters of this PhD focus on one specific group of celebrity audiences by
building on the findings of the previous chapter, namely that older audience members tend to
maintain strong connections with celebrities. This is especially interesting considering how
strongly this particular group is neglected in academic research, even in more general media
studies. Combining the findings of the previous chapter with the societal developments of global
aging and an increasingly omnipresent celebrity culture, there is a pressing need to explore the
relevance of celebrities for older adults. Chapters 5 and 6 focus on one specific group of older
adults nursing home residents as they are characterized by a lot of free time, restricted
mobility, and shrinking social networks, all factors which are believed to increase their media
and celebrity use. The study of celebrity in the context of the nursing home is innovative, not just
in terms of its focus but also by bridging the academic domains of gerontology and celebrity
studies. The exploratory study in chapter 5 examines the celebrity preferences of nursing home
residents and the social roles of celebrity by means of a combination of qualitative interviews
(individual semi-structured and focus group). The main social roles of celebrity for nursing
home residents are as food for social conversations, leads into moral discussions and help for
meaning-making processes, social companions, and aide-memoires.

AN EXPLORATORY STUDY OF THE SOCIAL ROLES OF CELEBRITIES IN THE

EVERYDAY LIVES OF NURSING HOME RESIDENTS

Nathalie Claessens
(A shortened version will be published in Aging, Media, Culture by C.L. Harrington, D.B. Bielby,
& A.R. Bardo [eds.])

As contemporary Western audiences live in societies characterized by global aging and a

growing celebrity culture, this chapter combines the disciplines of gerontology and
celebrity studies to examine their junction: the importance of celebrities in the lives of
older adults. With notable exceptions, this has not obtained much research attention,
presenting a gap in both domains. This chapter empirically explores the meaning of
celebrities for nursing home residents. In individual interviews and focus groups with a
sample of nursing home residents, their celebrity preferences and celebrities social roles
as food for conversations, leads into moral/social discussions, and sources for
meaning-making, companions in a decreasing social network, or aide-memoires are
explored.

158
While contemporary Western societies are increasingly characterized by processes of
global aging and an omnipresent celebrity culture, academic research has hardly
addressed the combination of these issues celebrities in the lives of older adults with
some notable exceptions (e.g., Harrington & Bielby, 2010; Harrington, Bielby, & Bardo,
2011; Van den Bulck & Van Gorp, 2012). Even within the larger domain of media studies,
little research has been conducted with regard to older adults (Carmichael, 1976).
However, this topic is in need of extensive exploration because, as people live longer and
grow older, they have more time to potentially invest in media and celebrity (news).
This may entail that celebrity culture and individual celebrities become important in the
lives of older adults in a variety of social roles and meanings. This is especially the case
for a specific subgroup of the older adult population nursing home residents who are
the focus of this study because their specific circumstances (leisure time, mobility, &
social networks) may increase their mediated encounters with and interest in celebrities
and, hence, the importance of celebrities (and the accompanying social roles) in their
lives.
Previous research demonstrated the importance of celebrities for nursing home
residents as food for gossip strengthening social relationships and replacements for
lost real-life social contacts in the form of parasocial relationships (De Backer et al.,
2007). In more general celebrity research, celebrities are believed to not just provide
food for entertainment or gossip, but to fulfill social functions for audiences and have
meaning beyond mere pleasure: as a topic for conversation, a lead into moral
discussions, a source for meaning-making, or social companions in the form of
parasocial relationships (Dyer, 1998; Evans & Hesmondalgh, 2005; Giles, 2003;
Marshall, 2006; Rojek, 2001; Turner, 2004). A qualitative approach is used here to
examine whether the celebrity roles for the general audience can be translated to the
context of nursing homes. The combination of individual and focus group interviews is
ideal here as we are interested in the personal celebrity preferences of the individual
residents as well as the roles and meanings of celebrity in a social space, namely a
nursing home.

159
Celebrity in the Lives of Older Adults
The academic neglect of older adults in media research has been criticized before
(Carmichael, 1976; van der Goot, 2009) and can be extended to celebrity studies
(Harrington & Bielby, 2010). Indeed, although there are several audience studies on
celebrities and older adults (e.g., Harrington, Bielby, & Bardo, 2011; Stever, 2011; Van
den Bulck & Van Gorp, 2012), their applicability to the general older adult population is
limited as they focus exclusively on fans, a specific segment of the audience with an
intense involvement with their fan object (Sandvoss, 2005) which cannot be equated
with the general audience. This is where this chapter wishes to contribute, by providing
insight in the roles of celebrity in the lives of the not necessarily fan older audience.
As stated above, this chapter focuses on one subgroup of older adults, namely
nursing home residents, as they are found to have a lot of leisure time, to be restricted in
terms of mobility, and to have decreasing social networks (Drageset, 2004), all factors
which may increase their use of (celebrity) media. The typical nursing home resident
can be described as an eighty-year-old widow with three or more chronic conditions
that render independent living impossible (Hajjar, 1998). Within the broad and diverse
older adult population, nursing home residents are often defined as the old-old (>80 for
Carmichael 1976, 125), the dependent and unhealthy, the fourth age (Baltes, 1997;
Harrington, Bielby, & Bardo 2011, 9), or the frail recluses (Moschis, 1996). Misleadingly,
research on older adults in general and nursing home residents in particular usually
focuses on the losses related to old age, namely physical problems, memory deficits, loss
of spouse or friends, changing social relations, approaching death and less time to live,
and the need to evaluate life (Baltes, 1997; Vandebosch & Eggermont, 2002; van der
Goot, 2009; van der Goot, Beentjes, & van Selm, 2012). However, aging also entails
advantages which should be included in academic research, such as more free time,
more leisure possibilities, a lot of (life) knowledge, social confidence, and more
emotional experience and skills (Carstensen, Fung, & Charles, 2003; Harrington, Bielby,
& Bardo, 2011; Vandebosch & Eggermont, 2002).
Contemporary Western societies are characterized by an increasingly
omnipresent celebrity culture, which is not limited to gossip magazines and popular
television talk shows, but extends to almost every medium (popular and quality,
newspapers, radio, television, cinema, Internet) and every domain of life (popular and
high culture, politics, business, advertising, charity, and so forth) (Holmes & Redmond,

160
2010; Marshall, 2006). This growing presence of celebrity in media can be linked to
changes in the importance of news values, where personalization and soft news
embodied in celebrity news are gaining importance (Turner, 2004; Whannel, 2010). In
this chapter, the phenomenon of celebrity is defined as a construction created and
maintained by four actors the combination of whom constitutes the celebrity
apparatus namely the famous person and his/her entourage, the media, and the
audience (Marshall, 2006; Rojek, 2001). The celebrity construction is described as the
celebritization process (Turner, 2004), in which a public figures private life becomes (at
least) as important to media and audiences as his/her professional activities. There are
three personae in the celebrity construct that are of interest for media and audiences:
public (professional activities), private (the official private life), and real persona
(behind the celebrity faade) (Dyer, 1987; Holmes, 2005). Within the large variety of
celebrity, Rojek (2001) distinguishes between ascribed (predetermined by birth, e.g.,
royalty), achieved (based on accomplishments or talent, e.g., sportsmen), and attributed
(result of extensive mass media attention, e.g., reality TV) celebrity. Despite the apparent
triviality of celebrities, they have multiple meanings for audiences that go beyond
entertainment and include food for social interactions, means to discuss moral issues
and representations of what it is to be human (Dyer, 1998: 99), and social companions
in the form of parasocial relationships (Dyer, 1998; Evans & Hesmondalgh, 2005; Giles,
2003; Harrington, Bielby, & Bardo, 2011; Marshall, 2006; Rojek, 2001; Stever, 2011;
Tulloch, 1991; Turner, 2004). The latter can be described as relationships in which
audience members have the illusion of friendship with a celebrity, which is ultimately
mediated and one-sided, but entails an emotional connection and is similar to social
relationships (Claessens & Van den Bulck, 2011; Giles, 2002). Celebrities thus have
important meanings and social functions that, on a theoretical level, are generally
accepted within the academic domain of celebrity studies. However, with the exception
of several case studies (e.g., Butler Breese, 2010; Feasey, 2008; Van den Bulck &
Claessens, 2013), there is little empirical research into the roles of celebrities in the
everyday lives of audience members. Even less academics have looked into the meaning
of celebrities in the lives of older adults, and more specifically, nursing home residents.
Therefore, this chapter examines whether the social roles described above are also
relevant for nursing home residents and whether there are any additional meanings that
are characteristic of the older adult group. For instance, considering the memory-related

161
issues that older adults and, in particular, nursing home residents are faced with,
celebrities may act as an aide-memoire, a means to foster reminiscence in a positive,
fun, and light manner and thereby enhance self-esteem (Bender, Bauckham, & Norris,
1999). Celebrities could thus help older adults to structure the memories of their life
course by linking particular life events with particular celebrities and their life events
and celebrities could stimulate reminiscence. Studies have found that reminiscence is
also an important aspect of the identity construction when people age, especially for
older adults (van der Goot, Beentjes, & van Selm, 2012).
In addition to the roles of celebrity, little is known about the celebrity preferences
of older adults. Some exceptions are the studies by De Backer et al. (2007) and
Vandebosch and Eggermont (2002) in Flanders (Belgium). De Backer et al. (2007) found
a preference for local, Flemish, celebrities and for royals, soap actors, and sportspeople
among nursing home residents. Vandebosch and Eggermont (2002) also found that
older adults like royalty and they refer to the long-lasting consistency of the Royal
Family throughout (Belgian) history as a possible explanation. Van der Goots (2009)
study on older adults TV viewing habits further showed that they tend to prefer older
characters. In this chapter, celebrity preferences are studied in the context of nursing
home residents by means of preliminary individual interviews and celebrities social
roles as food for conversation, leads into moral discussions, means for meaning-
making, social companions, and aide-memoires are addressed in focus group
interviews.

Method
In research with nursing home residents, it is difficult to choose a method adapted to
older adults and their specific abilities and limitations, for instance issues with hearing,
vision, writing, energy, memory, pain, and discomfort (Barrett & Kirk, 2000; Hajjar,
1998: 21). This is a first reason why individual and focus group interviews are employed
in this study. Second, focus groups are less intrusive for nursing home residents as they
fit into their daily schedules quite easily, mimic animation activities in the nursing home,
and are more time-efficient for researchers in comparison to participant observation
(Bloor et al., 2001; Morgan, 1997). Third, the combination of more in-depth and
controllable individual interviews and interactive focus groups helps to understand the
personal preferences of residents as well as the everyday human interactions and

162
underlying group meanings, processes, and norms (Bloor et al., 2001: 4). Finally, the
selection of nursing home residents who live together in one nursing home enhances the
natural character of the focus group discussions. However, conducting individual and
focus group interviews with nursing home residents is more time- and labor-intensive
than with other groups as multiple (physical and mental) health-related factors
complicate the interviews (Barrett & Kirk, 2000). Indeed, not all nursing home residents
who appear able to participate in interviews do actually contribute. Further, conditions
inherent to old age often complicate interviews with older adults, for example when
they forget the subject of focus group discussions or do not follow actively.
This study is situated in Flanders (Belgium), where forty semi-structured
individual interviews are conducted to gain insight in the preferences of nursing home
residents with regard to celebrities and to select participants for focus group discussions
with residents who are (at least) familiar with the celebrities (cf. Bender, Bauckham, &
Norris, 1999). The participants in the individual interviews are forty residents of two
similar (in size and population) nursing homes (twenty residents/home) one rural (in
Essen) and one urban (in Antwerp) and consist of 35 women and 5 men between the
ages of 68 and 101 (M=85), which is representative of the nursing home population in
Flanders (Vanden Boer & Pauwels, 2005). Several staff members of the nursing homes
were consulted for the selection of the participants and, following their instructions on
potential participants, only residents with relatively good mental capacities were
selected. Indeed, it proved difficult to interview less healthy residents or those suffering
from dementia, because of their unwillingness or limited capacities to participate
(Claessens forthcoming). Therefore, the results below do not represent the whole range
of nursing home residents, which (unfortunately) is common practice in research on this
particular group (Hajjar, 1998: 65). In the interviews, the residents are asked about
socio-demographic variables and whether they have favorite celebrities and, if so, to
provide their names.
Four focus groups are conducted with six to seven residents and last
approximately one hour and a half. The participating residents live in the two nursing
homes described above. Only residents who are able and willing to participate in the
research are selected, resulting in a total of 27 residents (13 for the rural and 14 for the
urban nursing home) with an average age of 85,6 (ranging from 65 to 97). There is a
large overlap between the participants in the individual and focus group interviews.

163
Unfortunately, not all participants in the individual interviews were able or willing to
participate in the focus groups. With regard to gender, there is no male resident in the
first focus group (rural), one man in the second (rural) and third (urban), and two men
in the fourth (urban). The focus groups are conducted by the researcher (a young
woman), assisted by a female activities staff member of the nursing home and a female
student who takes notes and transcribes the interviews verbatim. After an introduction,
the focus group discussion is opened with a general question (what is the first thing
that comes to mind when you think about celebrities?), followed by a series of twenty
celebrity pictures. For each celebrity picture, a set of questions returns: Who is this
[celebrity]? What do you know about [celebrity]? Do you like [celebrity]? What do you
think about [the theme in the life of the celebrity]? How old were you when [celebrity]
was famous? Did/do you talk about [celebrity] with other people? These questions
relate to the nursing home residents knowledge of celebrities as well as to the various
roles of celebrity described in the theoretical part, respectively celebrities as social
companions (cf. parasocial relationships), lead into moral discussions and means of
meaning-making, aide-memoire by situating the celebrity in the life course, and food for
conversation.
The celebrities in the pictures are selected on the basis of the individual
interviews and the celebrities that are named more than once are taken into
consideration for the focus groups. The sample is composed of nine non-themed
celebrities who have not been covered in the media in the context of a moral theme
and eleven themed celebrities who obtained extensive media coverage in the context
of a moral theme. There is a variety of themes adultery, homosexuality, divorce, drugs,
child abuse, accident, and suicide with a focus on the themes of adultery and
homosexuality. Further, efforts are made to achieve a balance between male/female,
past/contemporary, local/global, royalty/film/music/TV/sports celebrities (see Table
1).

164
Table 1. Selection of celebrities
Nr. Name Gender Time Place Domain Theme
1 Prince Charles and and Both UK; Global Royalty Adultery, divorce,
Princess Diana accident
2 King Albert and and Both Belgium; Local Royalty Adultery
Queen Paola
3 Elizabeth Taylor Past US; Global Film Divorce
4 Rock Hudson Past US; Global Film Homosexuality
5 Luc Appermont and Now Flanders; Local TV/Music Homosexuality
Bart Kall
6 Walter Capiau Both Flanders;Local TV Child abuse (sexual)
7 Elvis Presley Past US; Global Music Drug abuse, divorce
8 Yasmine Now Flanders;Local Music Homosexuality, suicide
9 Will Ferdy Past Flanders;Local Music Homosexuality
10 Eddy Merckx Past Belgium;Local Sports Drugs
11 Tom Boonen Now Belgium;Local Sports Drugs
12 Prince Filip and and Now Belgium; Local Royalty /
Princess Mathilde
13 Prince William and and Now UK; Global Royalty /
Princess Kate
14 John Wayne Past US; Global Film /
15 Greta Garbo Past Sweden; Global Film /
16 Tony Corsari Past Flanders;Local TV /
17 Eddy Wally Both Flanders;Local Music /
18 Kim Clijsters Now Flanders;Local Sports /
19 Rik Coppens Past Flanders;Local Sports /
20 Roger Federer Now Switzerland; Sports /
Global

The focus group are transcribed and the transcripts are qualitatively analyzed by
means of open and axial coding (Strauss, 1987) to identify the different roles or
meanings of celebrities that transpire from the residents discussions. The segments
covering the celebrity preferences as well as each potential role of celebrity food for
conversation, leads into moral discussions and means for meaning-making, personal
involvement (parasocial relationships), and aide-memoire are analyzed separately and
the findings are presented below.

Celebrity Preferences
Of the forty participants in the individual interviews, 31 residents indicate that they
have a favorite celebrity. The celebrities mentioned by the residents are a combination
of local celebrities and legendary international celebrities, as shown in the overview of
celebrities used for the focus groups (Table 1). In the context of global versus local
celebrities, several residents in the focus groups express that they feel distanced with
regard to international celebrities and closer to the local ones, which is similar to the

165
findings by De Backer et al. (2007) who found that nursing home residents prefer the
proximity of local celebrities over the glamour of Hollywood stars.

Researcher: Were you sad when she [Princess Diana] had the accident?
[]
Ma. ( 88): It was another country
Researcher: Yes, that isnt What if it happened in Belgium?
M. ( 76): Yes
Researcher: Would it be different?
M.: Yes
Mad. ( 90): Yes
Researcher: Is that the case for everyone?
J. ( 80): Yes, yes
Ma.: Yes
Researcher: So it is different because she comes from Great Britain?
Ma.: Yes
D. ( 91): It is quite distant (FG1)

This feeling of closeness to local celebrities can be linked to the relevance of

cultural proximity in parasocial relationships and interpersonal liking (Claessens & Van
den Bulck, 2011; Tian & Hoffner, 2010) and to the local character of celebrity culture in
the past (during their youth), when globalization was not yet a dominant process in the
media landscape and society in general (Chapman, 2005).
In terms of age, the residents mostly indicate older celebrities as their favorites,
often those of their own generation, which is similar to van der Goots (2009) findings
that older adults prefer older TV characters. In a study on television in the nursing
home, Hajjar (1998) found that earlier media habits often persist in later age, which can
explain why the residents in the current study prefer celebrities from their own
generation, who they have (probably) liked since their youth. However, when the
residents refer to sports celebrities as their favorites, the younger sportspeople who
are still competing today are mentioned more often than the older ones. This may be
related to the fact that they continue to see these celebrities in the media on television,
in newspapers, or magazines and are still able to cheer for them in competitions. In the
focus group interviews, there was almost no difference in the discussions of older and
younger celebrities, indicating that, even though they do not indicate the younger ones
as their favorites as often as older celebrities, they still know quite a lot about the former
and are keen to discuss their lives.

166
 A variety of celebrities and domains is mentioned by the nursing home residents
in both the individual interviews and the focus groups music, TV, film, royals, and
sports but it is clear that the ultimate version of ascribed celebrity, royalty, is of special
importance to most residents: Kings and Queens, not Ministers because they change all
the time (J. 78 FG1). This confirms the findings of De Backer et al. (2007) and
Vandebosch and Eggermont (2002) and can be linked to the stability of royals in Belgian
history (Vandebosch & Eggermont, 2002). The royalty preference is especially strong
among the female residents in the focus groups, who state that kings, queens, princes,
and princesses are among the first things that come to mind when thinking about
celebrities. The royals that are mentioned most often are the local Belgian Royal Family,
both from the past and the present. Sportspeople the archetype of achieved celebrity
are the most-mentioned celebrities among the male participants, which confirms De
Backer et al.s (2007) findings on sports celebrities as the favorites of nursing home
residents and previously found gender-based preferences (Gantz & Wenner, 1991).
Throughout the focus groups, two men repeat that they know most about sports
celebrities and ask several times: When will we talk about sportsmen?
Interestingly, the celebrities discussed by the residents are all ascribed or
achieved and none are famous for merely being famous (attributed celebrities or
celetoids; Rojek, 2001). In addition, when asked about the first thing that comes to mind
when thinking about celebrities, one of the residents answered: What they can do, what
they are worth (L. 84 FG3), which is in line with the idea behind achieved celebrity.
Indeed, despite the multitude of attributed celebrities in contemporary society (e.g.,
reality stars), the residents do not once mention or show interest in them. This may
indicate a preference for the types of celebrities that have been there since their youth
ascribed (royalty) and achieved (sportspeople) rather than the newer types such as
attributed celebrities or celetoids (Rojek, 2001).
Most male residents state that they mainly talk about the professional aspect of
sports because you can never hurt anyone this way (L. 78 FG2), namely by focusing
on the (clean) professional rather than (messy) private life: I will discuss the positive,
he was a great sportsman (C. 77 FG4). Men thus appear to prefer to focus on
celebrities professional (and sometimes also the official private) lives, whereas women
have no hesitations to discuss all three celebrity personae: public, private, and real. This
distinction also applies to the knowledge displayed by the residents throughout the

167
focus groups. Both male and female residents have an extensive knowledge of the
celebrities in the pictures and are keen to show it. Some residents even have a quiz
mentality and are happy to provide very detailed descriptions of the celebrities lives.
For men, this is usually limited to the professional careers of sportsmen: He recently
lost the finale at the Olympics, lost to Andy Murray [] He has 17 Grand Slams, he is the
record holder of the most Grand Slams (L. 78 FG2, on Roger Federer). Women,
however, demonstrate their knowledge of all kinds of issues (professional and private)
for all sorts of celebrities, with a preference for the private and the real celebrity
personae: I think about their personality, whether they are selfish or compassionate
(L. 84 FG3).
In the focus groups, an interesting difference is found between rural and urban
nursing home residents in terms of (knowledge of) celebrity types. Whereas movie stars
are well-known among urban residents, almost no rural residents recognize Rock
Hudson and they have a hard time recognizing Elizabeth Taylor. Further, in the rural
focus groups, one resident states that we did not really talk about movies (J. 86 FG1).
This can be explained by the findings on urban versus rural nursing homes in a study on
residents media use, namely that urban residents frequently visited cinemas as children
and young adults, whereas rural residents only went sporadically as children (Claessens,
forthcoming).

Social Roles of Celebrity for Nursing Home Residents

Food for Conversation
One of the social roles that are theoretically accepted by celebrity scholars and have
been confirmed by previous research (e.g., Feasey, 2007) is the celebrity as a topic for
conversation. Indeed, celebrity can be an easy lead into social interaction, as a shared
parasocial connection. The focus group discussions themselves demonstrate that
celebrities are often used as food for conversations within the nursing home. Indeed, by
simply showing a celebrity picture broad social interaction on a variety of topics is
initiated between the residents. When explicitly asked whether they talk about a
particular celebrity with other people, most residents state that they do use it as a topic
for conversation, sometimes even a lot (M. 79 FG2; R. 92 FG3), but some residents
are rather reluctant to admit it: I dont really talk about it with other people (M. 87
FG2). Surprisingly though, even those who are reluctant about using celebrities as food

168
for conversations, continue to actually use it in the focus group setting. The
unwillingness to admit to talking about celebrity with other people may be related to the
low cultural status of celebrity magazines (Hermes, 1995) as people are expected to talk
about less trivial issues: O no, that is certainly not the first thing [we talk about] (M.
88 FG1) or We have other things to discuss (L. 84 FG3). When they talk about
celebrities with other people, they state that those are no profound conversations,
rather a fait divers (R. 92 FG3). Discussions about particular celebrities are often
triggered by large events or big news (M. 88 FG1), from sports games, over a royal
marriage or a gay outing, to a celebritys death. They state that it is the talk of the town
for a day or maybe a few days, but that it does not last longer: The next day, we did not
talk about it anymore (R. 92 FG3) or It passes rapidly (L. 85 FG4). The media play
an important role here as their news stories often trigger the residents use of celebrities
as food for conversations as they discuss celebrities in the nursing home, when we read
something about it or see it on TV (J. 78 FG1). As television and magazines are the
most important media for nursing home residents (cf. Claessens, forthcoming), it is not
surprising that one resident states that now that is has been in a magazine, everyone
knows (M. 88 FG1). Media are thus the primary link between celebrities and the
residents and play a powerful role in terms of access as the nursing home residents (and
the broader audience) depend on them to fulfill their need for information on celebrities.
In the past, the residents used celebrities as food for conversations with their
family, acquaintances, and colleagues, but also in bars and on the streets: When you
went out for a beer, Dianas death was the talk of the day (R. 92 FG3). Now that they
live in the nursing home, celebrities are easy leads into social interaction with their
family, especially their children, although they stress that this is not the first thing that
they discuss with them, rather when there are no other topics for conversation. Further,
celebrities are discussed with other residents, if they have similar interests, e.g., they
discuss sports if the people at their dining table are also interested in sports. The
residents further state that celebrities are not just good to start conversations but also
as means for entertainment, laughter, and gossip: People laughed at it (J. 86 FG1).

169
Moral Discussions and Means for Meaning-Making
In addition to their role as food for conversation, celebrities are important for nursing
home residents as an easy lead into moral discussions, which help the residents give
meaning to the world and construct their own identity. Indeed, discussing the life events
of celebrities is a safe way into wider discussions about morality, as they are shared
parasocial acquaintances, known by a lot of people, whom the residents can criticize
without any real-life consequences in contrast to friends, family, or colleagues (Feasey,
2008).
In the focus groups, a large variety of themes is discussed, both provoked by the
selection of celebrities and unprovoked, newly introduced by the residents themselves.
The issues vary from rather superficial themes such as the celebrities appearances, over
their personalities, to moral issues as provoked by the celebrity pictures such as
adultery, homosexuality, health, addiction, drug use, marriage, divorce, suicide, child
abuse, or aids. In addition to these anticipated themes, the residents introduce new
issues such as the celebrity system and the heavy toll of fame, money, and the changing
world. Interestingly, these issues are not explicitly linked to the individual celebrities
but go broader. Indeed, discussions usually start specific and small, but often lead to
general issues such as cultural or societal changes over the course of the residents lives
or long-lasting truths that remain constant throughout life, for example: there are three
things [in life]: money, sex, and God (M. 79 FG2).
The statements on the celebrities appearances usually come up at the beginning
of the discussions, when a new celebrity picture is presented:

Researcher: What do you know about Prince Charles and Diana?

I. ( 85 FG3): That Diana was a very pretty girl

One resident even replies to the opening question What is the first thing that
comes to mind when thinking about celebrities? by referring to appearances:
Someone with long hair (L. 85 FG4). The celebrities looks or appearances are
mainly discussed by the female residents, but sometimes men also offer their evaluation.
The looks of both male and female celebrities are evaluated by the residents and most
statements positively describe celebrities as pretty women, handsome men, and
beautiful couples. Some female residents admit to having (had) a crush on the handsome
men in the movies: I thought he was very handsome. I always told my husband that he

170
[Rock Hudson] would not have to propose twice (J. 78 FG1). However, sometimes the
appearances of the male (but never the female) celebrities are criticized ugly man (J.
86 FG1) or considered too ordinary An everyday face, it is an everyday face like any
other person, right? (J. 90 FG4). The only time when a female celebrity is criticized in
the context of her appearances is when she is arrogant: She was a pretty girl, but she
knew it (L. 84 FG3). Interestingly, celebrities appearances are often linked to
homosexuality in different ways for male and female celebrities. Whereas the former are
considered too polished [] to be a normal man (I. 85 FG3) or feminine (J. 78
FG1), the latters homosexuality is regretted because I always thought she was pretty
enough to get a man (J. 78 FG1).
In response to the opening question, one resident states that the first thing that
comes to mind when thinking about celebrities is how their personality is [] whether
they are selfish or compassionate, or friendly (L. 84 FG3). Most other statements on
the celebrities personalities are given in response to Do you like [celebrity]? The
celebrities personalities are usually evaluated in a positive manner, when they are
sweet, humble, and act normal. However, when the male celebrities are arrogant or
hypocritical in the case of Eddy Merckx: showing off and meanwhile he took some
[drugs] as well (M. 87 FG2) or do not act in accordance with their status in the
case of Prince Filip as he is considered to be way too soft (J. 86 FG1) and not able to
handle it [being king] (J. 97 FG2) they are criticized. The female celebrities are
criticized less but when they are, it is usually about their arrogant or diva-like
behaviour, as in the case of Yasmine: She was always the best one and she could easily
belittle others (L. 84 FG3). As stated above, several moral themes are discussed in the
focus groups, both provoked by the selection of the celebrities and newly introduced by
the residents. For the former group, the two most-discussed themes adultery and
homosexuality are discussed briefly. With regard to adultery, two royal couples are
discussed, namely Prince Charles and Princess Diana (UK) and King Albert and Queen
Paola (Belgium). The adultery is mostly evaluated critically and the residents think it is
sad, but they accept is as a part of life and something which they cannot control. Here,
acquiescence is present which can be linked to the large amount of life experiences and
the higher emotional complexity that older adults tend to master, which helps to put
things in perspective and become relativistic (Carstensen, Fung, & Charles, 2003;
Harrington, Bielby, & Bardo, 2011).

171
 The second most-discussed and also the most-taboo theme introduced by the
researcher into the focus group discussions is homosexuality which is discussed with
regard to Rock Hudson, Will Ferdy, Yasmine, Bart Kall, and Luc Appermont. Here, the
residents state that they do wish to not talk about the taboo topic of homosexuality
because it is bad enough the way it is (M. 87 FG2):

Researcher: And do you ever talk about homosexuality or homosexual

celebrities?
M. ( 87): No, no
J. ( 97): No
M.: I dont tell anyone, I dont talk about it (FG2)

They further explicitly refer to the sensitivity of discussing homosexuality

especially in the past: Taboo, and if you talked about it, you got a severe punishment
(C. 77 FG4). In the discussions, the overall tendency is to promote and support
celebrities coming out and their homosexuality is accepted by most residents, although
they regret it and consider it to be a sad illness because they were born with the wrong
hormones (I. 85 FG3).
Interestingly, the presentation of the celebrity pictures did not only evoke
discussions on the themes that are associated with the celebrities, but also new moral
themes that go beyond the context of the specific celebrity. Indeed, the wider celebrity
system is discussed by the residents. For instance, they state that royal celebrities are in
such a strict system that, in fact, they are directed by the government and their
environment (R. 93 FG4). When discussing the tragic demise of Elvis Presley, a
resident states: In fact there are a lot of tragic figures, among the famous people []
because they are glorified too much and lead lives that are too hectic (R. 93 FG4).
Another example of the toll of fame is provided when a resident links the suicide of
Yasmine to the fact that she cannot handle the fame (C. 77 FG4). Here, the celebrities
are considered the victims of the celebrity system and their behavior is excused, which
can be linked to the residents adoption of a fan attitude (cf. Van den Bulck & Van Gorp,
2011).
Further, following a discussion of doping use in sports, the residents introduce a
new moral theme, namely the change in contemporary society with a decrease of
solidarity:

172
 M. ( 79): Not much is fair in the world now
Researcher: No?
M.: No
L. ( 78): No, it disappears. It disappears with Yes, it gets worse with a lot, a lot
of percentages
M.: Yes, in the past the people were happy when they could help each other. You
did not have to give anything or something you could miss yourself, they were
happy then. But now, you can no longer make a child happy. (FG2)

Another newly introduced theme is the issue of money, which is discussed in the
focus groups in two ways. First, celebrities especially sportspeople are discussed as
being extremely rich: you do know that those American basketball players are all
millionaires, right? (R. 92 FG3). Second, celebrities are seen as victims of the
(celebrity) system, especially in the context of sports, which ultimately uses celebrities
to make as much money as possible. Money is almost always evaluated negatively as it
does not make people happy, on the contrary, it often brings out the worst: Money is so
important in soccer, that they demand more and more from those sportsmen. But it is
not their fault, it is the fault of the sports world (R. on doping in sports 93 FG4) or
They are paid by their brand, but those brands pay the sportsmen a certain amount a
year if they win several times. And if you cannot win by natural means, you look for
something else [] But the reason is that they make a lot of money and they cannot do it
naturally (L. 78 FG2).
Interestingly, throughout the moral discussions in the focus groups, it becomes
clear that most residents are more relativistic than critical in their judgments of the
celebrities, which can be linked to the multitude of life experiences and the higher
emotional skills and complexity that characterize older adults and help them to put
things in perspective (Carstensen, Fung, & Charles, 2003; Harrington, Bielby, & Bardo,
2011).
Throughout these moral discussions, the residents often refer to their own
personal experiences or their social network. For instance with regard to homosexuality,
residents state: In my building two of them are living together and they are very good
men, very friendly (L. 85 FG4) or I have a cousin and her daughter did not know
either [that she was gay] so she told her daughter there is something about you and she
is a lesbian (L. 85 FG4). Further, the use of doping in sports evokes personal stories:
They used to do that here as well [] My father had pigeons, o and [] he did it in their
food and their water (M. 79 FG2). By providing these personal stories in response to

173
celebrity pictures and the similarity between celebrity gossip and gossip about real-life
acquaintances, the residents appear to see an analogy between celebrities and (real)
social companions.

Social Companions
Celebrities are meaningful for nursing home residents in terms of social companionship,
as a personal connection, a parasocial relationship, between the residents and the
celebrity. In the focus groups, the residents refer to both positive and negative
parasocial relationships with celebrities and, interestingly, this distinction can be linked
to the ordinary-extraordinary paradox of celebrity (Dyer, 1998; Holmes, 2005).
Residents positive expressions are found with normal celebrities, who are humble,
friendly, and sweet, namely the ordinary side of the paradox. The negative personal
connections are expressed toward celebrities who are arrogant, hypocritical, dishonest,
and behave like stars (e.g., Elizabeth Taylor, Greta Garbo), namely the extraordinary
side of the paradox.

M. ( 79): I really dislike this guy [Eddy Merckx]. I think he is a ridiculous little
man
Researcher: Yes, why?
M.: Being arrogant and cocky and taking drugs at the same time. And now hes too
afraid to do it
Ma. ( 87): I still think Tom Boonen is the most sympathetic
Researcher: Yes?
Ma.: Of all those men who have been cyclists
L. ( 78): Yes, hes more a man of the people (FG2)

Most residents have favorites among the celebrities that are discussed he is my
good friend (G. 94 FG3) or he is my favorite (M. 87 FG2) especially in the
context of sports which is not surprising as, in sports, one is usually pro one particular
player and contra the opponent (cf. Theodoropoulou, 2007): I was so sad that he lost [a
tennis game] (L. 84 FG3). Although they clearly indicate their favorite celebrity, they
are reluctant to label themselves as fans: O I was not a fan (L. 84 FG3) which can be
linked to the low cultural status of fans that is still present in contemporary society
(Jenson, 1992). Some female residents state that they had a crush on a celebrity: When I
was young, I always said I would marry him (J. 78 FG1) or You were always in love
with those men [] but it always passed because it would never happen for us (D. 91
FG1).
174
 Further, the residents state that they are shocked after a favorite celebritys
death: I think about Queen Astrid. I can never forget that she died when the car hit that
tree (I. 85 FG3). This is striking as Queen Astrid died in 1935, which indicates that
this event a queens (mysterious) death has a lasting impression and importance in
the life of this particular resident. Similarly, the residents feel sad when the celebrities
quit their job, especially in the context of sports: I think it is sad that she [Kim Clijsters]
quits [playing tennis] (I. 83 FG3). Further, the residents are disappointed after
certain events, e.g., the outing of a homosexual celebrity I always liked Bart Kall the
most [] but that changed since I know [that he is gay] (J. 78 FG1), the confession of
drug use in sports Eddy Merckx is sympathetic, but the day I found out he used drugs,
I lost my respect for him. He has no willpower (C. 77 FG4) and Tom Boonen, I always
liked him, until he took the drugs (I. 85 FG3), or child abuse:

Researcher: And what do you think about Walter Capiau, is he sympathetic?

C. ( 77): I used to think so
L. ( 85): I used to think so
R. ( 93): Not anymore
L.: Before
C.: Before
Researcher: So you liked him before, but what after you heard what he did [child
abuse]?
L.: That is very sad
R.: Yes, yes
Researcher: Does that change your image of him?
L.: Yes! (FG4)

The residents are aware that their image of the celebrities depends on the media
coverage, indicating their insight into the celebrity construction (Marshall, 2006):

R. ( 92): As an outsider, you cannot decide

L. ( 84): Whether she is sympathetic, you know?
R.: Whether someone is sympathetic or not
L.: That is right
R.: You let the press guide you, right?
L.: Yes
G. ( 94): That is the way it is, we only see them on film or on And you never
really know What she is really like (FG3)

They also state that they do not really know the celebrities, but this does not hold
them back to express their personal both positive and negative feelings toward them:
Without knowing the man of course, but he looks trustworthy and sympathetic (R.

175
92 FG3) or I dont know her personally but I never liked her (L. 84 FG3). Further,
they tend to be aware of the one-sided nature of the parasocial relationship with the
celebrity: He doesnt listen to me. I look at him, but not the other way around, right (L.
85 FG4).
Interestingly, several urban residents have actually met some of the local
celebrities discussed in the focus groups, other residents lived close to the celebrities in
their youth, and one resident was even married to a famous Flemish singer/TV
presenter. Especially in the latter case, the parasocial relationship is transformed into a
social relationship, where there is actual two-sided contact. However, in the case where
the residents only met the celebrity once, the concept of parasocial relationships is still
viable because of the one-off character of the meeting. In these one-off meetings, the
residents feel like they get an unmediated image of the celebrity which is almost always
positive: well, he was very friendly (L. 84 FG3). However, the resident who was
married to a Flemish celebrity was not always positive about the Flemish celebrities she
met throughout her late husbands career. She describes some of her husbands
colleagues as very sympathetic (A. 89 FG3) but others are like a devil [] he always
acted very bad, was in fights with everyone (A. 89 FG3).

Aide-Memoire
Finally, celebrities can help the residents to reminisce and link celebrities to stages in
their life course. Throughout the focus groups, the residents state that it is rather hard to
situate the celebrities life events in their own life courses: Oops, we do not remember
that (L. 85 FG4). This can be linked to the memory issues that older adults are often
faced with (Vandebosch & Eggermont, 2002; Van der Goot, 2009) and to which the
residents refer themselves: I told you, my short-term memory (R. 92 FG3). Indeed,
recall of the specific year of an event may be a problem for the residents, but they try to
outweigh this by pointing to other elements: I dont remember the year in which they
married, but I know they have not been married for long (J. 78 FG1). Further, the
residents often ask the birth date or age of the celebrity to compare it with their own age
or that of their siblings:

Ir. ( 83): You [researcher] ask a lot about our age

L. ( 84): We cannot remember that. How old is he? (FG3)

176
 Another strategy to compensate for their limited memory is to think about their
broad life circumstances, namely in terms of being married I was already married
then (M. 88 FG1), having (young/older) children I had small children then (I. 83
FG3), and being at home or at the nursing home No, I was not here yet, I was still at
home (J. 78 FG1). Further, the media play a role in this process of reminiscence: I
remember that we had our own TV and saw it (L. 84 FG3), Wasnt it black and white
television back then? (M. 88 FG1) or I went to the cinema with my children then (J.
86 FG1).
In terms of reminiscence, some residents remember watching some of the movies
played by the celebrities: I remember her well in the role of Cleopatra (R. 92 FG3).
Most residents also remember television programs: He played in a television series []
Dynasty, I think (J. 78 FG1) or He used to be on a show where children could
perform (C. 77 FG4). For some female residents, the picture of John Wayne brought
up memories of their late husbands fandom: My husband was a fan of him. Whenever
there was a John Wayne movie, I no longer had a say in things [] We bought a second
television for John Wayne (I. 83 FG3) or My husband had a picture of him in his
office [] and then those porcelain plates of John Wayne in cowboy movies and my
husband put them on a cupboard, which irritated me because it did not fit there. I has
Louis XIV and then John Wayne, Haha (J. 78 FG1).
Only one resident has a very clear memory of where she was when something
important happened to a celebrity, more specifically, when Elvis died: I was already
married and visiting the coast with a friend when I heard that he died (J. 78 FG1). This
resident further has quite specific memories with regard to several other celebrity
events: He became king after his brother 19 years ago or she died three years ago (J.
78 FG1)
When reminiscing, the residents often use celebrities to compare the present and
past culture and society, for instance with regard to the world of sports: He [Rik
Coppens] did not look after himself. In the current state of soccer, he probably would not
have been a great talent, because he went drinking too often and did not take care of
himself (R. 93 FG4) or I always used to follow cycling [] but not anymore, because
they are no longer honest. The one who is the best at cycling should be able to win and
that is no longer the case (L. 84 FG3). One resident compares the current dominance
of the United States in the sports world to the past situation: It was the same in the past

177
with the Eastern Bloc, the Germans, the GDR. At the Olympics and world championships,
they got one victory after the other. Afterwards it came out, when it was no longer
communist and the people flew to other countries, to America. Then they all confessed
to using doping (R. 92 FG3). Further, a discussion of Elvis Presley leads to a
comparison between the status of marriage and divorce in the past and present:

Researcher: And he divorced his wife

M. ( 79): Yes, but that is nowadays warp and weft
Staff member: But back then it wasnt, right?
M.: No
L. ( 78): The Americans brought it here. We have to be honest, when we look at
the past
Researcher: Who brought it in?
L.: The movie stars, marrying multiple times. And now the families here are
almost all reconstituted families. One week at dad, one week there (FG2)

In addition, a discussion of the taboo status of homosexuality in the past it took

a lot of courage back in those years [to come out] (R. 92 FG3) leads to the statement
that having babies was also something that was not talked about in the past, you almost
had to hide it if you had a baby. But that is in the past (L. 84 FG3). Thus, by presenting
residents with a picture of a celebrity, memories of past taboos are triggered.

Conclusion
To compensate for the gaps in both gerontology and celebrity studies, this chapter aims
at providing insight into the meaning of celebrities in the lives of older adults and, more
specifically, nursing home residents by means of individual and focus group interviews.
The current body of knowledge on the celebrity preferences of nursing home residents
is scarce but provides some valuable findings preferences for royalty, sportspeople,
local and older celebrities which are mainly confirmed and, importantly, expanded by
this study. Further, the limited knowledge of the social roles of celebrities for nursing
home residents as food for conversation and social companions (De Backer et al.,
2007) is confirmed and extended in this study, both by transferring social roles of
celebrities for the general audience to the nursing home context in the case of moral
discussions and means for meaning-making and by looking for new roles specific for
the older adult and nursing home population in the case of aide-memoire.

178
 With regard to the celebrity preferences of nursing home residents, the findings
in this chapter mostly confirm and nuance previous studies. Cultural proximity is
important for the residents under study as they feel closer to local than to global
celebrities (cf. Claessens & Van den Bulck, 2011; De Backer et al., 2007). In terms of age,
the residents appear to prefer celebrities of their own generation, confirming van der
Goots (2009) findings, but this should be nuanced as the favorite sports celebrities tend
to be the younger ones who are still competing today. The preference for older
celebrities is further put in perspective as the knowledge of the residents is (at least) as
extensive with regard to the current celebrities, compared to the older ones, and they
are (at least) as keen to discuss them. In terms of celebrity types, the residents favorites
are mostly royalty () and sportspeople () which confirms the findings of De Backer et
al. (2007) and Vandebosch and Eggermont (2002) and gender-based preferences (Gantz
& Wenner, 1991). Interestingly, all favorite celebrities are ascribed or achieved and none
are attributed, which can be explained by the long-standing history of ascribed and
achieved celebrity in contrast to the recent phenomenon of attributed celebrity (e.g.,
reality stars). This can be linked to the fact that previously developed media habits and
preferences often persist over time (Hajjar, 1998). Some interesting distinctions are
found between the male and female celebrities. In addition to the preferences for royal
and sports celebrities, men and women discuss celebrities differently in terms of
personae. Whereas female residents are keen to discuss all three personae (public,
private, and real), most male residents feel more comfortable focusing on the
professional activities of the celebrities and, in some cases, their official private lives.
Celebrities fulfill a variety of social roles and meanings for the nursing home
residents, consisting of roles that are found for the general audience of celebrities food
for conversations, moral discussions and means for meaning-making, and social
companions and a new role, specific for this group aide-memoire. With regard to the
first three roles, the nursing home population thus appears to be similar to the general
audience although the specific interpretation of the roles is likely to differ whereas,
for the latter, the nursing home population is a distinct group in society, with its own
needs and wishes and where celebrities can help compensate for memory loss in old
age.

179
 The role of celebrities as food for conversations which was also found by De
Backer et al. (2007) is clearly demonstrated in this studys focus groups by residents
explicitly confirming this role and by the discussions themselves, with celebrity pictures
evoking social interactions. Some residents are not keen to admit using celebrities as
food for conversations and state that it is certainly not the first thing that they discuss
with other people. This can be explained by the view of celebrity as a low form of
culture, which is still present in contemporary society, despite the omnipresence of
celebrities in all domains of life (Hermes, 1995; Marshall, 2006). When the residents use
celebrities as food for conversations, these are not profound conversations, rather the
talk of the town for a few days which is usually linked to an media-covered event. The
latter confirms the importance of media as the primary link between celebrities and
audiences and a crucial element in the celebrity apparatus (Marshall, 2006; Rojek,
2001). Conversational partners with whom to discuss celebrities in the past differ from
those in the nursing home as the former consist of practically everyone (family, friends,
colleagues, people on the streets, in bars, ) and the latter of family and residents with
similar interests. As the social networks of nursing home residents become smaller and
their mobility decreases (Drageset, 2004; Hajjar, 1998), it is not surprising that the
availability of conversational partners becomes limited in the nursing home.
Celebrities further help nursing home residents to make meaning of the world by
evoking moral discussions on a large variety of themes. This social role is important as it
allows nursing home residents to discuss (sensitive) moral themes in the context of a
shared acquaintance the celebrity without the risk of any real-life consequences for
criticizing an acquaintance (Feasey, 2008). The themes in the focus groups vary strongly
and are both provoked by the researcher and newly introduced by the residents, both
specifically linked to a celebrity and broadly incorporating society, and both superficial
(celebritys appearances) and profound (homosexuality, adultery, ). Interestingly,
whereas the viewpoints of the residents usually differ, their moral judgments are more
often relativistic than critical. This can be related to the acquiescence that characterizes
older adults and which is linked to the large amount of life experiences and the
emotional complexity that they master (Carstensen, Fung, & Charles, 2003; Harrington,
Bielby, & Bardo, 2011). Finally, the residents often talk about their personal experiences
or those from their social network in the discussions of moral themes, which may imply
that they consider celebrities as similar to their real-life acquaintances. The latter is one

180
of the main elements of a parasocial relationship (Claessens & Van den Bulck, 2011),
which refers to another social role of celebrity for nursing home residents.
Indeed, the residents consider celebrities as a form of social companions as they
(appear to) maintain parasocial relationships with them. These relationships are either
positive (like friendship) and negative (like enmity) and this distinction is linked to the
ordinary-extraordinary paradox of celebrity (Dyer, 1998). When celebrities are
perceived as ordinary the residents have very positive feelings for them, but when the
celebrities act extraordinary they evoke negative feelings. Most residents indicate that
they have a favorite celebrity, but they refrain from labeling themselves as fans. This can
be linked to the low cultural status of fandom which appears to remain dominant in
contemporary society (Jenson, 1992). In some cases, the urban residents discuss
personal contact with a celebrity, which can change the para-social aspect of the
relationship. However, as most meetings are highly controlled and one-off, it is
questionable that the parasocial relationship is transformed into a social one in these
cases.
Finally, a new social role of celebrity is found, specifically for nursing home
residents, as celebrities help the residents to reminisce as an aide-memoire. Although it
is hard for residents to link celebrities events to a specific time in their lives, due to
memory issues inherent to old age (Barrett & Kirk, 2000), the residents compensate for
these limitations by comparing the celebrities age with their own age or that of their
siblings, think in broader life stages rather than specific years (marriage, children,
nursing home), and refer to media changes (movies, television, black/white).
Reminiscence triggered by celebrity pictures includes memories of media consumption,
mostly movies and television, and the sharing of memories of the past culture and
society which results in a comparison of past and present. This role of celebrities as an
aide-memoire, triggering memories, is important as reminiscence enhances older adults
self-esteem and helps with identity construction (Bender, Bauckham, & Norris 1999; van
der Goot, Beentjes, & van Selm 2012).
This chapter thus provides exploratory insight into the junction of two major
processes in contemporary Westerns societies (global aging and a growing celebrity
culture): the meaning of celebrities in the lives of nursing home residents. Considering
the societal relevance of this theme, this chapter is an important contribution to the
(very) limited body of knowledge on the celebrity preferences of nursing home

181
residents and the social roles of celebrities in nursing homes and wishes to be an
inspiration for future research. This study shows that celebrities fulfill similar roles for
the general audience and nursing home residents with regard to food for conversation,
means for meaning-making (moral discussions), and social companions. However,
celebrities perform a new role adapted to the characteristics of the nursing home
population as an aide-memoire to promote reminiscence despite memory issues.
To expand the academic knowledge on this topic, more empirical studies are
needed to gain a profound understanding of the meaning of celebrities in the lives of
older adults not just nursing home residents but also independently-living older
adults. Further, as only a limited set of celebrities is employed in this study, future
research would benefit from using a greater variety of celebrities. The relevance of
gender-, class-, location-, or age-related differences with regard to celebrities should also
be explored in the future. Finally, the different social roles of celebrities that are found
here for nursing home residents should be explored in more detail in future research.

182
References
Baltes, Paul B. 1997. On the Incomplete Architecture of Human Ontogeny: Selection,
Optimization, and Compensation as Foundation of Developmental Theory.
American Psychologist 52 (4): 366-380.
Barrett, Julia, and Stuart Kirk. 2000. Running Focus Groups with Elderly and Disabled
Elderly Participants. Applied Ergonomics 31: 621-29.
Bender, Mike, Paulette Bauckham, and Andrew Norris. 1999. The Therapeutic Purposes
of Reminiscence. London: Sage.
Bloor, Michael, Jane Frankland, Michelle Thomas, and Kate Robson. 2001. Focus Groups
in Social Research. London: Sage.
Butler Breese, Elizabeth. 2010. Meaning, Celebrity and the Underage Pregnancy of
Jamie Lynn Spears. Cultural Sociology 4: 337-55.
Carmichael, Carl W. 1976. Communication and Gerontology: Interfacing Disciplines.
Western Speech Communication 40 (2): 121-29.
Carstensen, Laura L, Helene H. Fung, and Susan T. Charles. 2003. Socioemotional
Selectivity Theory and the Regulation of Emotion in the Second Half of Life.
Motivation and Emotion 27 (2): 103-23.
Chapman, Jane. 2005. Comparative Media History: An Introduction. Cambridge: Polity
Press.
Claessens, Nathalie. forthcoming. Nursing Home Residents Media Use from a Life
Course Perspective. In Northern Lights: Film & Media Studies Yearbook 11.1 Age,
Generation and the Media, edited by Eli Skogerb & Gran Bolin. Bristol: Intellect
Press.
Claessens, Nathalie, and Hilde Van den Bulck. 2011. Personal Bonding with Celebrities
Through Media Reception: The PSR Scale. Paper presented at the 7th Media
Psychology Conference. Bremen.
De Backer, Charlotte J.S., Marc Nelissen, Patrick Vyncke, Johan Braeckman, and Francis T.
McAndrew. 2007. Celebrities: From Teachers to Friends. Human Nature 18 (4):
334-54.
Drageset, Jorunn. 2004. The Importance of Activities of Daily Living and Social Contact
for Loneliness: A Survey among Residents in Nursing Homes. Scandinavian
Journal of Caring Sciences 18: 65-71.
Dyer, Richard. 1987. Heavenly Bodies: Film Stars and Society. London: MacMillan.

183
Dyer, Richard. 1998. Stars. London: British Film Institute.
Evans, Jessica, and David Hesmondhalgh. 2005. Understanding Media: Inside Celebrity.
Maidenhead: Open University Press.
Feasey, Rebecca. 2008. Reading heat: The Meanings and Pleasures of Star Fashions and
Celebrity Gossip. Continuum 22 (5): 687-99.
Gantz, Walter, and Lawrence A. Wenner. 1991. Men, Women, and Sports: Audience
Experiences and Effects. Journal of Broadcasting and Electronic Media 35: 233-
43.
Giles, David. 2002. Parasocial Interaction : A Review of the Literature and a Model for
Future Research. Media Psychology 4: 279-305. doi:
10.1207/S1532785XMEP0403_04
Giles, David. 2003. Media Psychology. London : Routledge.
Hajjar, Wendy J. 1998. Television in the Nursing Home: A Case Study of the Media
Consumption Routines and Strategies of Nursing Home Residents. London:
Routledge.
Harrington, C. Lee, and Denise D. Bielby. 2010. A Life Course Perspective on Fandom.
International Journal of Cultural Studies 13 (5): 429-50.
Harrington, C. L., Bielby, D. B., & Bardo, A. R. (2011). Life course transitions and the
future of fandom. International Journal of Cultural Studies, 14(6), 567-590.
Hermes, Joke. 1995. Reading Womens Magazines: An Analysis of Everyday Media Use.
Cambridge: Polity.
Holmes, Su. 2005. Off-guard, Unkempt, Unready?: Deconstructing Contemporary
Celebrity in Heat Magazine. Journal of Media & Cultural Studies 19 (1): 21-38.
Holmes, Su, and Sean Redmond. 2010. A Journal in Celebrity Studies. Celebrity Studies 1
(1): 1-10.
Jenson, Joli. 1992. Fandom as Pathology: The Consequences of Characterization. In The
Adoring Audience: Fan Culture and Popular Media, edited by Lisa A. Lewis, 9-29.
London: Routledge.
Marshall, P. David. 2006. The Celebrity Culture Reader. London: Routledge.
Morgan, David L. 1997. Focus Groups as Qualitative Research (2nd ed). London: Sage.
Moschis, George P. 1996. Gerontographics: Life-stage Segmentation for Marketing
Strategy Development. Westport: Greenwood.
Rojek, Chris. 2001. Celebrity. London: Reaktion Books.

184
Sandvoss, Cornel. 2005. Fans. Cambridge: Polity.
Stever, Gayle S. 2011. Fan Behavior and Lifespan Development Theory: Explaining Para-
social and Social Attachment to Celebrities. Journal of Adult Development 18 (1):
1-7.
Strauss, Anselm L. 1987. Qualitative Analysis for Social Scientists. Cambridge: Cambridge
University Press.
Theodoropoulou, Vivi. 2007. The Anti-Fan within the Fan: Awe and Envy in Sport
Fandom. In Fandom: Identities and Communities in a Mediated World, edited by
Jonathan Gray, Cornel Sandvoss, and C. Lee Harrington, 316-27. New York: New
York University Press.
Tian, Qing, and Cynthia A. Hoffner. 2010. Parasocial Interaction with Liked, Neutral, and
Disliked Characters on a Popular TV Series. Mass Communication and Society 13
(3): 250-69. doi: 10.1080/15205430903296051
Tulloch, John. 1991. Approaching the Audience: The Elderly. In Remote Control:
Television, Audiences & Cultural Power, edited by Ellen Seiter, Hans Borchers,
Gabriele Kreutzner, and Eva-Maria Warth, 180-203. . London: Routledge.
Turner, Graeme. 2004. Understanding Celebrity. London: Sage.
Vanden Boer, Lut, and Koenraad Pauwels. 2005. 75-Plussers: een Profielschets.
Accessed September 17, 2012.
https://wvg.vlaanderen.be/applicaties/kenniscentrum/pdf/webbijdragen/2005
_75-plussers_profielschets.pdf.
Van den Bulck, Hilde, and Nathalie Claessens. 2013. Celebrity Suicide and the Search for
the Moral High Ground: Comparing Frames in Media and Audience Discussions of
the Death of a Flemish Celebrity. Critical Studies in Media Communication 30 (1):
69-84.
Van den Bulck, Hilde, and Jasmijn Van Gorp. 2011. Eternal Fandom: Elderly Fans, the
Media, and the Staged Divorce of a Schlager Singer. Popular Communication 9
(3): 212-26.
van der Goot, Margot J. 2009. Television Viewing in the Lives of Older Adults. Nijmegen:
Radboud University.
van der Goot, Margot J., Johannes W.J. Beentjes, and Martine van Selm. 2012. Meanings
of Television in Older Adults Lives: an Analysis of Change and Continuity in
Television Viewing. Ageing & Society 32: 147-68.

185
Vandebosch, Heidi, and Steven Eggermont. 2002. Elderly Peoples Media Use: At the
Crossroads of Personal and Societal Developments. Communications: The
European Journal of Communication Research 27: 437-55.
Whannel, Gary. 2010. News, Celebrity, and Vortextuality: a Study of the Media Coverage
of the Michael Jackson Verdict. Cultural Politics 6 (1): 65-84.

186
187
 Chapter 6
The final empirical chapter of this PhD examines one specific group of celebrity audiences that is
strongly neglected in the current body of celebrity (and media) studies, namely nursing home
residents, by bridging gerontology and celebrity studies. Indeed, building on the findings in
chapter 5, this study explores one particular social role that celebrities fulfill for nursing home
residents, namely as a lead into moral discussions. The focus groups in this study show that
celebrity goes beyond mere entertainment or commodity as it provides nursing home residents
with a forum for discussion of a variety of moral themes. The most-discussed themes are
adultery and homosexuality and a qualitative framing analysis provides insight in the
viewpoints on these topics among the nursing home residents under study. These frames are
compared to those found in chapter 2 and 3 and, although they are similar to some extent, there
are some interesting differences related to the specific characteristics of nursing home residents.

NURSING HOME RESIDENTS AND CELEBRITIES: A TALE OF MORALITY

Nathalie Claessens
(Submitted to Ageing & Society)

In contemporary Western societies, characterized by global aging and an omnipresent

celebrity culture, surprisingly little is known about the role of celebrities in the lives of
older adults. This study bridges gerontology and celebrity studies and uses this
framework to explore the specific social role of triggering moral discussions that
celebrities fulfill for nursing home residents. The latter are chosen as target group
because they have a lot of free time and restricted mobility which potentially augments
their celebrity interest. This is done by means of four focus groups with twenty-seven
residents. In particular, two themes are explored in depth in the focus groups adultery
and homosexuality and the discussions hereof are subjected to a qualitative framing
analysis. Results show that celebrities indeed trigger moral discussions both provoked
and unprovoked among the nursing home residents. Looking particularly at adultery
and homosexuality, different frames could be identified in the discussions but, in general,
the older adults were shown to retain values from their youth, often combining them
with contemporary values (cf. Westerhof, 2010). Further, the residents frames proved to
be relativistic, which can be linked to the multitude of life experiences and complexity of
emotional skills that characterize older adults.

Keywords: nursing home residents celebrity morality adultery homosexuality

focus groups

188
Strikingly, in times of global aging, many [media] researchers reflect societal attitudes
in wanting to ignore the existence of old people (Carmichael, 1976, p. 125). Looking at
the current body of research, this quote does not only apply to media and
communication studies, but also to celebrities studies, where the older adult group is
severely underrepresented, if not neglected. Although there are some studies looking at
older celebrities (Harrington & Brothers, 2008) and celebrities and older adults (e.g.,
Harrington, Bielby, & Bardo, 2011; Van den Bulck & Van Gorp, 2011), they are limited as
the former study focuses on older soap opera actors and the latter studies on the specific
group of older fans instead of considering the broader, more general older audience.
This creates a gap in both celebrity studies and gerontology. Therefore, this study
wishes to contribute to the limited body of literature on celebrities and older adults by
investigating one specific role of celebrity as a lead into moral discussions and aid for
meaning-making in the context of nursing home residents. This will be done by means
of a qualitative framing analysis as it allows direct insight into the underlying
viewpoints that structure the nursing home residents discussions.

When Older Adults Meet Celebrities

In contemporary Western societies, characterized by global aging and an increasingly
present celebrity culture, it is surprising that so little research has been conducted on
the role of celebrities in the lives of older adults. Indeed, the older adult population in
general and the nursing home population in particular are growing extensively in
contemporary Western societies (Moschis, 2003, p. 516; see also Harrington et al.,
2011). In the past, older adults were often studied as a homogenous group but the
growth and consequently the increasing variety requires a better understanding of
the different segments within the older population (Harrington et al., 2011; Moschis,
2003; Nelson & Dannefer, 1992). Older adults have further (and misleadingly) been
presented predominantly in terms of physical and mental losses, whereas aging also
entails gains (Baltes, 1997; Harrington et al., 2011; Vandebosch & Eggermont, 2002). In
this particular study, it is important to acknowledge that older adults do not only
experience physical problems and memory deficits but also have more leisure time, life
knowledge, and emotional skills. Older adults are often considered migrants in time,
living in a culture which strongly differs from the one in which they grew up. Westerhof
(2010) distinguishes several acculturation strategies with migrants in time either

189
ultimately focusing on their previous (past) culture (separation), fully adapting to the
new (present) culture (assimilation), balancing both (integration), or doing neither
(marginalization). In relation to media, Tulloch (1991) found that older soap fans engage
with soaps in terms of the values of their own generations and often underline the
clean and nice aspects, which can be situated within Westerhofs (2010) separation
strategy. This study focuses on nursing home residents, a specific subgroup within the
larger population of older adults. Hajjar (1998, p. 14) describes the typical nursing home
resident as an eighty-year-old widow with three or more chronic conditions that render
independent living impossible and who stays averagely 2.6 years in the nursing home.
This specific group of older adults has so far been largely ignored in the context of
celebrity studies but tends to have a lot of leisure time, to be restricted in terms of
mobility, and to have decreasing social networks (Drageset, 2004), all factors which may
increase their use of (celebrity) media and thus the importance of celebrities in their
lives.
Celebrity has become an omnipresent phenomenon in contemporary Western
societies, not only in media outlets, as a result of processes of tabloidization and
personalization of news (Turner, 2004), but also in peoples everyday lives, for instance
with social conversations about celebrities. Following Turner (2004), a celebrity is
defined here as a public figure whose private life becomes (at least) as important to
media and audiences as the professional activities (see Van den Bulck & Claessens,
2013a). The construction of celebrity takes place in the interaction between famous
people, entourage, media, and audience (Marshall, 2006; Rojek, 2001) which entails that
celebrities are essentially mediated phenomena as they depend on the media to reach
audiences. The meaning of celebrities for audiences goes beyond mere consumer
capitalism or entertainment as they provide media and society with a forum for
discussion of a wide range of social and moral topics. This theoretical role of celebrities
is widely accepted by scholars in celebrity studies: celebrities are symbols by which we
narrate, negotiate, and interpret our collective experience, and establish moral
boundaries (Butler Breese, 2010, p. 337) and the lives of celebrities provide parables
of instruction for fans and general audiences (Rojek, 2007, p. 73; see also Cashmore,
2006; Evans & Hesmondalgh, 2005; Gorin & Dubied, 2011; Marshall, 2006). Indeed, by
reading about and discussing celebrities lives, audiences are believed to make sense of
their own identity and the world around them. Yet, with the exception of some case

190
studies (e.g., Butler Breese, 2010; Van den Bulck & Claessens, 2013b), there is little
empirical work on the role of celebrities in the everyday lives of general audience
members and even less in the lives of older adults. While the large body of fan studies
can provide some valuable insights, it is not straightforwardly useful here as fan
research focuses on one particular group of audience members, namely those with an
intense investment in celebrities. In addition and with some exceptions (Harrington et
al., 2011; Van den Bulck & Van Gorp, 2011), fan studies have neglected the older adult
group, limiting its use for the current study. Looking at research into older adults and
celebrities, even less studies have been conducted on the particular group of older
adults under study here: nursing home residents.
This study thus wishes to help fill the current gaps in both gerontology and
celebrity studies the lack of research into the meaning of celebrity in the former and on
older audience members in the latter by bridging the two bodies of literature and use
this framework to analyze the role of celebrities in the lives of older adults. Previously,
De Backer, Nelissen, Vyncke, Braeckman, and McAndrew (2007) found that celebrities
are important for nursing home residents, either as food for gossip strengthening
social relationships or as replacements for lost real-life social contacts. In this study,
we wish to explore the extent to which celebrities are a source of meaning-making and
an easy lead into discussions of moral issues for nursing home residents.

A Tale of Two Themes: Adultery and Homosexuality

To gain in-depth insight in the role of celebrities as triggers of moral discussions among
nursing home residents, this study focuses on two themes adultery and homosexuality
and analyzes these more profoundly by means of a qualitative framing analysis. The
selection of these themes is motivated by two main reasons. First, the legal and/or
public stance toward both topics has evolved quite dramatically in the past century and
thus throughout the nursing home residents life courses, from a criminal act to a private
matter in the case of adultery and from a taboo topic to a, to some extent, accepted
element of society in the case of homosexuality (Halman et al., 2008; Stevens & Hooghe,
2003). This allows for a valuable insight into the evolution of viewpoints in the context
of Westerhofs (2010) theory of acculturation strategies for older adults as migrants in
time. In the second half of the 1960s, the sexual revolution began spreading in Western
countries, but in Belgium (where this study is conducted) there were no legal changes

191
until 1972 with regard to public morality and sexual behavior (Stevens & Hooghe,
2003). For homosexuality, legal sexual liberalization only started in 1985 when
discriminative laws (e.g., on homosexual seduction) were removed and, for adultery, in
1987 when it was no longer considered a criminal act but a private matter (Stevens &
Hooghe, 2003). With regard to public opinion, Halman et al. (2008, p. 235; 244) found
that, for adultery, there has been a rather stable percentage of Belgians considering
adultery justifiable since 1981 (46-49%), whereas, for homosexuality, the percentage
has increased from 48% since 1981 to 74% in 2000. The combination of these two
themes in the focus groups thus allows for comparison of a relatively stable theme in
terms of public acceptance and one that has changed dramatically.
A second reason for selecting these two themes is that both can be addressed
from a multitude of viewpoints which relate to religion, society, the human psyche,
biology, amongst others. Further, previous research has shown that the themes of
adultery (Gamson, 2001; Starn, 2011; Van den Bulck & Claessens, 2013c) and
homosexuality (Bailey, 2011; Brady, 2011; Dow, 2008; Van den Bulck & Claessens,
2013b) in relationship to celebrities both evoke vivid discussions. These studies can
serve as a wider framework in which to place this particular study. In his study on three
celebrity sex scandals, Gamson (2001) found three media frames that are employed to
make sense of the celebrities adultery: hypocrisy, recklessness and amorality, and
disloyalty. Interestingly, all these frames consider the adultery as a sin but link this to
institutional decay rather than personal inauthenticity, hereby blaming the morally
bankrupt environment rather than the celebrity. Further, a qualitative study on online
celebrity news and accompanying audience reactions in the context of celebrity sex
revealed strong discussions of adultery that could be brought back to seven views (or
frames) (Van den Bulck & Claessens, 2013c): adultery as addiction, as a result of flawed
personality, as a sin, as a societal phenomenon (i.e., moral decay), as part of human
nature (monogamy as unnatural/mistakes are human), as the ultimate sign of
masculinity, and as the womens fault (wife and/or mistress). In addition to the seven
frames, several online readers state that the adulterous sex lives of celebrities are none
of our business and should not be discussed further.

192
 In the context of homosexuality, Dows (2008) study of Ellen DeGeneres and
Bradys (2011) study of Adam Lambert show that a celebritys coming out is almost
invariably applauded by critics and audiences because of its political potential. However,
this potential is hardly ever realized as the outings remain within a frame of
heteronormativity that is still dominant in contemporary Western society. Further, in
Baileys (2011) analysis of the homophobic Greys Anatomy incident, she found that the
homophobia of actor Washington was harshly criticized and the coming out of actor
Knight was evaluated very positively, in terms of authenticity and personal honesty.
However, the homophobic incident and coming out were both framed in confessional-
therapeutic terms, focusing on the personal rather than the political or societal. Indeed,
Knights coming out is described as finding an ideal balance between telling the truth
about himself (authenticity) and remaining personal and apolitical to avoid challenging
dominant heteronormative cultural norms: gays are visible but not too much so
(Bailey, 2011, p. 13). Finally, in a study on the media coverage and online audience
reactions on the suicide of a Flemish gay icon (Van den Bulck & Claessens, 2013b),
several frames are identified to discuss homosexuality. The media discussed the
homosexuality within a gay icon frame, focusing on the positive role of the celebrity as a
role model for young gays. Several audience members adopt this positive media frame
but most of them develop counter frames, such as the lesbian exception and gay
rejection frames. In the former, the acceptance of gay people in society is questioned,
which is often linked to the problematic situation of and the higher-than-average suicide
rates among lesbians. This frame does not judge homosexuality, but focuses on the
difficult situation of these people in contemporary society. The latter, gay rejection,
frame condemns homosexuality as such and considers it unacceptable and unnatural
(Van den Bulck & Claessens, 2013b; cf. homophobia e.g., Bailey, 2011; Brady, 2011).
The in-depth analysis of these two themes will help to gain insight in the social
roles of celebrities in the context of moral discussions among nursing home residents.
Therefore, the discussions of the nursing home residents in the four focus group
interviews a method that was proven to be suitable and valuable for this type of study
with this type of participants (Barrett & Kirk, 2000) will be subjected to a qualitative
framing analysis to capture the underlying viewpoints of these moral themes.

193
Method
One difficulty in researching nursing home residents is the choice of method as it should
fit the unique abilities and limitations of the population issues with hearing, speech,
attention, vision, writing, energy, memory, pain, and discomfort (Barrett & Kirk, 2000;
Hajjar, 1998, p. 21). This is one of the reasons why this study opts for focus group
interviews as they provide a multitude of interactions in a limited time frame (time-
efficiency) and are less intrusive for the nursing home and the residents than, for
instance, participant observation. This enhances the access to the field and the
cooperation of the nursing homes (Bloor, Frankland, Thomas, & Robson, 2001; Morgan,
1997). Further, the more in-depth and controllable method of individual interviews does
not prove adequate here as it is the interactive character of focus groups which enables
us to understand everyday human interactions (Morgan, 1997) in the context of
celebrities, namely how they evoke discussions on moral issues in the nursing home.
Indeed, focus groups help reveal underlying group meanings, processes, and norms
(Bloor et al., 2001, p. 4). Finally, the selection of residents living together in a nursing
home strengthens the natural character of the focus group discussions. However,
conducting focus group interviews with nursing home residents is more time- and labor-
intensive than with other participants as there are multiple complicating factors
(Barrett & Kirk, 2000): not all nursing home residents who appear able to participate
actually contribute and participants often suffer from memory- or attention-related
conditions inherent to old age (Barrett & Kirk, 2000).
For this study, four focus groups are conducted in a private meeting room in the
nursing home with six to seven residents and lasted approximately one hour and a half
to ensure a familiar setting, a manageable amount of participants, and a limited time
frame to avoid fatigue (Barrett & Kirk, 2000). The focus group sessions are fitted into the
activities schedule at the nursing homes in order to avoid disturbance of the residents
daily schedules. The participating residents live in two nursing homes one in a rural
(Essen) and one in an urban area (Antwerp) that are similar in terms of size and
population.
Only residents who were able and willing to participate in the research are
selected, resulting in a total of 27 residents (13 for the rural and 14 for the urban
nursing home) with an average age of 85,6 (ranging from 65 to 97). The residents are
mainly female with no male resident in the first focus group (rural), one man in the

194
second (rural) and third (urban) and two men in the fourth (urban). After consulting the
nursing home staff about potential participants, only the residents with relatively good
mental capacities are selected. Consequently, the results below do not represent the
whole range of nursing home residents, which (unfortunately) is often the case in
research into nursing home residents (Hajjar, 1998, p. 65).
The focus groups are conducted by the researcher (a young woman), assisted by
an activities staff member of the nursing home and a student taking notes and
transcribing the interviews verbatim afterwards. After introducing the research and
researcher, the residents were asked to introduce themselves (name and age). The focus
group discussion is opened with a general question (what is the first thing that comes
to mind when you think about celebrities?), followed by a series of twenty celebrity
pictures. For each picture, a series of questions returns: Who is this [celebrity]? What
do you know about [celebrity]? Do you like [celebrity]? What do you think about [the
theme in the life of the celebrity]? How old were you when [celebrity] was famous?
Did/do you discuss [celebrity] with other people?
The celebrities in the pictures are selected on the basis of preliminary semi-
structured interviews in which the residents could indicate their favorite celebrities. The
celebrities that were named more than once were taken into consideration for the focus
groups. The sample is composed of nine non-themed celebrities who have not been
covered in the media in the context of a moral theme and eleven themed celebrities
who obtained extensive media coverage in the context of a moral theme. These themes
are not limited to the two main topics focused on in this study adultery and
homosexuality but comprise additional issues (such as divorce, drugs, child abuse,
accident, and suicide). Further, efforts were made to achieve a balance between
male/female, past/contemporary, local/global, royalty/film/music/TV/sports
celebrities (see Table 1).

195
Table 1. Selection of celebrities
Nr. Name Gender Time Place Domain Theme
1 Prince Charles and and Both UK; Global Royalty Adultery, divorce,
Princess Diana accident
2 King Albert and and Both Belgium; Local Royalty Adultery
Queen Paola
3 Elizabeth Taylor Past US; Global Film Divorce
4 Rock Hudson Past US; Global Film Homosexuality
5 Luc Appermont and Now Flanders; Local TV/Music Homosexuality
Bart Kall
6 Walter Capiau Both Flanders;Local TV Child abuse (sexual)
7 Elvis Presley Past US; Global Music Drug abuse, divorce
8 Yasmine Now Flanders;Local Music Homosexuality, suicide
9 Will Ferdy Past Flanders;Local Music Homosexuality
10 Eddy Merckx Past Belgium;Local Sports Drugs
11 Tom Boonen Now Belgium;Local Sports Drugs
12 Prince Filip and and Now Belgium; Local Royalty /
Princess Mathilde
13 Prince William and and Now UK; Global Royalty /
Princess Kate
14 John Wayne Past US; Global Film /
15 Greta Garbo Past Sweden; Global Film /
16 Tony Corsari Past Flanders;Local TV /
17 Eddy Wally Both Flanders;Local Music /
18 Kim Clijsters Now Flanders;Local Sports /
19 Rik Coppens Past Flanders;Local Sports /
20 Roger Federer Now Switzerland; Sports /
Global

The focus group were transcribed and the transcripts are qualitatively analyzed
by means of open and axial coding (Strauss, 1987) to enable the selection of fragments
in which the celebrity pictures evoked discussions of moral issues. A variety of issues
are discussed in these fragments, but as expected (and provoked), the majority of
comments is on the topics of adultery and homosexuality. As stated above, these themes
are analyzed more profoundly by means of a qualitative framing analysis. Although
framing analysis is a common research tool in content analyses in the context of political
communication, it is rather rare in this type of research (in contrast to, for instance,
discourse analysis). However, this tool is selected here as it provides direct insight into
the essential aspects of moral themes, namely problem, cause, moral judgment, and
solution (Entman, 1993). Further, as Pan and Kosicki (1993, p. 56) point out, we all
actively classify, organize, and interpret our life experiences to make sense of them. By
using framing analysis to investigate the focus group discussions of adultery and
homosexuality in this study, the residents underlying viewpoints are revealed which
enhances our understanding of their use of frames as schemata of interpretation to
make sense of events and moral issues (Goffman, 1974, p. 21). Further, the use of

196
framing analysis allows us to place our findings within a larger body of studies on
audience frames (e.g., Gamson, 1992; Huang, 2009; Van den Bulck & Claessens, 2013b,
2013c).

Celebrities and Moral Discussions among Nursing Home Residents

The core of this article focuses on how celebrities views and actions lead to discussions
of moral issues among nursing home residents. In the focus groups, a variety of moral
issues both provoked by the selection of celebrities and unprovoked or introduced by
the residents are discussed in response to the celebrity pictures. The themes vary from
more superficial issues such as the celebrities appearances, over their personalities, to
moral issues as provoked by the celebrity pictures such as adultery, homosexuality,
health, addiction, drug use, marriage, divorce, suicide, child abuse, or aids. In addition to
these provoked themes, the residents also discuss new, unprovoked issues such as the
celebrity system and the heavy toll of fame, money, charity, and the changing world,
society, and culture. Interestingly, these issues are not limited to those explicitly linked
to the individual celebrities Indeed, discussions usually start specific and small, but often
lead to broad, general issues such as cultural or societal changes over the course of the
residents lives or long-lasting truths that remain constant throughout life, for example:
there are three things [in life]: money, sex, and God (M. 79 FG2). As anticipated by
the selection of celebrities, the focus group discussions concentrate on the themes of
adultery and homosexuality, the qualitative framing analysis of which is discussed
below.

Adultery
In the focus groups, the residents address the topic of adultery with regard to late
British Princess Lady Di (1961-1997) and current British Prince Charles (1948) and the
Belgian King Albert (1934) and Queen Paola (1937). In these cases, both the man and
woman are known to have had affairs. Some residents take a firm none of our business
stance: No, I dont discuss their private lives (C. 77 FG4), We are not the ones to
judge (J. 90 FG4) or And you can judge that? Everyone has their own life and we
cannot judge that (A. 91 FG4). However, most residents are eager to discuss the
adultery and the discussions revealed five frames.

197
 First, the residents frame the adultery as the result of a flawed personality (cf.
Van den Bulck & Claessens, 2013c). They blame the cheaters horrible personality and
criticize the adultery: Yes, he [Prince Charles] was not a good person, he always used to
go out with the other woman (J. 78 FG2) or He is the type of man who is just no
good (L. 78 FG2). Second, adultery is considered to be a royal phenomenon, placing
adultery in the context of moral decay in royal circles (cf. Gamson, 2001). The residents
relate the loss of sexual norms to the large amount of free time that royals are believed
to have: They dont have anything to do, to keep busy (L. 92 FG3). Third, several
residents consider adultery to be the (wo)mens fault (Van den Bulck & Claessens,
2013c). One female resident blames the mistress for seducing the adulterer: That
Camilla Parker-Bowles must have had something to do with it (J. 78 FG1). One male
resident refers to the wife and a bad marriage as the cause of the adultery: This proves
that he was not happy in the beginning of his marriage or When you are fighting daily,
you try to find someone else to forget that (both R. 92 FG3). The womens fault frame
is transformed by one male and several female residents into a mens fault frame: Well,
if you have a bad husband, yes, then you go and find someone else, of course. But if you
are a good couple, you dont have any reason to start an affair (M. 79 FG2) or She
[Diana] had other men because he [Charles] never gave her any attention (L. 84 FG3).
The female residents thus appear to empathize with the female celebrities as adulterers
rather than with the men. The only male resident using this type of frame expresses both
versions (mens and womens fault) which may imply that he empathizes with all
adulterers, male or female, or that he adapts himself to the female majority in the focus
groups and therefore also uses the mens fault frame. Fourth, one resident, a deeply
religious Mormon woman (J. 78 FG1), discusses adultery as a sin, from a religious
point of view (cf. Gamson, 2001; Van den Bulck & Claessens, 2013c): I dont think that it
[adultery] is normal [] Once you are married, you stay married. This frame can be
positioned within Westerhofs (2010) separation strategy in which the focus lies on the
values of the past (i.e., adultery as a criminal act (Stevens & Hooghe, 2003)). Considering
the strong stance of the Roman Catholic church toward adultery and the widespread
nature of Roman Catholicism in Flanders (Billiet & Dobbelaere, 1976) especially in the
past and thus the residents youth it is interesting that the other residents never frame
adultery as a sin. They thus appear to adhere to more modern viewpoints with regard to
adultery. In Westerhofs (2010) terms, the residents can be seen to assimilate and focus

198
on the present culture rather than the past. Another explanation here may be that the
residents have acquired a lot of life experience and emotional skills throughout their
lives (Carstensen, Fung, & Charles, 2003; Harrington et al., 2011) which helps them put
things in perspective and react less strongly toward issues such as adultery. Fifth, six
female residents discuss adultery within an acquiescence frame not found in previous
research and consider the adultery to be an unfortunate event, but nothing to worry
about because you cannot do anything about it (J. 86 FG1) and they will have to
accept it anyway (M. 90 FG1). This frame can be linked to the higher emotional
complexity and skills that older adults are believed to master and that imply a larger
focus on positive emotions as a result of a shortened time perspective (Carstensen et al.,
2003).
Interestingly, throughout the discussion of adultery the residents often compare
themselves (normal people) to celebrities. On the one hand, they state that celebrities
are not normal people as there are different value systems by which to judge
celebrities and normal people and as celebrities lives are under greater scrutiny: For
royalty, this [adultery] easily turns into a scene (J. 86 FG1). They also mention that
celebrities life events, in this case adultery, are used as food for gossip and a source of
entertainment: We tend to talk more about them (D. 89 FG1) or We have a laugh
about it (J. 86 FG1). This can be linked to Gamsons (2001) analysis of the Hugh Grant
sex scandal in which celebrities are considered to differ from regular people as the
former have the difficult task of keeping their public image in line with their private life.
The residents also position themselves in a separate life world as the royals have plenty
of time (J. 86 FG1) but are severely restricted by their royal status: He does not
make it [his illegitimate daughter] official because those people [royals] are so
restricted (R. 92 FG4). On the other hand, the residents consider celebrities to be
normal people because they go through the same life events and experience similar
difficulties: I think it is sad that things like this happen to them as well. Also among the
higher people, you know? (D. 91 FG1) or No, I believe that when you get married,
you stay married. Whether you are royalty or not. This distinction between and
combination of distance/closeness or distinction/similarity between celebrities and
normal people fits the ordinary-extraordinary paradox that is often discussed in
celebrity studies (Dyer, 1998; Holmes, 2005). This paradox describes the tension
between, on the one hand, the celebrity as a human being just like the rest of us, and, on

199
the other hand, the celebrity as especially talented and/or as living in wealth and luxury,
free from (material) problems. Interestingly, the latter part of the paradox is used quite
differently by the residents. Rather than focusing on the wealth and talent of the
celebrities, they discuss the different life worlds, value systems, and scrutiny of
celebrities. For the residents in this study, the ordinary-extraordinary paradox also
entails a paradox in their stance toward the celebrity: on the one hand, they feel
empathy because of the perceived similarity with the celebrities (I think it is sad that
things like this happen to them as well (D. 91 FG1)), while on the other hand, they use
the celebrity stories as food for gossip and entertainment, creating an emotional
distance (We have a laugh about it (J. 86 FG1)). Surprisingly, the moral high ground
often found to be taken by respondents in previous studies on moral discussions
following celebrity news (Butler Breese, 2010; Lull & Hinerman, 1997; Van den Bulck &
Claessens, 2013b, 2013c) is not very dominant here: the residents try to understand the
celebritys actions rather than condemning them. This can once again be linked to the
residents high amount of life experience and emotional skills, which help them to put
events in perspective (Harrington et al., 2011).

Homosexuality
The second theme that is discussed profoundly in the focus groups is homosexuality.
The residents discuss this topic with regard to five celebrities, one global (Rock Hudson)
and four local stars (Bart Kall, Luc Appermont, Yasmine, and Will Ferdy). Rock Hudson
(1925-1985) was an American film and TV actor and one of the first celebrities to (be
known to) have died of AIDS. He kept his homosexuality secret until his death, despite
rumors during his life (DeCrescenzo, 1997). Bart Kall (1960, Flemish singer) and Luc
Appermont (1949, Flemish TV presenter) did not come out as gay until 2010 (despite
earlier rumors), when they publicly announced their long-time relationship. Flemish
singer and TV presenter Yasmine (1972-2009) was one of the main gay icons in
Flanders and served as a role model for young lesbians until her death by suicide in
2009 which media reports linked to her divorce. Will Ferdy (1927), a Flemish singer,
was the first Flemish celebrity to come out and publicly claim his homosexuality.
Interestingly, in all focus groups the residents talk about the high taboo-status of
homosexuality, especially in the past (cf. Halman et al., 2008; Stevens & Hooghe, 2003):

200
 M. ( 87): Yes, now they are coming out. In the past, they did not dare to say it
Ma. ( 79): No
L. ( 78): Yes
M.: You did not know back then because people did not talk about it
Ma.: No, nobody talked about it in the past
L.: It started in the 1960s
M.: And now, they just come out and say it
L.: Yes
M.: That they are it
L.: Yes, it started in the 1960s. Flower power (FG2)
or
Researcher: Didnt people discuss it?
R. ( 92): No, it wasnt
J. ( 90): Oh, no in those days
C. ( 77): Taboo. And if you talked about it, you got a severe punishment (FG4)

The residents further discuss homosexuality by means of three homosexuality-

related frames, next to some expressing a none of our business view, the latter
implying that they do not feel comfortable talking about the celebrities homosexuality
as they cannot interfere with that (L. 78 FG2). However, only two residents
expressed this view, while the others are eager to discuss the homosexuality of
celebrities:

D. ( 91): We do not need to know all that, right?

J. ( 78): But we are all curious, because we want to know. Haha (FG1)

Three frames transpire. First, two female nursing home residents discuss
homosexuality within a homophobic gay rejection frame (cf. Bailey, 2011; Brady, 2011;
Van den Bulck & Claessens, 2013b). This strongly negative frame sees gay people as
unnatural and not normal, right? (J. 86 FG1) as men are created to be with a
woman (J. 78 FG1). These two residents find it disgusting (J. 86 FG1) or sad,
because she was pretty enough to get a man (J. 78 FG1). They can be found to adopt
the separation strategy (Westerhof, 2010) as they adhere to the dominant values of the
past when homosexuality was taboo and wrong (Halman et al., 2008; Stevens &
Hooghe, 2003). Second, some residents employ the gay acceptance frame to discuss
homosexuality (cf. Bailey, 2011; Brady, 2011). Here, homosexual people are considered
to be the same as heterosexual people and respect is central: They are people like us
(L. 85 FG4) and Everyone has their taste and we should respect that (L. 84 FG3).
They greatly support celebrities coming out of the closet and state that they are happy
and that is the most important thing (L. 84 FG3). The residents using the gay
201
acceptance frame can be said to use the integration strategy as described by Westerhof
(2010) in which they combine the dominant values of their past (homosexuality as
taboo, negative) with the dominant values of the present (homosexuality as accepted).
Third, several nursing home residents frame homosexuality as a gay illness not found
in previous research in which it is considered a disease determined by birth. Even
though they see it as sad and regrettable, these residents accept homosexuality because
it is a done deal and cannot be fixed:

I. ( 85): They always told me that they [gay people] cannot help it
L. ( 84): No
Ir. ( 83): They are born that way
I.: Because they were born with the wrong hormones
L.: That is possible
Ir.: You are born that way, right?
I.: Yes
Ir.: And, I heard that you should not work against it [homosexuality], because that
is very, very dangerous, they say, to work against it (FG3)

The residents often refer to their own relief of having escaped the gay illness,
both for themselves and their children: Yes, I think it is sad to be like that [homosexual].
I am happy that none of my children are like that (D. 91 FG1). This frame can once
again be considered part of an integration strategy (Westerhof, 2010) in which the
residents combine the values of the past (homosexuality as wrong) with those of the
present (homosexuality as accepted).
Interestingly, throughout the discussions it becomes clear that the residents
always support the coming out of homosexual celebrities as honesty prevails over all
other (even more homophobic) values: I think it is nice that he came out (J. 78 FG1).
However, homosexual people should not draw too much attention to their sexuality:
Thousands of people are normal couples, as woman and man, and people dont go and
spread that news everywhere. Why would they [gay people]? If they are a couple, both
women or men, why do they need to make it so public? (R. 93 FG4) or I dont like the
way they ride on the streets these days [Gay Parade], I think that is vulgar (L. 85
FG4). Here, the integration strategy (Westerhof, 2010) is once again visible: the
residents adapt to the dominant values of contemporary culture (promote coming out)
to some extent, combining it with the dominant values from the past (taboo status of
homosexuality: not too much attention). However, in this combination of acceptance and
taboo a double standard is visible in which people accept homosexuality but do not wish
202
to be confronted with it in their everyday lives: gays are visible but not too much so
(Bailey, 2011, p. 13). This can be linked to Baileys (2011) findings that homosexuality
and coming out is accepted as long as it is confined to the boundaries of the personal
sphere and remains apolitical, not challenging dominant cultural norms, namely
heteronormativity.
The residents often refer to (gay) people they know in real-life to discuss
homosexuality. Interestingly, all references to real-life gay acquaintances are positive,
even for the residents using the gay rejection frame to discuss homosexuality in general.
For example, L. ( 78 FG2) talks about his mate who has that [homosexuality] too. But
he has been in the same relationship for twenty years, with a guy from the city. A good
guy, because he saved C. from lying on the streets. The residents further describe real-
life gay acquaintances as sweet, friendly, helpful, and especially nice to their mothers.
These links to personal experiences or to people in their social networks were also
found in Tullochs (1991) study on older soap fans and he describes it as a personal
memory emphasis (see also Van den Bulck & Claessens, 2013b, 2013c). The difference
between the residents descriptions of real-life acquaintances (positive) and
homosexuality in general (neutral to negative) is striking, which stresses the importance
of celebrity role models as a way of familiarizing people who do not normally meet gay
people with the issue of homosexuality (cf. Schiappa, Gregg, & Hewes, 2005).

Discussion and Conclusion

By bridging gerontology and celebrity studies, this study confirms the theoretically
assumed social role of celebrities for nursing home residents as an easy lead into
discussions of moral themes and as an aid for meaning-making. By simply showing
pictures of celebrities, the residents start discussing a variety of topics, both provoked
by the researcher and newly introduced by the residents, going from the celebrities
appearances or general personality, over the specific events in those celebrities lives, to
broader issues of morality or topics of general societal interest. Although multiple issues
are discussed in the focus groups, the focus is on two topics, as intended by the celebrity
selection, adultery and homosexuality which are subjected to a qualitative framing
analysis to understand the underlying viewpoints. A variety of frames are employed by
the residents to discuss these two themes. In both cases, a few residents adopt a none of

203
our business stance, but this is rare in the focus groups as most residents are eager to
talk about the celebrities and themes.
The residents discuss adultery as the result of a flawed personality, a royal
phenomenon, the (wo)mens fault, a sin, or within a frame of acquiescence. Some of
these findings are similar to previous research (Gamson, 2001; Van den Bulck &
Claessens, 2013c). The flawed personality, sin, and womens fault frames were also
found among audience members in the study by Van den Bulck and Claessens (2013c)
which indicates a similarity of viewpoints between the residents and the (presumably
younger) online readership of celebrity news. Further, Gamson (2011) also found the sin
frame in his study and the royal phenomenon frame is similar to the institutional frames
used in the three sex scandals, which point to the moral decay of the environment rather
than the adulterer himself as causing the adultery. The more apologetic addiction and
part of human nature frames and the positive sign of masculinity frame found by Van
den Bulck and Claessens (2013c) are not mentioned by the residents. This may indicate
that older adults tend to relate to adultery in terms of the values of their own generation
rather than the present, namely adopt the separation strategy (Westerhof, 2010). This
confirms Tullochs (1991) findings that older adults continue to adhere to the values of
their own generation, strongly influenced by Roman Catholicism, in which adultery is
considered to be non-acceptable. However, as the religious sin frame is only used by one
resident, the influence of the religious condemnation of adultery should not be
overestimated. Indeed, six residents introduce the more relativistic acquiescence frame,
not found in previous research, which can be explained by referring to older adults high
emotional skills and life experiences, through which they have learned to put events in
perspective (Carstensen et al., 2003; Harrington et al., 2011). In addition, considering
how public opinion toward adultery remained relatively stable since 1981, no real
acculturation strategies can be found here, with the exception of the sin frame as a
separation strategy.
For homosexuality, three frames are used by the residents, two of which were
found in previous research. First, the homophobic gay rejection frame was present in the
studies by Bailey (2011), Brady (2011), and Van den Bulck and Claessens (2013b). As
this frame is only used by one resident, it should not be considered dominant among the
participants. Second, the less negative gay acceptance frame (equating homo- and
heterosexuality without celebrating the former) was also found by Bailey (2011) and

204
Brady (2011). Finally, the residents introduced the gay illness frame (homosexuality as a
disease). With regard to homosexuality, the residents thus mainly use the integration
strategy, combining the dominant values of their own generation (homosexuality as
taboo, wrong) with contemporary values (homosexuality as accepted). The residents
frames indicate the persisting heteronormativity in society, as they continue to
distinguish between us (heterosexuals) and them (homosexuals). Indeed, although
homosexuality is increasingly accepted in contemporary Western society, there remains
a distance. In addition, as long as most celebrities frame their coming out in the personal
and apolitical sphere, the dominant cultural heteronormativity will not be challenged
(Bailey, 2011; Brady, 2011). Indeed, in contrast to the study by Van den Bulck and
Claessens (2013b), the gay icon frame is never mentioned by the residents, indicating
that the image of homosexuality as positive deserving an icon may not fit their value
system. This can be linked to Westerhofs (2010) separation strategy in which the
residents focus on the past (cf. the taboo status of and negative stance toward
homosexuality). Rather than considering gay celebrities as role models, the nursing
home residents use them as a lead into more general discussions about homosexuality.
Further comparing the discussions of the two themes, an interesting distinction is
found. In the context of adultery, the ordinary-extraordinary paradox (Dyer, 1998) is
clearly present in the discussions, with residents distinguishing themselves from
celebrities in terms of life worlds, scrutiny, and value systems while, at the same time,
discussing the similarity with celebrities, as they experience the same difficulties in life.
This paradox translates in another paradox of distance celebrities as food for gossip or
entertainment and empathy for the celebrities as they are just human underneath the
faade. This comparison of celebrities and normal people is never mentioned in the
context of homosexuality. In the latter, however, there are references to real-life
acquaintances (cf. Tullochs (1991) personal memory emphasis), which are never
present in the adultery discussions. Interestingly, the residents are more positive in the
discussion of real-life gay acquaintances and the specific celebrities than in the
discussion of homosexuality in general.
By means of a combined framework of gerontology and celebrity studies, this
study shows that celebrities play an important social role as they help the nursing home
residents and by extension, all audience members to construct their identities and
views of the social world by providing the means to discuss moral issues. Further, the

205
findings show that the nursing home residents discussions of the two moral themes are
quite different, for instance in terms of acculturation strategies, and provide insight into
various elements of the meaning of celebrities, for instance the ordinary-extraordinary
paradox for adultery and the personal memory emphasis for homosexuality. The role of
celebrity for nursing home residents in terms of moral themes is thus more complex and
should be investigated further with regard to a variety of moral topics to gain a better
understanding.

206
References
Bailey, C. W. (2011). Coming out as homophobic: Isaiah Washington and the Greys
Anatomy scandal. Communication and Critical/Cultural Studies, 8(1), 1-21.
Baltes, P. B. (1997). On the incomplete architecture of human ontogeny: Selection,
optimization, and compensation as foundation of developmental theory.
American Psychologist, 52(4), 366-380.
Barrett, J., & Kirk, S. (2000). Running focus groups with elderly and disabled elderly
participants. Applied Ergonomics, 31, 621-629.
Billiet, J., & Dobbelaere, K. (1976). Godsdienst in Vlaanderen. Van kerks katholicisme naar
sociaal-kulturele kristenheid [Religion in Flanders. From churchly Catholicism to
socio-cultural Christianity]. Leuven: Davidsfonds.
Bloor, M., Frankland, J., Thomas, M., & Robson, K. (2001). Focus groups in social research.
London: Sage.
Brady, A. (2011). This is why mainstream America votes against gays, Adam Lambert:
Contemporary outness and gay celebrity. Celebrity Studies, 2(3), 292-304.
Butler Breese, E. (2010). Meaning, celebrity and the underage pregnancy of Jamie Lynn
Spears. Cultural Sociology, 4, 337-355.
Carmichael, C. W. (1976). Communication and gerontology: Interfacing disciplines.
Western Speech Communication, 40(2), 121-129.
Carstensen, L. L., Fung, H. H., & Charles, S. T. (2003). Socioemotional selectivity theory
and the regulation of emotion in the second half of life. Motivation and Emotion,
27(2), 103-123.
Cashmore, E. (2006), Celebrity culture. New York: Routledge.
De Backer, C. J. S., Nelissen, M., Vyncke, P., Braeckman, J., & McAndrew, F. T. (2007).
Celebrities: From teachers to friends. Human Nature, 18(4), 334-354.
DeCrescenzo, T. (1997). Gay and lesbian professionals in the closet: Whos in, whos out,
and why? New York: Haworth Press.
Dow, B. J. (2008). Ellen, television, and the politics of gay and lesbian visibility. In C.
Brunsden, & L. Spigel (Eds.), Feminist television criticism: a reader (pp. 93-110).
Berkshire: Open University Press.
Drageset, J. (2004). The importance of activities of daily living and social contact for
loneliness: a survey among residents in nursing homes. Scandinavian Journal of
Caring Sciences, 18, 65-71.

207
Dyer, R. (1998). Stars. London: British Film Institute.
Entman, R. M. (1993). Framing : Towards clarification of a fractured paradigm. Journal of
Communication, 43(4), 51-58.
Evans, J., & Hesmondhalgh, D. (2005). Understanding media: Inside celebrity. Maidenheid:
Open University Press.
Gamson, J. (2001). Normal sins: Sex scandal narratives as institutional morality tales.
Social Problems, 48(2), 185-205.
Gamson, W. A. (1992). Talking politics. Cambridge: Cambridge University Press.
Goffman, E. (1974). Frame analysis. Harmondsworth: Penguin Books.
Gorin, V., & Dubied, A. (2011). Desirable people: Identifying social values through
celebrity news. Media, Culture & Society, 33(4), 599-618.
Hajjar, W. J. (1998). Television in the nursing home: A case study of the media consumption
routines and strategies of nursing home residents. London: Routledge.
Halman, L., Inglehart, R., Diez-Medrano, J., Lujix, R., Moreno, A., & Basnez, M. (2008).
Changing values and beliefs in 85 countries. Leiden: Brill.
Harrington, C. L., Bielby, D. B., & Bardo, A. R. (2011). Life course transitions and the
future of fandom. International Journal of Cultural Studies, 14(6), 567-590.
Harrington, C. L., Brothers, D. (2010). A life course built for two: Acting, aging, and soap
operas. Journal of Aging Studies, 24, 20-29.
Holmes, S. (2005). Off-guard, unkempt, unready?: Deconstructing contemporary
celebrity in Heat magazine. Journal of Media & Cultural Studies, 19(1), 21-38.
Huang, H. (2009). Frame-rich, frame-poor: An investigation of the contingent effects of
media frame diversity and individual differences on audience frame diversity.
International Journal of Public Opinion Research, 22(1), 47-73.
Lull, J., & Hinerman, S. (1997). Media scandals: Morality and desire in the popular culture
marketplace. New York: Columbia University Press.
Marshall, P. D. (2006). The celebrity culture reader. London: Routledge.
Morgan, D. L. (1997). Focus groups as qualitative research (2nd ed.). London: Sage.
Moschis, G. P. (2003). Marketing to older adults: an updated overview of present
knowledge and practice. Journal of Consumer Marketing, 20(6), 516-525.
Nelson, E. A., & Dannefer, D. (1992). Aged heterogeneity: Fact or fiction? The fate of
diversity in gerontological research. The Gerontologist, 32(1), 17-23.

208
Pan, Z., & Kosicki, G. M. (1993). Framing analysis: an approach to news discourse.
Political Communication, 10(1), 55-75.
Rojek, C. (2001). Celebrity. London: Reaktion Books.
Rojek, C. (2007). Cultural studies. Cambridge: Polity Press.
Schiappa, E., Gregg, P. B., & Hewes, D. E. (2005). The parasocial contact hypothesis.
Communication Monographs, 72(1), 92-115.
Starn, O. (2011). The passion of Tiger Woods: An anthropologist reports on golf, race, and
celebrity scandal. Durham: Duke University Press.
Stevens, L., & Hooghe, M. (2003). The swing of het pendulum: The detraditionalisation of
the regulation of sexuality and intimacy in Belgium (1973-2003). International
Journal of the Sociology of Law, 31(2), 131-151.
Strauss, A. L. (1987). Qualitative analysis for social scientists. Cambridge: Cambridge
University Press.
Tulloch, J. (1991). Approaching the audience: The elderly. In E. Seiter, H. Borchers, G.
Kreutzner, & E.-M. Warth (Eds.), Remote Control: Television, Audiences & Cultural
Power (pp. 180-203). London: Routledge.
Turner, G. (2004). Understanding celebrity. London: Sage.
Van den Bulck, H., & Claessens, N. (2013a). Of Local and Global Fame: A Comparative
Analysis of News Items and Audience Reactions on Celebrity News Websites
People, Heat, and HLN. Journalism [online]. Available at
http://jou.sagepub.com/content/early/2013/05/28/1464884913488725
Van den Bulck, H., & Claessens, N. (2013b). Celebrity Suicide and the Search for the
Moral High Ground: Comparing Frames in Media and Audience Discussions of the
Death of a Flemish Celebrity. Critical Studies in Media Communication, 30(1), 69-
84.
Van den Bulck, H., & Claessens, N. (2013c). Guess who Tiger is having sex with now?
Celebrity sex and the framing of the moral high ground. Celebrity Studies, 4(1),46-
57.
Van den Bulck, H., & Van Gorp, J. (2011). Eternal fandom: Elderly fans, the media, and the
staged divorce of a Schlager singer. Popular Communication, 9(3), 212-226.
Vandebosch, H., & Eggermont, S. (2002). Elderly peoples media use: At the crossroads of
personal and societal developments. Communications: The European Journal of
Communication Research, 27, 437-455.

209
Westerhof, G. J. (2010), During my life so much has changed that it looks like a new
World to me A narrative perspective on migrating in time. Journal of Aging
Studies, 24, 12-19.

210
211
 Conclusion

The final part of this PhD summarizes the main findings of the six empirical chapters,
first, by relating the results to the main research question that was posited in the
introduction: What is the social or cultural meaning of the mediated phenomenon of
celebrity for contemporary audiences? and, second, by demonstrating how these
studies have addressed gaps in the current body of celebrity studies.

ADDRESSING THE RESEARCH QUESTION

As stated in the introduction, the main research question throughout this PhD has been:

What is the social or cultural meaning of the mediated phenomenon of celebrity

for contemporary audiences?

The six empirical chapters have addressed this question from multiple angles, either by
combining the elements of text (mediated celebrity in the form of celebrity news) and
audience as in chapter 1 to 3 or by focusing on the celebrity audience in chapters 4 to 6.
Following the theoretical framework, all six empirical studies in this PhD start by
conceptualizing celebrity as a construction, built in the interplay between four actors:
the famous person, the entourage, the media, and the audience. The focus of the studies
is on the bottom of the trapezium-shaped celebrity apparatus (figure below), with
chapters 1 to 3 looking at the relationships between celebrity media and audiences and
chapters 4 to 6 at the celebrity-audience relationships. The wider societal context is
always brought into consideration in the studies, either by referring to societal
processes such as tabloidization and globalization or by linking celebrity
(news/audience) frames to wider societal viewpoints.

212
 Famous person Entourage

CELEBRITY

Audience Society Media

In the following sections, the main research question will be addressed in two parts: first
by summarizing the findings on the defining aspects of the celebrity concept and,
second, by looking at the debate between the celebrity as a commodity or as fulfilling
social and cultural roles for audience members.

Conceptualizing Celebrity
Throughout the PhD, the definition of celebrity that has been described in the theoretical
framework celebrity as a well-known mediated persona, constructed by the celebrity
apparatus, whose private and real personae attract at least as much attention as the
public persona has been the starting point for the conceptualization in the empirical
studies. These studies can thus be considered empirical tests of the applicability of this
definition and its different aspects. The results of these empirical validations are
described below.
First, the criterion of well-knownness is validated by the studies as the
celebrities that are examined are typically (well-)known beyond what is normally
expected of them considering their status (cf. Giles, 2000). This criterion can be
specified, though, as the studies show that the well-knownness of a celebrity is highly
dependent on the media attention granted to him/her. Indeed, one of the few (if not the
only) ways to reach an audience larger than expected is via the media, either in the form
of the classic mass media or new social media. In this PhD, the former are the main
definers or gatekeepers of fame as the first three studies focus explicitly on celebrity
news and demonstrate the medias power to define what celebrity is. Here, celebrities
are considered to be those people who are discussed in celebrity news, because media
213
do not tend to report on people who are not well-known in the celebrity news section.
While Rijt, Shor, Ward, & Skiena (2013) already linked well-knownness to newspaper
coverage, this PhD goes further and includes all media that report on celebrities, from
newspapers over radio or television talk shows to online celebrity news websites. In
addition, social media are expected to play a valuable role for becoming and being well-
known (cf. Justin Biebers rise to celebrity through Youtube) which is why future
research should look further into this aspect. In sum, the criterion of well-knownness is
confirmed to be important for a definition of celebrity and it is strongly linked to the
mediated aspect, which is discussed next.
Second, the celebrity as a mediated persona is confirmed throughout all studies
in this PhD as celebrities need the media to reach their audiences and vice versa because
celebrities and audience members hardly ever meet in person. Chapters 1 to 3 provide
interesting insights in this regard as they explicitly examine the mediated form of
celebrity namely celebrity news and thereby naturally illustrate the crucial nature of
media exposure for celebrity. The other chapters in this PhD also demonstrate the
importance of media for the concept of celebrity. Indeed, not only are media the
essential intermediary between celebrities and audiences (who do not typically
encounter one another in person), chapter 4 also shows that audience members
celebrity news interest is a strong predictor of the strength of parasocial relationships
maintained with celebrities. Thus, the more interested a person is in celebrity news
the mediated form of celebrity the stronger the parasocial relationship with a favorite
celebrity will be. Chapter 5 focuses on nursing home residents and here, the role of the
media is also discussed. The residents themselves are aware that they depend on the
media to learn things about celebrities, to use them as food for conversations or moral
discussions, and to develop and maintain parasocial relationships with them. The media
are thus an indispensable actor in the celebrity apparatus (see below) that strongly
affects the relationships between the other actors. However, the power of the media
should not be overstated, as audiences always combine media frames with their own
dispositions, experiences, (parasocial) relationships, and interaction with peers. Further,
the awareness of the media construction of celebrity among the nursing home residents
shows how critically these audience members deal with celebrity media coverage.

214
 Third, the celebrity construct is the result of negotiations within the celebrity
apparatus, namely between famous people, their entourage, media, and audiences.
Because this PhD has not examined the production of celebrity, the elements of the
famous person and entourage are not tested empirically and should be included in
future research. The relevance of media and audiences is empirically demonstrated
throughout the studies in this PhD. On the one hand, the interdependence of these actors
is confirmed by illustrating how audiences need the media to learn about celebrities and
how media use celebrity (news) to attract audiences. On the other hand, though, there is
a struggle between these actors in the construction of celebrity and this is also
demonstrated throughout the different studies. Indeed, the specific conceptualization of
celebrity differs for media and audiences in various aspects. In the first three chapters,
the specific conceptualization of celebrity by media and audiences is examined by
looking at who is featured in celebrity news and accompanying audience reactions as
well as how they are framed. Indeed, the analysis of the selection of people and
framing of the news allows to grasp an understanding of the celebrity concept that is
dominant for the media and audience members. Combining a systematic (chapter 1) and
in-depth (chapters 2 & 3) understanding of celebrity news and the accompanying
audience reactions, these studies present some valuable findings that confirm the
constructed nature of celebrity within the celebrity apparatus and deepen our
understanding hereof.
In terms of who, the large-scale systematic analysis in chapter 1 shows that this
is similar for media and audiences with the same A-listers (Lindsay Lohan, Lady Gaga,
Cheryl Cole, ) in the top five and the majority of celebrities being white, heterosexual,
and working in the music, film or TV/radio industries. One difference here is that there
is a clear US dominance for the Flemish HLN website with a neglect of local, Flemish
celebrities, whereas audiences discuss both American (global) and Flemish (local)
celebrities. Thus, cultural proximity or the preference for local celebrities is one element
in which the concept of celebrity differs for media and audiences. Chapter 4 provides
insight into the audiences specific conceptualization of celebrity by asking about their
favorite celebrity. Here, male celebrities are mentioned more often and the celebrities
usually work in the music, film, and TV/radio industries. Further, the element of cultural
proximity also plays a role here, with higher-educated and younger audience members

215
preferring global celebrities and lower-educated and older respondents indicating local
celebrities as their favorites.
Further, with regard to how, comparisons of the media coverage and the
accompanying audience reactions in the first three empirical chapters show that
audiences do not always adopt the frame that is presented by the media, either in terms
a complete frame or its reasoning devices: mainly the issues (problems) and judgments.
In other words, audiences are active decoders of media messages. Indeed, chapter 1
demonstrates that whereas media mainly cover professional and love-related themes,
audience members respond rather superficially by focusing more frequently on the
celebrities appearances or personalities. In chapter 3, the media offer a wide range of
sex-related topics but the audience members choose to focus on the scandalous adultery
stories. Further, audiences tend to be much more judgmental than the media in the first
three chapters, which is not surprising considering the circumstances in which media
and audiences operate. Indeed, media wish to (appear to) provide objective news
coverage, have to maintain good working relationships with the celebrities, and want to
avoid law suits, whereas audience members are usually anonymous, which means they
cannot be punished for their reactions, and their responses are ultimately aimed at
expressing judgments. Several mediating factors are found that can explain why media
frames are not just adopted by audience members, but often negotiated and even
rejected. First, peoples personal experiences with a certain theme can mediate the
adoption of media frames. Second, the interaction with peers online (i.e., the other
readers) can evoke counter frames among the audience members, as discussions (both
negative and positive) often culminate in polarized opinions. Third, audience members
are found to refer to their personal connections with celebrities which can be described
as parasocial relationships and this can also influence the adoption, negotiation, or
rejection of the frames offered by the media. There is thus an interesting interplay
between media and audience frames that is mediated by a number of audience-related
factors which confirms the activity of the audience in interpreting celebrity news.
Despite several similarities in the specific conceptualization of celebrity by media and
audiences, there are thus also differences which prove that negotiations take place in the
construction of celebrity, in other words: in the celebrity apparatus.

216
 In addition to a media-audience distinction, the specific conceptualization of
celebrity can also differ between various media. One example is found in chapter 2
where the popular and quality media report on different aspects surrounding a celebrity
suicide and in different manners, with the former focusing on the real celebrity
persona and the latter on the private persona. The first chapter also found differences
between the celebrity websites with People being the friendliest (most positive
judgments) and Heat the most hostile (most negative judgments) toward celebrities.
Similarly, different audience groups tend to have diverse visions of celebrity. For
instance, chapter 4 shows that cultural proximity is especially important for lower-
educated and older audience members. Further, chapter 5 illustrates how nursing home
residents differ from the general audience as they prefer local and older celebrities
whose fame is ascribed or achieved (Rojek, 2001). This study also shows that male and
female nursing home residents celebrity preferences (and thus how they think about
celebrities) differ as they prefer respectively sports and royal celebrities. The
construction or specific conceptualization of celebrity is thus a complex and continuous
interplay between and within the actors of the celebrity apparatus.
The final element of the celebrity definition contains the criterion that celebrities
private and real personae should attract at least as much attention as their
professional activities for both media and audiences. Interestingly, the large-scale
analysis of celebrity news and audience reactions demonstrates that media cover
celebrities most often in professional themes (28.3%). For audiences, celebrities
professional lives are the second most frequently discussed theme (19.4%). At first
sight, these findings tend to refute the at least as much attention to the private or real
personae aspect of a celebrity definition. However, this criterion is still be considered
valuable because chapter 1 demonstrates how private themes (summing up the
different categories) are most often the subject of celebrity news (61.5%) and audience
reactions (73.7%). Further, the cases in chapters 2 and 3 confirm that the private and
real aspects of celebrity are the most interesting ones for audiences (and media), either
in the form of a celebrity suicide as the result of a soap of life (chapter 2) or sex scandals
(chapter 3). Chapters 5 and 6 further show that the (female) nursing home residents are
also mainly interested in the private and real personae of celebrity. This aspect of the
celebrity definition is thus also validated empirically throughout this PhD, with a
dominance of attention to the private and real celebrity personae compared to the

217
public (professional) persona. This is true for both media and audiences, although the
latter tend to focus more on the private and real than the former. This can be linked to
the good working relations that media have to maintain with celebrities and the
negotiations that take place here. Indeed, whereas the celebrities want to promote their
professional activities (new CD, film, sports event, ), the media are ultimately aimed at
gathering information about their private life because the audience is primarily
interested in the latter. As a compromise, the media can agree to publish news on the
public personae, as long as celebrities provide new information in the context of their
private and real personae. In sum, this aspect of a celebrity definition is important, but it
would be more accurate to formulate a slightly different view than Turners (2004: 8)
description of celebritization as: media interest in their activities is transferred from
reporting on their public role [] to investigating the details of their private lives.
Indeed, instead of a transfer of attention, this PhD found that it is better to describe a
combination of attention to public and private roles, with the latter gathering
(quantitatively) more media and, especially, audience attention and evoking
(qualitatively) deeper discussions within media and among audiences.
In sum, the celebrity definition as described in the theoretical framework in the
beginning of this PhD is validated and specified by the empirical studies. Celebrity can
be considered a well-known mediated persona, constructed in the celebrity apparatus
with a complex interplay between media and active audiences whose private and real
personae gather more attention from media and audiences than its professional
activities and evoke deeper discussions among audiences. Within this broad definition,
the specific conceptualizations of celebrity tend to vary between media and audiences as
well as between different media and audience groups. One important aspect here is
cultural proximity, which is found to be more important for audiences than media, and
especially for certain audience groups, namely the lower-educated and older people.
Further, the ordinary-extraordinary paradox that is often linked to celebrities
also proved valuable in this PhD. Indeed, the empirical studies show that, on the one
hand, celebrities are considered as idols, to be loved or hated, who live in a luxurious
world that is free of material problems but in which they are heavily restricted by
constant scrutiny of media and audiences. On the other hand, celebrities are considered
to be people like us, who have similar problems in their personal lives. In addition,
Rojeks (2001) distinction between ascribed, achieved, and attributed celebrity has been

218
useful in the empirical studies. Indeed, in the current celebrity culture with a multitude
of different types of celebrities working in different domains, it is valuable to have a
categorization that is not too restrictive and is able to capture the whole variety of
celebrity in contemporary societies. This is where Rojeks (2001) distinction proves to
be helpful.
The combination of in-depth, small-scale and systematic, large-scale analyses of
celebrity (news) has provided interesting insights with regard to the defining aspects of
celebrity that should be developed further in future research. These future studies
ideally combine case studies with more large-scale, quantitative approaches of the topic
of celebrity. Indeed, while the value of high profile case studies is undeniably high, it is
important to situate these cases in the wider celebrity news context which audience
members are confronted with every single day in a variety of media. The adoption of the
framing concept and method, naturally rooted in (political) communication studies, has
proved to be highly valuable in the context of celebrity news analyses. Future research
could thus benefit from using framing analysis as it provides direct insight in the
essential aspects of a frame, allows the combination of encoding and decoding analyses,
and makes it possible to link the frames found in media and audience items to broader
societal viewpoints, uncovering the underlying mechanisms of frames. Finally, the active
audience concept is clearly demonstrated here, with audiences not just following media
frames but often constructing their own counter frames, based on personal experiences,
parasocial relationships with celebrities, and interaction with peers. The meaning of
celebrity for audiences will be discussed further in the following section.

Celebrity as Commodity or Fulfilling Social/Cultural Roles?

Throughout this PhD, the chapters have looked into and offered valuable knowledge
with regard to the debate on the celebrity-as-commodity versus celebrity as fulfilling
important social and cultural roles in society.
In the empirical studies, celebrity has been demonstrated to operate as a
commodity, distracting audiences from the real issues in society by diverting them to
celebrities antics. Indeed, the superficiality of the audience reactions in chapter 1
focusing on celebrities appearances and personalities could be interpreted in such a
way that celebrity is mainly a commodity that is aimed to draw audiences away from
real-life issues and inequalities and to compensate for their alienation and anonymity as

219
well as the frustrations in capitalist society by directing them to the trivialities of
celebrities lives. The focus on melodrama in chapter 2 and scandal in chapter 3 add to
this thesis as they divert attention from problematic aspects of contemporary society
and public themes that need to be addressed. This confirms Couldry and Markhams
(2007) findings that celebrity (news) is rarely linked to public issues by audience
members. Indeed, throughout the different studies in this PhD, celebrity is only
sporadically discussed on the level of public, societal issues as it naturally stays within
the private or personal sphere (cf. celebrity definition). Celebrity can thus be considered
a commodity to the extent that it does not (usually) question the fundaments of society,
in this case, Western capitalism and consumer culture.
However, this celebrity-as-commodity thesis does not tell the whole story of
celebrity and its meaning for audiences. Indeed, whereas this thesis may apply to the
themes that are discussed by the audience members in the systematic analysis (chapter
1), this is not the case for the other framing functions such as moral judgment. Indeed,
the celebrity-as-commodity claim is not fit to explain the strong judgments in the
audience reactions in contrast to the rather neutral media stories that indicate that
celebrities are not just pure forms of escapism from the problems in capitalist society,
but are employed by audiences to discuss a variety of moral and social themes (cf.
Alberoni, 1972; Alexander, 2010; Cashmore, 2006; Couldry & Markham, 2007; Cowen,
2000; Drake & Miah, 2010; Duits & van Romondt Vis, 2009; Dyer, 1979, 1986; Evans &
Hesmondalgh, 2005; Feasey, 2008; Gorin & Dubied, 2011; Holmes, 2005b; Johansson,
2006; Marshall, 1997, 2006, 2010; Rojek, 2001; Turner, 2004). Although this moral
function of celebrity is not especially dominant in the systematic analysis of everyday
celebrity news coverage (which confirms Couldry and Markhams (2007) findings), this
celebrity role does become very clear in the analyses of high profile cases such as the
suicide of a Flemish celebrity in chapter 2 and the adultery stories of Tiger Woods, Jesse
James, and so on in chapter 3. Here, celebrities do not just distract audiences from real-
life problems by providing scandalous stories on celebrities lives, but the audiences
employ celebrity (news) as a forum of discussion of a myriad of social and moral themes
that can certainly be considered real-life issues in contemporary societies, albeit on a
personal level.

220
 Indeed, the social and cultural roles of celebrity are definitely present in
contemporary Western societies and their importance is confirmed in this PhD, but they
are mainly confined within the boundaries of the private, personal sphere. Some
exceptions are the political lobbying or philanthropic work of celebrities such as
Angelino Jolie, George Clooney, and so on, but these are rare. Therefore, it can be stated
that, paradoxically, celebrity is both a commodity and fulfills social and cultural roles in
society, but that these roles (co-)exist on different levels. The commodity thesis is valid
for the public, societal level as celebrities, celebrity news, or audiences hardly ever
question public issues related to the fundaments of society, consumer culture, or
Western capitalism. On a personal, private level, celebrities can and do fulfill important
social and cultural roles but they continue to operate within the boundaries of the larger
frame of Western consumer society.
This focus on the private is not surprising considering that this is a crucial
element of the definition of celebrity and that celebrity (news) is often linked to
processes of tabloidization and personalization. Here, hard news (~public-level issues)
is increasingly replaced by soft news (~private-level issues) and the focus shifts from
society or societal groups to individuals. However, the focus on the private and personal
does not mean that celebrities social or cultural roles are not valuable, on the contrary,
they are of great importance for audience members, as demonstrated throughout the
studies in this PhD. Further, it is not because people are interested in or discuss celebrity
news with regard to personal themes, that they are unaware or ignorant of societal
issues. Indeed, audience members can turn to celebrity news to make meaning of
themes on a private level and at the same time turn to other sources (e.g., general news)
to understand public themes. Within the active audience paradigm, this is especially
important: it is not because celebrity is a commodity, aimed to distract audiences from
societal issues, that these audiences will just comply and neglect public themes.
In sum, celebritys main value should be situated on the private level, not
subverting the fundaments of Western capitalist society (cf. commodity on a public
level), but operating within them while focusing audience attention to themes or issues
within the personal sphere. Here, celebrity represents societal ideas, values, and norms
that can help audiences make meaning of themselves and the world (on a private level).
The opium for the masses view that is often expressed by proponents of the celebrity-

221
as-commodity thesis is thus too strong in our view because celebrity does fulfill
important social and cultural roles in society, albeit on a personal and private level.
One of celebrities social or cultural functions in society as confirmed in this PhD
is to help audience members make meaning of the world and their own identities by
evoking social discussions on a variety of moral themes that are mainly situated within
the personal, private life sphere. The excellent suitability of celebrities for these issues of
morality can be explained by the fact that celebrities are the ideal subjects of gossip and
moral discussions as they are shared acquaintances well-known among a large group
of people but can be criticized without risking any real-life consequences in contrast
to family and friends (Feasey, 2008). Throughout the PhD, the references to personal
experiences in the context of discussions evoked by celebrity (news) indicate that
celebrities do indeed help meaning-making of ones own identity and the world (on a
personal/private level). The social and cultural role of celebrity as a lead into moral
discussions is further confirmed in the context of nursing home residents in chapters 5
and, especially, 6. Indeed, chapter 6 demonstrates how the moral themes of adultery and
homosexuality that are often sensitive, especially for older people because of the past
taboo status can be discussed rather easily by linking it to a specific celebrity.
Conversations evoked by celebrities can thus shine light on the different frames or
viewpoints that live among the nursing home residents in chapter 6 and among the
online readers in chapters 2 and 3. Interestingly, celebrity news usually embodies the
dominant values in contemporary societies (cf. Gorin & Dubied, 2011) and can thus be
considered hegemonic. However, the active nature of the audience and, in particular, the
construction of counter(-hegemonic) frames among audience members demonstrate
that celebrity is not just a hegemonic tool as it has the potential to evoke profound
debates on moral themes among audience members that could challenge dominant
views in society. These debates mostly focus on personal-sphere themes, that do not
subvert the fundaments of Western consumer societies, but have the potential to evoke
debate and challenge hegemonic views for issues on a private level.
In addition to this social role of celebrity, several audience members in chapters 1
to 3 refer to their personal and often highly emotional involvement with particular
celebrities, in the form of parasocial relationships, which also indicates that the meaning
of celebrity goes far beyond the commodity claims (cf. Giles, 2002, 2003; Horton & Wohl,
1956; Marshall, 2010; Rojek, 2001; Turner, 2004). In chapter 1, the importance of

222
parasocial relationships is demonstrated with 44 percent of the audience reactions
containing references to these (both positive and negative) emotional connections to
celebrities. Chapter 4 provides more knowledge on this form of celebrity-audience
connection by illuminating the nature of parasocial relationships which consists of an
emotional connection and an analogy to social relationships and the way in which
audience members and celebrities (socio-demographic) characteristics correlate with
the strength of these relationships. This study shows that parasocial relationships with
celebrities are widespread in contemporary Western societies, which is not surprising
considering the omnipresence of celebrities. The relevance of celebrities as social
companions in the form of parasocial relationships is thus also an important social
function of the phenomenon of the celebrity. Surprisingly, parasocial relationships with
celebrities are stronger for older audience members, which contrasts the findings by
Couldry and Markham (2007) that celebrity audiences are mostly young. In addition, the
lower-educated audience members tend to have stronger parasocial relationships with
their favorite celebrities. In the selection of a favorite celebrity and the parasocial
relationships with him/her, the concept of cultural proximity proved to be important.
Indeed, audience members tend to have stronger parasocial relationships and
emotional connections with local than global celebrities. The latter is confirmed in the
study in chapter 5 where nursing home residents indicated stronger involvement with
local than global celebrities. Finally, in chapter 1 more references to positive parasocial
relationships are found for local than global celebrities, indicating a positive relationship
between cultural proximity and closeness to celebrities. Interestingly, the ordinary-
extraordinary paradox of celebrity also proved important for parasocial relationships in
chapter 5, as these are more positive when the celebrity acts normal and more
negative when the celebrity act like a star.
Other social roles of celebrity are found in the exploratory study in chapter 5 on
the meaning of celebrity for nursing home residents. Here, celebrities are found to not
just be leads into moral discussions or parasocial companions, but also to provide
audiences in general and nursing home residents in particular with food for
conversation (cf. Cashmore, 2006; Couldry & Markham, 2007; Drake & Miah, 2010;
Feasey, 2008; Johansson, 2006; Marshall, 2006, 2010; Turner, 2004). This food-for-
conversation role differs from the lead-into-moral-discussions function as the former
does not have to cover morally or socially relevant themes but can be superficial or

223
trivial, as long as a celebrity (story) promotes social interaction. This is especially
valuable for people who do not have a lot in common but wish to communicate with
each other or when there is a lack of other topics for conversation. Celebrity can thus be
used as a safe, reassuring and comforting way into social encounters (Feasey, 2008:
692). Similarly to the moral discussions function, the main value of celebrity as a topic
for conversation lies in its role as a shared but distant acquaintance, who provide
points of commonality for people to reconnect both with celebrities and with each
other (Marshall, 2006: 317). In addition, we can criticize these celebrities life choices
without any consequences because, for instance, Angelina Jolie or Brad Pitt will never
know and thus will not be angry for condemning them (Feasey, 2008).
A final social or cultural role of celebrity that is found in chapter 5 is as an aide-
memoire. This role is not expected to be as important for the general audience as for
nursing home residents considering the fact that reminiscence gains importance later in
life. As an aide-memoire, celebrity can help trigger memories and reminiscence, which
enhances the self-esteem of older adults and can be useful to reconstruct their life
course. The aide-memoire role of celebrity has not been (theoretically) suggested
previously in celebrity studies or gerontology which makes it one of the valuable
innovations of this PhD research. Future research should explore this further to see, for
instance, how celebrity can become a part of the animation activities in nursing homes
as a means to promote social conversations, discussions of moral themes, to trigger
parasocial relationships with celebrities, and as an aide-memoire to foster reminiscence.
In sum, the studies in this PhD have demonstrated that celebrity is a form of
entertainment and, on a public, societal level, a commodity aimed at distracting
audiences from their capitalist-related problems and frustrations by means of
celebrities trivialities, but it also fulfills a variety of important social or cultural roles in
contemporary Western societies, on a private, personal level. Indeed, celebrity (news)
can provide food for conversations and, thus, promote social interaction in an
increasingly individualized society where isolation is a growing problem (Rojek, 2001;
Turner, 2004). Further, audiences can use celebrities to make meaning of their own
identity and the world as a forum for discussion of a myriad of moral themes. These
themes are mainly situated within the personal life sphere which means that celebrity is
not typically used to question the fundaments of society, but that it helps to give
meaning to private-level issues. This role is especially important in secularized societies

224
 such as most Western countries in which the traditional institutions for meaning-
making are losing importance and people are searching for new means to construct a
view of their own identity and the world (Cashmore, 2006; Couldry & Markham, 2007;
Marshall, 2006, 2010; Rojek, 2001; Turner, 2004). In these two roles, celebrities are
ideal subjects as they are well-known by a lot of people and can be talked about without
any social risks of hurting or insulting any real-life acquaintances. Celebrities can also be
a form of social companionship, often in the form of parasocial relationships, as
audiences feel involved with and connected to them. This role is especially important for
lower-educated and older people and considering the increasing isolation in the strongly
individualized contemporary Western societies (especially for vulnerable groups such
as the lower-educated and older), this may an especially valuable aspect of celebrity
(Rojek, 2001; Turner, 2004). Indeed, parasocial relationships are characterized by an
emotional connection and are analogous to social relationships. A celebritys cultural
proximity also plays an important role here, especially for older and lower-educated
audiences. Finally, in the context of the nursing home, celebrities are found to be an
aide-memoire, helping older adults to reminisce and structure their life course by
relating their own life events to those of celebrities. These social roles may not
dominantly come to the surface by (structurally) investigating celebrity news although
high profile cases already reveal these functions more clearly than everyday celebrity
coverage which is why the combination of textual and audience studies is crucial for an
understanding of celebrity for contemporary audiences.
Despite its apparent triviality or superficiality, celebrity is thus an influential and
valuable part of contemporary Western societies and it is important to further research
and expand the understanding of their commodity aspect as well as the social and
cultural roles because their omnipresence in both media and society continues to
increase. Indeed, future research should investigate this from different angles, for
different groups in society, and in different contexts to reach a more comprehensive
understanding of the meaning of celebrity for contemporary Western and non-Western
audiences. In addition, the aspect of production in the celebrity apparatus which is not
addressed here as it would blur the focus of this PhD should be the subject of future
research, both in the context of large global and smaller local celebrity industries.

225
 ADDRESSING GAPS IN CELEBRITY STUDIES

This PhD and its six empirical studies offer a valuable contribution to the current body
of research in the academic domain of celebrity studies as they address a number of
shortcomings that are pointed out by, amongst others, Holmes and Redmond (2010) as
well as Turner (2010) in the first edition of Celebrity Studies.
First, this PhD addresses an important gap in celebrity studies by focusing on the
reception of celebrity and presenting empirical studies that examine celebrity and
audiences in the wider context of media texts and contemporary Western society.
Indeed, the current body of celebrity research is strongly rooted in media and cultural
studies which entails that there is a focus on textual analysis and a lack of attention to
the production or consumption of celebrity. This neglect is considered to be one of the
major gaps in celebrity studies, especially with regard to audience research (Holmes &
Redmond, 2010; Turner, 2010). While most celebrity scholars (Alberoni, 1972;
Alexander, 2010; Boorstin, 1961; Butler Breese, 2010; Cashmore, 2006; Dyer, 1979,
1986; Giles, 2002, 2003; Holmes, 2005a, 2005b; Holmes & Redmond, 2010; Marshall,
1997, 2006; Rojek, 2001; Turner, 2004, 2010) theoretically agree that celebrity is more
than just a form of entertainment or a commodity as it fulfills important social and
cultural roles in society, there is hardly any empirical audience research that validates
this. In response, the chapters in this PhD consist of empirical studies that either
combine the elements of text and reception (Chapters 1-3) or focus exclusively on
audiences (Chapters 4-6). These studies illuminate the conceptualization of celebrity,
with a complex interplay between media and audiences in the celebrity apparatus as
well as the co-existence (on different levels) of celebrity-as-a-commodity and as fulfilling
social and cultural functions of celebrities in contemporary Western societies.
Second, the combination of large-scale systematic studies in chapters 1 and 4
with more in-depth case studies in chapters 2, 3, 5, and 6 in this PhD helps to expand the
academic knowledge on celebrity that is, so far, mainly based on case studies that shine
light on high profile cases but lack a form of systematic or structural knowledge (Turner,
2010). This PhD thus allows to bring together the insights from more large-scale
quantitative (chapters 1 & 4) and more profound qualitative studies (chapters 5 & 6).
This is important to situate high profile cases within a larger whole, of everyday

226
celebrity news coverage, to gain a wider understanding of parasocial relationships, and
to allow for a deeper view of the relevance of celebrity for nursing home residents.
Chapters 2 and 3 unite quantitative and qualitative methods in their research by
employing the former to gain an understanding of the degree to which certain elements
or frames are present in celebrity news and audience reactions and the latter to provide
a more profound insight in the nature and meaning of these elements or frames.
Third, the focus on one specific group in society, namely older adults, in the final
two chapters of this PhD aims to compensate for celebrity studies neglect of certain
groups (such as older adults) in contemporary Western societies. The fourth chapter of
this PhD demonstrates the high relevance of celebrities in the lives of older adults and
the strong connections between them, which is why this group is selected as the focus
for the last two empirical chapters. This findings in chapter 4 are rather surprising as it
is mostly assumed that the celebrity audiences are young (Couldry & Markham, 2007),
which is another reason why this group is selected for the final two studies. In
particular, nursing home residents are selected as the group under study because of
their specific characteristics (in terms of free time, mobility, and social network) that
predict a special interest and higher use of celebrity (news). The studies on this
particular group allow for a deeper understanding of the meaning of celebrity for
nursing home residents.
Finally, the combination of multiple academic domains in this PhD contributes to
the multidisciplinary character of celebrity studies that is often described as one of its
main strengths (Holmes & Redmond, 2010; Turner, 2010). However, most research on
celebrity is still strongly positioned within cultural studies where the roots of celebrity
studies can be situated and rarely combines it with other academic disciplines. This is
why this PhD incorporates concepts, methods, and insights from three other academic
domains in the study of celebrity. Cultural studies is combined with communication
studies in the first three empirical chapters, as the latter allows for more structural
analyses of celebrity news and audience reactions. In particular, the concept and method
of framing that is mainly used in political communication studies is integrated in the
study of celebrity news and audience responses in chapters 1, 2, and 3 to provide direct
insight into the essential elements of a message and allow for a comparison between
encoding (media) and decoding (audience) processes. Chapter 6 also employs framing
analysis to examine the social role of celebrity as a lead into moral discussions for

227
nursing home residents. Further, chapter 4 examines a psychological concept
parasocial relationships in the context of celebrities, by adhering to the rigorous
standards of psychological research and maintaining a cultural studies view on
celebrity-audience involvement. Finally, the academic domain of gerontology is
combined with celebrity studies in the final two empirical chapters (5 & 6) as the focus
of these studies is on the role of celebrity in the lives of nursing home residents. Here,
insights from gerontology help to interpret the cultural and social functions of celebrity
in the nursing home context.

ISSUES FOR FUTURE RESEARCH

Although this PhD addresses a number of shortcomings in the current field of celebrity
studies, it also has certain limitations itself, which can and should be addressed in future
research on celebrity.
First, several chapters in this PhD employ online data to gain a wider
understanding of celebrity news and its audience reception. However, the use of online
data entails that only a segment of the audience is reached as not everyone is present on
the Internet and even a smaller segment actively reacts on celebrity news. Therefore, it
would be useful to examine the reception of celebrity in an offline form, aiming to reach
a wider segment of the audience. In addition, although online celebrity news is expected
to be highly similar to the offline versions, this is not yet validated by research. Future
studies should thus also analyze offline celebrity news and compare it to the online
equivalent. In addition, it would be interesting to not only use official celebrity news
sources, but also the increasingly popular blogs (e.g., Perez Hilton) or to examine the
celebrity sections in more generally-oriented news sources. A combination of systematic
and in-depth studies should be employed to provide a more complete view of celebrity
news, both online and offline.
Second, the data in the first and third chapter of this PhD are gathered on three
celebrity websites, hereby allowing the comparison of American, British, and Flemish
celebrity news and audience reactions. Although this comparison is useful as it includes
two large global celebrity industries and one small local industry, it would also be
fruitful to include other national contexts in future research, such as other Western
countries (with smaller or larger celebrity industries) or, even more interestingly, non-

228
Western contexts. Indeed, this would allow to create a more comprehensive view of the
phenomenon of celebrity worldwide. Here, both systematic analyses and case studies
are needed to provide a large-scale wider as well as a small-scale in-depth
understanding of the phenomenon of celebrity (news and audience reactions) in
different national contexts.
Third, the latter two studies of this PhD focus on one particular group in the
audience, namely nursing home residents. Although this choice is strongly needed and
well-founded, the selection of one particular audience group always implies that other
groups are not included. Therefore, future research would benefit from examining other
populations in the audience, such as teenagers, young adults, adults, (independently-
living) older adults, and so on. More systematic research could also be useful to validate
the findings of the exploratory study in chapter 5 of the social roles of celebrity, not only
for nursing home residents but also for other audience groups. In addition to age,
different audience groups can be distinguished in terms of gender, education, ethnicity,
and so on.
Finally, the aspect of celebrity production was not examined in this PhD as it
would distract from the main research question and blur the focus of the empirical
studies. However, production is an important part of the celebrity system that needs to
be explored thoroughly, which is why future research should focus on this aspect. This
PhDs trapezium-shaped presentation of the celebrity apparatus can be a useful starting
point for future production research that examines the relationships between the
famous person, the entourage, media, and audiences. In the study of celebrity
production, the multidisciplinarity of celebrity studies can be honored by combining
cultural studies viewpoints with those of political economy.

229
 SUMMING IT UP

Addressing the main research question, the studies in this PhD have
demonstrated that celebrity is a commodity on a public level and fulfills a variety of
important social or cultural roles in contemporary Western societies on a private or
personal level. These two paradoxical viewpoints are thus both correct as the roles of
celebrity as commodity and as fulfilling social/cultural roles are empirically proven to
co-exist in celebrity culture, albeit on different levels. Celebrity (news) can provide food
for conversations and help audiences to make meaning of their own identity and the
world. In these two roles, celebrities are ideal subjects as they are well-known by a lot of
people and can be talked about without any social risks of hurting or insulting any real-
life acquaintances. Celebrities can also be a form of social companionship, often in the
form of parasocial relationships, as audiences feel involved with and connected to them.
Finally, in the context of the nursing home, celebrities can be an aide-memoire, helping
older adults to reminisce and structure their life course by relating their own life events
to those of celebrities. Within the field of celebrity studies, this PhD has aimed to
compensate for the lack of audience research, the scarcity of systematic knowledge, the
neglect of particular groups in society, and the need to combine insights from different
academic fields to honor the multidisciplinary nature of celebrity studies. These
shortcomings should further be addressed in future research and the findings of these
PhD should be developed further both with regard to the conceptualization of celebrity
and its meaning for audiences as the celebrity phenomenon is increasingly
omnipresent in contemporary (Western) societies and thus continues to be a large part
of everyday life for almost all audience members.

230
 REFERENCES

Alberoni, F. (1972). The Powerless Elite: Theory and Sociological Research on the
Phenomenon of Stars. In D. McQuail (Ed.), Sociology of Mass Communications (pp.
75-98). Harmondsworth: Penguin Books.
Alexander, J. C. (2010). The Celebrity-Icon. Cultural Sociology, 4, 323-336.
Boorstin, D. J. (1961). The Image: A Guide to Pseudo-Events in America. New York:
Vintage.
Butler Breese, E. (2010). Meaning, Celebrity and the Underage Pregnancy of Jamie Lynn
Spears. Cultural Sociology, 4, 337-355.
Cashmore, E. (2006). Celebrity Culture. New York: Routledge.
Couldry, N. & Markham, T. (2007). Celebrity Culture and Public Connection: Bridge or
Chasm? International Journal of Cultural Studies, 10(4), 403-421.
Cowen, T. (2000). What Price Fame? Cambridge: Harvard.
Duits, L., & van Romondt Vis, P. (2009). Girls make sense: Girls, celebrities and identities.
European Journal of Cultural Studies, 12(1), 41-58.
Dyer, R. (1979) (reprinted 1998). Stars. London, BFI.
Dyer, R. (1986). Heavenly Bodies: Film Stars and Society, London: BFI.
Feasey, R. (2008). Reading heat: The Meanings and Pleasures of Star Fashions and
Celebrity Gossip. Continuum, 22(5), 687-99.
Giles, D. C. (2002). Parasocial Interaction: A Review of the Literature and a Model for
Future Research. Media Psychology, 4, 279-305.
Giles, D. C. (2003) Media Psychology. London: Routledge.
Gorin, V., & Dubied, A. (2011). Desirable people: Identifying social values through
celebrity news. Media, Culture & Society, 33(4), 599-618.
Holmes, S. (2005a). Off-guard, Unkempt, Unready?: Deconstructing Contemporary
Celebrity in Heat Magazine. Journal of Media and Cultural Studies, 19(1), 21-38.
Holmes, S. (2005b). Starring ... Dyer?: Re-visiting Star Studies and Contemporary
Celebrity Culture. Westminster Papers in Communication and Culture, 2(2), 6-21.
Holmes, S., & Redmond, S. (2010). A journal in Celebrity Studies. Celebrity Studies, 1(1),
Marshall, P. D. (1997). Celebrity and Power: Fame in Contemporary Culture. Minneapolis:
University of Minnesota Press.

231
Marshall, P. D. (2006). The Celebrity Culture Reader. London: Routledge.
Marshall, P.D. (2010). The promotion and presentation of the self: celebrity as marker of
presentational media. Celebrity Studies, 1(1), 35-48.
Rojek, C. (2001). Celebrity. London: Reaktion Books.
Turner, G. (2004). Understanding Celebrity. London: Sage.
Turner, G. (2010). Approaching celebrity studies. Celebrity Studies, 1(1), 11-20.

232
233
 Nederlandse samenvatting
Celebrity, Media, en Publiek
Sociale en Culturele Betekenis in hedendaagse Westerse Samenlevingen

Hedendaagse Westerse samenlevingen worden steeds sterker gekenmerkt door de

alomtegenwoordigheid van celebrities (of beroemdheden), zowel in de media als in
andere maatschappelijke domeinen (politiek, zakenwereld, ). Hierbij wordt celebrity
(nieuws) vaak gezien als iets triviaal of oppervlakkig of een vorm van lage cultuur, zowel
in de samenleving als in het academische veld. Dit wordt op het eerste gezicht bevestigd
wanneer men gaat kijken naar celebrity nieuws aangezien een groot aandeel inderdaad
gaat over oppervlakkige of triviale onderwerpen zoals het uiterlijk van celebrities. Dit
past binnen de celebrity-als-commodity thesis die deel uitmaakt van een breder debat
in het academische veld van celebrity studies. Volgens deze visie zijn celebrities
consumptiegoederen die gecreerd worden door de kapitalistische consumptiecultuur
om het publiek af te leiden van de echte problemen in de maatschappij. Door triviale
(schandaal-)verhalen over celebrities te verspreiden en het kapitalisme te promoten,
blijft de consumptiecultuur dominant in de Westerse samenleving. Echter, dit is niet de
enige visie op het sociale of culturele belang van celebrities in de hedendaagse
maatschappij. Een andere strekking binnen celebrity studies meent immers dat celebrity
meer is dan alleen entertainment of een commodity aangezien het ook belangrijke
sociale en culturele rollen vervult, zoals een bron voor sociale interactie, een manier om
sociale of morele themas te bespreken en betekenis te geven aan de wereld en de eigen
identiteit, en een vorm van (para-)sociaal gezelschap. Dit doctoraat wil dit debat van
naderbij bestuderen aan de hand van zes empirische studies om na te gaan of celebrity
alleen maar gaat om entertainment en consumptie of ook sociale en culturele functies
vervult. De centrale onderzoeksvraag luidt dan ook als volgt: Wat is de sociale of
culturele betekenis van het gemedieerde fenomeen celebrity voor het hedendaagse
publiek? Om deze onderzoeksvraag te beantwoorden, zijn verschillende academische
domeinen geraadpleegd. Celebrity studies staat hierbij centraal en aangezien dit domein
sterk geworteld is in cultural studies, is ook dit veld belangrijk voor de studies in dit
234
doctoraat. De inzichten vanuit celebrity en cultural studies worden verder aangevuld
met concepten, methodes, en inzichten vanuit communicatie studies, psychologie, en
(sociale) gerontologie. Dit promoot het multidisciplinaire karakter van celebrity studies,
dat vaak als n van diens sterktes wordt aangehaald.
Celebrity wordt in dit doctoraat geconceptualiseerd als een bekende,
gemedieerde personae, geconstrueerd binnen het celebrity apparaat (bestaande uit de
bekende persoon, diens entourage, de media, en het publiek), wiens private en echte
personae minstens evenveel aandacht genereren bij media en publiek als diens
professionele activiteiten. De accuraatheid van deze conceptualisering wordt doorheen
het doctoraat onderzocht door deze theoretische assumpties te toetsen aan empirische
studies. Om de centrale onderzoeksvraag te beantwoorden, focussen de studies in dit
doctoraat op het tekstuele (gemedieerde) en publieksaspect van celebrity. De eerste drie
empirische studies onderzoeken beide aspecten aan de hand van analyses van celebrity
nieuws en de bijhorende publieksreacties. Hierbij wordt een combinatie gebruikt van
systematisch kwantitatief en diepgaand kwalitatief onderzoek. De laatste drie studies
concentreren zich op het publiek. Hier is er dus geen expliciete analyse van celebrity
nieuws aanwezig maar wordt wel in het achterhoofd gehouden hoe belangrijk de media
zijn voor de celebrity constructie. Bij deze studies wordt opnieuw kwantitatief en
kwalitatief onderzoek gecombineerd.
Het empirische gedeelte van het doctoraat begint met een grootschalige
systematische studie van het tekstuele en publieksaspect van celebrity. Dit is zeldzaam
in het domein van celebrity studies, waar men meestal gebruik maakt van case studies
en focust op het tekstuele aspect. Deze studie vormt dus een waardevolle bijdrage aan
het academische veld omwille van het grootschalige karakter en de combinatie van tekst
en publiek. Dit eerste empirische hoofdstuk bestaat uit een kwantitatieve analyse van
een grote steekproef van online celebrity nieuws op HLN (Vlaanderen), Heat (Groot-
Brittanni), en People (Verenigde Staten) en de bijhorende online publieksreacties. De
gehanteerde methode hier is gebaseerd op de (in politieke communicatie studies
veelgebruikte) framing analyse, met name op de vier reasoning devices van framing:
probleem, oorzaak, oordeel, en oplossing. Op basis van deze elementen wordt er een
vergelijking gemaakt tussen de mediaberichtgeving over en publieksreacties op lokale
versus globale celebrities. Hier blijkt culturele nabijheid een interessante factor te zijn
waardoor men anders reageert op lokale celebrities. Daarnaast worden de framing

235
elementen van de media en het publiek vergeleken waaruit blijkt dat het publiek
voornamelijk praat over het uiterlijk van de celebrity terwijl media meer naar de
professionele activiteiten kijken. Dit kan genterpreteerd worden binnen de celebrity-
als-commodity thesis, waarbij de oppervlakkige of triviale nieuwtjes van celebrities
mensen moeten afleiden van de echte problemen in de samenleving en de
consumptiecultuur moeten versterken. Deze thesis wordt echter genuanceerd door de
sterke oordelen die geuit worden door het publiek, zowel in positieve als negatieve zin,
die erop lijken te wijzen dat mensen ook sterke opinies hebben over bepaalde themas in
de context van celebrity. Deze themas zijn niet alleen triviaal of oppervlakkig (zoals het
uiterlijk), maar ook sociaal en moreel wanneer ze gaan over meer fundamentele
aspecten van de hedendaagse samenleving of cultuur. Celebrity lijkt dus meer te zijn dan
puur entertainment of een consumptieobject en dit zal verder onderzocht worden in de
volgende studies.
De tweede studie bestaat uit een case studie van de mediaberichtgeving en
publieksreacties rond de zelfdoding van Bekende Vlaming Yasmine. Door middel van
een kwantitatieve en kwalitatieve framing analyse wordt duidelijk dat deze gebeurtenis
tal van sociale of morele discussies opwekt bij de media en het publiek. De themas die
hierbij aan bod komen zijn zelfdoding, scheiding, homoseksualiteit, enzovoort. Een
vergelijking van media en publieksframes maakt duidelijk dat lezers niet zomaar een
media frame overnemen, maar dit toetsen aan hun persoonlijke ervaringen, de
parasociale relatie die ze onderhouden met de celebrity, en de interactie met peers
(hier: andere lezers). Deze studie toont dus aan dat celebrity meer is dan een commodity
en belangrijke functies vervult in de hedendaagse Westerse samenleving, namelijk als
een manier om morele of sociale themas aan bod te laten komen of zelfs bespreekbaar
te maken.
Het derde empirische hoofdstuk van het doctoraat bevat een case studie van
celebrity nieuws over seks en de bijhorende publieksreacties. De data van deze studie
zijn een selectie van de data in de eerste studie, met name de artikels en reacties over
seks. Hier worden een kwantitatieve framing-based analyse en een kwalitatieve
framing analyse gecombineerd. De kwalitatieve analyse focust op overspel en hierbij zijn
zeven frames ge(re-)construeerd. Deze studie toont opnieuw aan dat media frames niet
zomaar worden overgenomen door het publiek, maar dat deze getoetst worden aan hun
persoonlijke ervaringen, parasociale relatie met de celebrity, en interactie met peers.

236
Verder wordt hier opnieuw duidelijk dat celebrity niet zomaar een commodity is,
ondanks de grote aandacht van het publiek voor (seks)schandalen, maar een belangrijke
rol vervult als trigger van morele of sociale discussies die verder gaan dan triviale of
oppervlakkige themas en echte problemen in de samenleving bespreekt.
Aangezien de vorige studies het belang van parasociale relaties met celebrities
aantonen bij het al dan niet overnemen van media frames, focust het vierde hoofdstuk
op deze relaties. Een parasociale relatie kan gedefinieerd worden als de illusie van een
vriendschap tussen een celebrity en het publiek, die eenzijdig en gemedieerd is, maar
een emotionele band omvat en gelijkaardig is aan sociale relaties. De vierde studie van
dit doctoraat bestaat uit een kwantitatieve online survey waarbij aan een
representatieve steekproef van 1000 Vlamingen gevraagd is naar hun favoriete
beroemdheid en hun parasociale relatie met deze persoon. Interessant hier is dat
culturele nabijheid vooral een rol speelt bij de keuze van favoriete beroemdheden bij
lageropgeleide en oudere respondenten. Aan de hand van een nieuwe parasociale relatie
(PSR) schaal van 13 items wordt verder duidelijk dat lageropgeleide en oudere mensen
met een sterke interesse in celebrity nieuws de sterkste parasociale relaties
onderhouden met hun favoriete beroemdheden.
Op basis van de bevindingen van de voorgaande studie, die het speciale belang
van celebrities voor ouderen aantoont, zullen de laatste twee hoofdstukken van dit
doctoraat zich focussen op ouderen en celebrities. Hierbij wordt n specifieke groep
ouderen geselecteerd, namelijk rusthuisbewoners, aangezien zij (meer dan andere
ouderen) veel vrije tijd hebben, beperkt zijn qua mobiliteit, en steeds kleinere sociale
netwerken hebben. Van deze factoren wordt gedacht dat ze het mediagebruik van deze
groep ouderen zouden verhogen en dus ook de mate waarin men in contact komt met
celebrities of er in genteresseerd is. In de vijfde empirische studie van dit doctoraat
wordt op exploratieve wijze nagegaan welke sociale of culturele rollen celebrities
kunnen spelen in het dagelijkse leven van rusthuisbewoners. Aan de hand van
individuele en focus groep interviews, wordt onderzocht welke celebrities de voorkeur
krijgen van de rusthuisbewoners alsook welke betekenis ze voor hen hebben. Hierbij
wordt duidelijk dat celebrities een gemakkelijk gespreksonderwerp zijn (en dus sociale
interactie kunnen vergemakkelijken), een goede manier om morele of sociale themas te
bespreken, een vorm van (parasociaal) gezelschap, en een aide-memoire waarbij
herinneringen aan celebrities helpen om andere herinneringen op te roepen. Deze

237
studie toont met andere woorden aan dat celebrities voor rusthuisbewoners meer zijn
dan een commodity of entertainment aangezien ze belangrijke sociale of culturele rollen
vervullen in hun dagelijkse leven.
Het laatste empirische hoofdstuk focust opnieuw op rusthuisbewoners en gaat
dieper in op n specifieke rol van celebrity in hun dagelijkse leven: het opwekken van
morele of sociale discussies. Aan de hand van focusgroepen waarin fotos van celebrities
getoond worden wordt duidelijk dat celebrities relatief gemakkelijk discussies oproepen
omtrent allerlei morele of sociale themas. In de focusgroepen van deze studie worden
voornamelijk de themas van overspel en homoseksualiteit besproken door de
rusthuisbewoners. Een framing analyse is hier gebruikt om de verschillende
standpunten van de rusthuisbewoners met betrekking tot overspel en homoseksualiteit
te onderzoeken. Hierbij worden rusthuisbewoners gezien als migranten in de tijd, die
zich ofwel kunnen aanpassen aan de nieuwe cultuur, blijven vasthouden aan de cultuur
vanuit het verleden, of beide strategien combineren. De framing analyse toont aan al
deze acculturatiestrategien gebruikt worden door de rusthuisbewoners, maar dat de
meesten een evenwicht zoeken tussen de hedendaagse en vroegere normen en waarden.
Deze studie toont opnieuw aan dat celebrity belangrijke sociale en culturele rollen
vervult in de hedendaagse Westerse samenleving.
Wat de conceptualisering van celebrity betreft, wordt het belang van het bekende
en gemedieerde karakter empirisch bevestigd en hierbij is duidelijk geworden dat deze
aspecten sterk aan elkaar gelinkt zijn, aangezien celebrities de media nodig hebben om
bekend te worden en blijven. Daarnaast wordt bevestigd dat celebrity een constructie is
en het resultaat van complexe onderhandelingen in het celebrity apparaat. Hierbij
hebben media en publiek vaak dezelfde visie op het concept celebrity, maar zijn er toch
belangrijke verschillen met betrekking tot wie zij als celebrity beschouwen en hoe die
dan besproken wordt. Het actieve publiek neemt dus niet steeds het beeld van de media
over, maar toetst dit af aan de eigen persoonlijke ervaringen, interactie met peers, en de
parasociale relaties met celebrities. Celebrity wordt met andere woorden continu
geconstrueerd in een wisselwerking tussen media en publiek. Daarnaast verwachten we
dat de bekende persoon en diens entourage ook deel uitmaken van deze wisselwerking,
maar aangezien dit doctoraat zich focuste op de tekst en het publiek, moet dit nog in
toekomstig onderzoek behandeld worden. Tot slot wordt celebrity vaak gedefinieerd
aan de hand van media- en publieksaandacht voor de private en echte personae, eerder

238
dan het professionele. Dit wordt bevestigd door de empirische studies, waarbij private
themas niet alleen vaker aan bod komen, maar ook diepere discussies opwekken.
De studies in dit doctoraat maken duidelijk dat het gemedieerde fenomeen
celebrity voor entertainment zorgt en op maatschappelijk en publiek niveau een
commodity is die ons moet afleiden van de kapitalistische problemen. Celebrity
(nieuws) focust immers op het private domein en stelt zelden de fundamenten van de
Westerse consumptiemaatschappij in vraag. In dit opzicht is celebrity dus een
commodity, aangezien het de aandacht van het publiek op persoonlijke themas vestigt
en hierbij afleidt van de brede maatschappelijke debatten. Hierbij moet wel worden
opgemerkt dat dit niet betekent dat het (actieve) publiek van celebrities zich zomaar
afwendt van maatschappelijke themas. Het is meer waarschijnlijk dat men andere
informatiebronnen hanteert voor zulke debatten en dat de sociale en culturele waarde
van celebrity voornamelijk op het private, persoonlijke vlak gesitueerd moet worden.
Alle empirische studies hierboven beschreven tonen immers aan dat celebrity allerlei
sociale en culturele rollen vervult in de maatschappij. Celebrity is ten eerste een bron
voor conversaties, die sociale interactie vergemakkelijkt in een tijd waarbij er veel
sprake is van isolatie en eenzaamheid. Daarnaast helpen celebrities om betekenis te
geven aan de eigen identiteit en de wereld door middel van het uitlokken of
vergemakkelijken van morele of sociale discussies over themas in de persoonlijke of
private levenssfeer. Dit is zeker van belang in de hedendaagse geseculariseerde
Westerse samenlevingen, waar de traditionele instituties hun macht verliezen en men
op zoek gaat naar nieuwe bronnen voor betekenisgeving. Celebrities zijn de ideale
invulling voor deze twee eerste rollen aangezien ze bekend zijn bij een grote groep
mensen en men vrij over hen kan praten en kritiek kan geven aangezien er nooit real-
life consequenties zullen zijn door de afstand en het eenzijdige karakter van celebrity
cultuur. Verder kunnen celebrities fungeren als een soort gezelschap, in de vorm van
parasociale relaties, waarbij mensen zich verbonden voelen met hen. In de context van
de groeiende eenzaamheid in de gendividualiseerde Westerse samenleving, is dit zeker
een waardevol aspect van celebrity. Tot slot kunnen celebrities, in de context van het
rusthuis, helpen bij het oproepen van herinneringen en het structureren van de
levensloop, wat bijdraagt aan het welbehagen van de rusthuisbewoners.
De combinatie van tekstueel en publieksonderzoek in dit doctoraat heeft het
mogelijk gemaakt om meer inzicht te verkrijgen in de betekenis van celebrities voor

239
mensen in de hedendaagse Westerse samenleving. Daarom is het belangrijk dat het
publieksaspect van celebrity verder onderzocht wordt in toekomstig onderzoek. Hierbij
is het waardevol om grootschalig systematisch en kleinschalig diepgaand onderzoek te
combineren om een zo compleet mogelijk beeld te creren van celebrity cultuur en de
betekenis ervan voor het publiek.

240