How Hospitality Industry Managers Characteristics Could Influence Hospitality Management Curricula

Sjoerd A. Gehrels CHN University, Leeuwarden, The Netherlands s.gehrels@chn.nl

Abstract: This doctorate research (EdD): work in progress describes the value systems and other driving powers of hotel upper segment restaurant managers (HUSRMs) and the way that these can influence hospitality management curriculum design. The transcripts of the digitally recorded semi-structured interviews were analysed applying grounded theory methodology (Glaser and Strauss) and using NVivo7 QSR software in order to conceptualize on the answers provided by the respondents. The research generated some elements in the ongoing construction of theory about the characteristics of this specific category of professionals and the application in education. By using grounded theory methodology, theory is constructed from the empirical data. The outcomes of the research provided an overview of the divers value systems and driving powers of the respondents. Recommendations were made to have these value systems, driving powers and other characteristics to be taken into account by hospitality management schools in the curricula content and proximaty to the industry. A moderate alert was issued in relation to schools size and format as perceived by the respondents. The research is going to be moved into a next phase that involves expanding the interview sample. Keywords: upper segment restaurants, value systems, driving powers, hospitality management curricula.

1. Introduction
In the hospitality industry and more specifically upper segment hotel restaurants providing hospitality, the enjoyment of high quality food and beverages is viewed as part of a special culture. In this culture hospitableness is one of the central themes. Philosopher Telfer (1996) refers to hospitableness as depending on devotion and a spirit of generosity rather than on skill. Practitioners or former-practitioners in the upper segment, like the researcher, can find reward, satisfaction and opportunities for personal learning and growth in the notion that hospitality and being hospitable carries substantive intrinsic value. Telfer (1996) quotes three historic reasons for offering hospitality in a cultural perspective. Firstly, hospitality originally involved meeting travellers needs before modern means of travel and facilities existed. A biblical sense is referred to here where strangers should be looked after if they are in need. Secondly, giving, receiving and sharing food is a bond of the trust and interdependency set up between host and guest. In some cultures this is a permanent bond between people (e.g. traditional Bedouin will not fight anyone with whom they have eaten salt). Thirdly, giving food (and beverages: note author) is a gesture of friendliness. Particularly where hosts try to give their guests agreeable food it can be considered as an act of pleasing as well as sustaining them. On the other hand however it would be fair to agree with Johnson et al. (2005) that because restaurants operate in a highly competitive business environment, it is often critical for the restaurateur to develop or try to develop a distinctive dining experience for increasingly discerning customers. The philosophy of one of the leading quality hotel chains of the world, Ritz-Carlton, is particularly interesting. This professional organisation, immerses its workforce in the organisational value of considering themselves as Ladies and Gentlemen Serving Ladies and Gentlemen (Henry, 2003). Interestingly, the verb Serving is written with a capital letter in order to refer to a higher meaning of hospitality. The dynamics and special characteristics of the strong entrepreneurial nature of upper segment quality restaurant operations are worthwhile further exploring. Morrison et al. (1999) emphasize the dynamism of the process of entrepreneurship in the hospitality industry and the potential positive outcomes of that process. In the upper segment of the hospitality industry, the quality of service and food and beverages is considered to be paramount. The following question initiated the research: Is it possible to research the upper segment restaurant entrepreneurs and managers in order to distil some of the important characteristics they share and feed this knowledge into the practice of hospitality management education? This research has been designed to generate findings and conclusions that can be used to educate and prepare starting professionals (i.e. students) for the hospitality industry or maybe even other service oriented industries. A fair amount of scepticism towards the possible findings of such a research should always be apparent. It might turn out as a result of this research that the characteristics (values systems and other driving powers)
ISSN 1477-7029 37 Academic Conferences Ltd Reference this paper as: Gehrels, Sjoerd A. (2007) How Hospitality Industry Managers Characteristics Could Influence Hospitality Management Curricula. The Electronic Journal of Business Research Methods Volume 5 Issue 2, pp 37 - 48, available online at www.ejbrm.com

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (37-48)

of entrepreneurs in this restaurant practice are rooted in unethical foundations merely connecting to selfish, purely financial, self-indulging motives that should not be reproduced in education. Knowledge about the characteristics of the successful entrepreneurs and managers in the upper segment restaurants is very limited and merely of descriptive and anecdotic nature. Dominantly journalists have looked at their restaurants (the scenery), their menus and their professional careers (in a quantitative manner; where have they worked and for how long). Not many attempt have yet been made however to look at their specific value systems and driving powers in order to define the grounding principles that brought these successful entrepreneurs and managers in the upper segment to the point they are now. The outcomes generated by this research will provide a contribution to the existing body of knowledge in the field of hospitality management. An even more unique contribution of this research lies however in the theory building on how these entrepreneurial and managerial characteristics can influence hospitality management curricula and therefore the education and development of future professionals for an important industry. The programmes in this sector originated from a strongly industry influenced orientation which Airey and Tribe (2000 in: Lashley and Morrison, 2000) labelled vocational and action in nature as opposed to liberal and reflection. In the past decade however programmes in hospitality management particularly those at the (under)graduate level have been adopting a more reflective, liberal and academic content. While the development of hospitality management curricula is increasingly heading in the direction of academic and reflective modes the risk may be presenting itself that the connection with the hospitality industry and its professional orientation will become looser. Comments to the spirit and manner in which hospitality management education in The Netherlands is offered can be heard among professionals from the industry. One of the concerns they raise is the lack of quality in hospitality (management) education and availability of trained staff. The connection of the curriculum with the professional hospitality world is important and can be phrased as follows as an example in one of the hospitality offering universities in The Netherlands (Bosko, Dekker and Van der Hoek, 2005): our education can be a help to the progress of the professional practice. It is a case of interdependence. In our programme this is personified by lecturers who join us as staff members after a former career in the hospitality industry. The search for the constant factors in hospitality managers actions and attitude have been important for the direction an institute offering a hospitality management programme (Bosker, Dekker and Van der Hoek, 2005). In an even stronger manner hospitality management can be defined as a way of life. A translation from these constants into competencies that guide an educational programme could very well be a leading motive for designing the curriculum. Research into competencies (of successful and representative hospitality practitioners) has been formulated as one of the main points of attention in keeping the educational programme both current as well as legitimate (Bosko, Dekker and Van der Hoek, 2005). The relation between research and curriculum design can be visualized in the following framework.

Figure 1: Relation between research and hospitality industry/competencies/curriculum (modified version of framework Bosko, Dekker and Van der Hoek, 2005) As can be seen in the upper part of the model the relation between the hospitality industry and the competencies derived from it, is presented in the first balloon. From the competencies a (educational) profile is derived that eventually is translated into the curriculum. A major source of influence is research in the first phase applied to the hospitality industry to outline and continuously monitor the competencies needed for

www.ejbrm.com

38

Academic Conferences Ltd

Sjoerd A. Gehrels

practitioners in that industry. A second phase of research is used to keep track what the output and evaluation of the curriculum is in order to modify the competencies definition from that side. An extensive exploration of the value systems and other driving powers that influence successful entrepreneurs and managers in the upper segment restaurants and the influence this can have on curriculum design would be yielding a contribution to the existing body of knowledge in the field of hospitality management education. A better understanding of value systems and driving powers as generated by this EdD research would serve as beneficial for the first phase of problem analysis in policy making as defined in figure 1.

1.1 Definition of terms

The relevant terms in this research and their interpretation are: Hospitality industry = the industry, professional field that is generally considered to be part of the bigger tourism industry. International tourism is expected to increase in revenue from almost 500 billion US$ in 1995 to 2000 billion US$ in 2020 (WTO, 2006). Employment figures showed 207 million jobs in 2000 which was 8,2% of total global employment (UoL, 2006). HUSRM = Hotel Upper Segment Restaurant Managers; the specific, relatively small, group of entrepreneurs/practitioners within the hospitality industry with an significant achievement in operating high quality restaurant operations for more than 5 years and recognized by quality assessing institutions (Michelin, Lekker) for their performance (author) Value systems = the system of values that people have incorporated and from which personal behaviour and actions are triggered (definition author combining different sources) Driving powers = the combination of factors that consciously or subconsciously support an individuals behaviour and actions in life (author) Characteristics of HUSRMs = the combination of value systems, driving powers, sociodemographic variable and biographic notions relating to this specific group of practitioners (author) Hospitality management curricula = educational programmes that prepare students at diploma or degree level in the interdisciplinary subject area of managing service and hospitality offering oriented organisations (author) Grounded theory = the constructionist research methodology (Charmaz, 2006) oriented towards the inductive generation of theory from empirical data that have been systematically obtained and analyzed (Glaser and Strauss, 1967)

1.2 Aim and research questions

The central aim for this research is to define the influence that the characteristics of hotel upper segment restaurant managers (HUSRMs) could have on hospitality management curriculum design. Research questions Which value systems drive HUSRMs in their business and life. Which other driving powers have brought the HUSRMs to performing in their business and life as they do. How can the characteristics of the HUSRMs be used in hospitality management curricula.

2. Literature review
Morrison et al. (1999), looking at the effectiveness and success of entrepreneurship, consider energy, health, emotional stability, intelligence, capacity to inspire and personal and ethical values as desirable but not so acquirable. Some of these characteristics relate to personality traits i.e. energy, health, emotional stability, intelligence and capacity to inspire while personal and ethical values obviously fall within the values domain. Olver and Mooradian (2003) tried to discover and present the relation and possible integration of personality traits and personal values. The former were defined as endogenous basic tendencies tied to underlying biophysical response systems, they are strongly heritable, surprisingly immune to parental and social influences and remarkably stable throughout adulthood. The latter, personal values, are considered to be learned beliefs about preferred ways of acting or being which serve as guiding principles in the life of a person or other social entity. When looking at the relationship between personality traits and personal values,

www.ejbrm.com

39

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (37-48)

Olver and Mooradian (2003) concluded that personality and values both capture distinct and differential characteristics of the individual. They also provide a base of evidence that suggests that personality traits and values are relatively strongly interrelated. Because personality traits are considered to be endogenous variables (nature) they cannot be significantly influenced or modified. Personal values however are part of the nurture elements that constitute a persons behaviour and actions. In this study a strong focus was on personal values that sit within the nurture dimension while in the driving powers elements of personality traits (the nature element) are represented. Tepeci and Bartlett (2002) concluded that workers preference for a particular organisational culture is by far the strongest influenced by personal interests including personal values. For the researcher being a practitioner-researcher in (hospitality management) education it is then, profession oriented, the most interesting in terms of research activity to concentrate on a topic where the nurture (teaching and learning) element can be defined. Lingsma and Scholten (2001) confirm the importance of values for learning, where they refer to McClellands iceberg model describing learning processes of people. Values are considered to be part of the large underwater part of the ice mountain that most strongly influences the learning and changing of people over time. If the underwater part of the ice mountain in peoples learning and changing that contains their values can be influenced, the strongest effect is achieved because it is connected to the will of people to learn or change. The empirical grounding for the assumed relation between values and behaviour is still in progress and to date has not produced a clear cut and fully covering evidenced concept yet. Bardi and Schwartz (2003: 1218), although recognizing correlations between values and behaviour, hold reservation before emphasizing causality. Similarly, Tepeci and Bartlett (2002: 152) make a distinction between situationalist and personalogist streams of thinking. The situationalists hold the view that behaviour can be predicted by assessing the persons situation or organisation while personalogists suggest that personality traits, values and beliefs are primarily responsible for behaviour. Also conflicts between attitude and actions in a particular context may be activated that do not necessarily follow the general values a person holds (Feather, 2002: 447). A wealth of literature has been published on leadership in major companies and multi-nationals and research institutes have been set up to study the subject (Hou, 2001: 262 and Dawes, 1999: 20). In 2000 over 2000 books were published on aspects of leadership (Gillespie, 2001: 167). There is however limited research specifically about value systems and other driving powers of hotel upper segment restaurant managers although examples are present like Morrison (2001) who looked at the biography of two entrepreneurs in the hospitality industry and Legohrel et. al. (2004) examined hotel/restaurant managers, their personality characteristics, attitude towards risk and decisional orientation. Both studies however had no correlation with the upper segment of the hotel and restaurant business. There is no research available on the effect that these characteristics can have on curriculum design in hospitality management education.

3. Methodology
This study is aimed at finding the connection between the characteristics of a particular group of professionals and curriculum design in hospitality management education. It focuses on the core of the hospitality industry. High quality restaurants are a very competitive segment within the hospitality industry that traditionally has been left outside the academic research world with a few exceptions such as Gehrels (2004), Johnson (2005), Gehrels, Kristanto and Eringa (2006) and Gehrels and Kristanto (2007). Looking from an epistemological point of view the orientation in this research is within constructionism (Crotty, 2003 and Charmaz, 2006) following the definition of Flick (2006): constructionism informs a lot of qualitative research programs with the approach that the realities we study are social products of the actors, of interactions and institutions. Knowledge is seen as being constructed in a process of social interchange between the subject (the researcher) and the object (in this case the managers being researched). It is important to define the founding principles for doing this research being in the constructivist tradition. In this way an objectivist/positivist approach is rejected. This type of research would traditionally start from the theory/literature to produce hypotheses that can be tested. The essential philosophy of objectivism/positivism is that research should be value free and objective. In this particular EdD research approach however the results, analysis and theory building generated interacts with the researchers understanding and interpretation. In the encounter with the research sample (the HUSRMs) reality is constructed and not value free.

www.ejbrm.com

40

Academic Conferences Ltd

Sjoerd A. Gehrels

Grounded theory as applied in this research was first presented by Glaser and Strauss (1967) and taken further by Glaser, Strauss, Strauss and Corbin, Locke (2001) and many others. The essence of grounded theory is that theory is derived from data, systematically gathered and analyzed through the research process (Strauss and Corbin, 1998). Clearly this research is done in a very specific context which does not assume grand generalization. The sample in this research consisted of the earlier described specific hotel upper segment restaurant managers (HUSRMs). As presented in paragraph 1.1, HUSRMs are defined as: hotel upper segment restaurant managers; the specific, relatively small, group of entrepreneurs/practitioners within the hospitality industry with an significant achievement in operating high quality restaurant operations for more than 5 years and recognized by quality assessing institutions Michelin Guide (Michelin, 2007) and Lekker Top 500 Restaurant Guide (Nijssen, 2007)) for their performance. The assumption underlying this research is that there must be significant points of interest in the characteristics of this specific upper segment category of managers that could influence curriculum design in education in order to benefit future managers in the professional world. As an important component in the definition of the sample lies in the description upper segment which in this research is connected to two reputable publications in The Netherlands that are considered to set the tone: Michelin Guide and Lekker Top 500 Restaurant Guide. To pinpoint the successful hotel upper segment restaurants and thereby identify their HUSRMs, the following criteria in the two quality assessing guides were taken: 1. Michelin: 4/5 hotels black or 3/4/5 hotels red (H) and possibly star(s) 2. Lekker: quoted (preferably in the top 100 or quoted as being among the best restaurants in their city Amsterdam: BoA) For this sample 6 hotel upper segment restaurants in Amsterdam in the category were taken and their restaurant managers were interviewed. Table 1: Companies of interviewees
Hotel Amstel The Grand LEurope Barbizon Palace Bilderberg Garden The Dylan Restaurant La Rive Caf Roux Excelsior Vermeer Kersentuin The Dylan Michelin 2007 HHHHH Red / HHHHH Black HHHH Red / HHHH Black / HHHH Black HHH Red Lekker 2007 5 BoA 100+ BoA 92 BoA 22 BoA 100+ BoA 70 BoA

Data were generated through semi-structured interviews. This format allows the researcher to ask supplementary questions to attain deeper insights to the meaning of the answers given (Johnson et al., 2005). The interviews were digitally recorded and the recordings were filed into the research project as defined with NVivo7 software to make sure the rich data could be handled.

4. Results and analysis

The six interviews of this first stage of the study were organized between September and December 2006. The interviewing was done by a junior researcher who incorporated the results in her own dissertation research. The respondents were first approached by telephone and when they agreed to participate an appointment was made for the interview. The anticipated time for each interview of maximum one hour was not exceeded. It turned out that the respondents were quite busy people who aimed at a efficient interaction with the interviewer which resulted in an average time of interviewing of between 22 and 46 minutes. The age of the respondents who were all employed as either restaurant manager or maitre dhotel was between 32 and 52. In this first stage of the research we wanted to see if there would be a correlation between the age and years of experience of the interviewees and the length of the interviews. There was no correlation because it turned out to be the youngest interviewee that generated the longest interview while the oldest interviewee generated the second interview duration. The length of the interviews seems to be more related to the preparation and background of the interviewer because in two pilot interviews done by the senior researcher interview times were 1 to 2,5 hours with similar respondents. After analyzing the interviews it became clear that the number of follow-up questions is considerably higher when the researcher/interviewer is really into the topic and background of the research. This notion is confirmed by Arksey and Knight (1999: 104) where they highlight the importance of background knowledge for competent interviewing. Furthermore Rubin and Rubin (2005: 136) stress the crucial nature of asking follow-up questions in order to obtain depth and detail

www.ejbrm.com

41

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (37-48)

in the interview but also acknowledge the difficulty novice interviewers have in working out follow-up questions while doing the interview. The analysis of the interviews was done by reading and re-reading the transcripts that were generated from the digital recordings there by following the suggestion of Easterby-Smith, Thorpe and Lowe (2006: 122) when applying grounded theory approach of systematically analysing the data as to tease out themes, patterns and categories. In order to systematically register the information gained from the transcripts NVivo7 software was used. The interesting elements in the respondents answers were coded in the text and then stored in the so-called free nodes in NVivo7. This procedure allowed the researcher to keep a proper overview of the themes and concepts that appeared. Prior to creating the free nodes a basic structure of socalled tree nodes was designed that followed the main questions put forward to the interviewee. The main questions essentially captured the research questions as they were defined at the beginning of the research. After analyzing the first two interviews the free nodes partly could be fitted into the tree nodes after which the answers created the branches of the trees. For example the first interview question got the interviewees to talk about their education and experience. The tree node initially created for this consisted of education and work experience. After analyzing the first two interviews branches were created such as: secondary education and higher education and after that sub-branches such as general secondary education and vocational secondary education. The creating of more sub-tree nodes generated refinement to the examination of the respondents answering. The free nodes not only generated tree nodes in a predictable direction as given above in the example of education but also led the researcher in new directions exploring the respondents characteristics. For example some respondents recalled particular critical incidents in their life that stimulated their interest for the hospitality industry. These specific new notions captured in free nodes were then transferred into new main categories (tree nodes). The main categories created after looking at the research questions and additionally incorporating the information gained from the detailed reading of the interviews were: 1. Value systems 2. Driving powers 3. Other characteristics 4. Important elements of the profession 5. Influence on hospitality management education 6. Education 7. Work experience 8. Other issues of importance Although education and work experience showed an apparently great diversity, careful examining showed that all respondents had a background in secondary vocational education having been through either the working-learning part-time format (2 respondents) or the full-time MHS (Middle-level Hotel School) format (4 respondents) including one year of internship. Work experience and career paths were all generated in a diversity of hospitality companies (hotels and restaurants) and dominantly for all respondents in food and beverage department of hotels and/or independent restaurants. One respondent (the oldest) indicated to have been working in a variety of other companies outside the hospitality industry. In the category value systems and driving powers the six interviewees were asked to talk about their values and driving powers related primarily to their professional life (Interview questions: Can you tell me something about the things that you value in the profession of restaurant manager in the hotel upper segment restaurants? and What drives you in your life as a restaurant manager in the hotel upper segment restaurants? Could you tell me about your most important values?) We found out that the hotel upper segment restaurant managers held a wide range of value systems and driving powers. In value systems most notably guest oriented, enjoying work and being curious (having an urge to learn) were mentioned. All three values were mentioned by three respondents. The value guest oriented was illustrated as the highest value is probably to be guest oriented and every guest should really feel the warmth of coming home and feeling at home, and should be welcomed in that way. With an even stronger meaning of this guest oriented value one respondent described it as: if guests come to a restaurant it will be under different circumstances and they may be nervous, it is then very important to greet them and make them feel at ease. It is our duty to take away feelings of anxiety. Welcoming and saying goodbye to guests are the most crucial moments. The strong connection the value guest oriented holds to the

www.ejbrm.com

42

Academic Conferences Ltd

Sjoerd A. Gehrels

profession was explained by one restaurant manager as follows: I think that trainees should be inspired by restaurant managers, they should learn the values of the classic cuisine. I want to teach our trainees that the guests are most important and we should fulfil all their wishes, even when you have to be the underdog. Clearly this statement indicates a firm dedication to defining this value of being guest oriented into professional practice. Enjoying work was an interesting value to be picking up from the interviewees because in another category that was touched upon in the interviews respondents indicated that their work environment is challenging. Working a lot of hours (sometimes 10 to 11 hours a day), working at other peoples holidays like Christmas and experiencing stress were mentioned as features of the work. Still three respondents explained that they see enjoying the work as a core value: you have to really, really like the work, I really enjoy myself and I always tell my staff that is of the utmost importance that they have a good working time here. Curiosity as the denominator of the urge to learn was another important value mentioned by three respondents: I learn new things every day and I am curious and read a lot about the business,..I try to keep myself up-to-date. The other values that the respondents were mentioning are shown in the model below.

Figure 2: Value systems grounded in the interviews In the category of driving powers a rather diverse range of issues were mentioned by the respondents. As a stimuli to go into the hospitality industry and into the specific profession some special experiences at young age were mentioned by the respondents: a first experience with an upper segment restaurant on a date with a girlfriend, the hospitable environment created by a grandmother that organized parties and meals, being involved in hospitality work by two girls next door at a young age, helping a brother who was already working in the industry and getting acquainted with the hospitality industry through excursions are all examples of particular experiences that stimulated the respondents to choose for a career in restaurant management. Another category of driving powers lies within the pull factors related to the assumed attractiveness of the hospitality industry and push factors relating to negative experiences related to other industries (or segments within the hospitality industry). Examples were given as: working with the specific products in the upper segment such as wines and good food and the ambiance in the companies within that segment. Also the intrinsic characteristics of the profession have provided the driving powers for some as the examples taken from the transcript illustrate: to organize everything perfectly within the right frames which includes coaching the staff to deliver the products at the right way and time, we are able to set a very high service standard here, which is for me a motivation and driving power to work here, in the end I always liked the service side more, because of the interactions with the guests, but for me this social aspect always has been important and what further more is important as restaurant manager is coaching and to be the host, the person in

www.ejbrm.com

43

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (37-48)

charge to approach and for me to approach the guest, have contact with the guest. One respondent expressed an initial feeling of ambiguity about his profession because of the fact that his father is an economist and he is carrying plates as he described it.

Figure 3: Categories of driving powers In addition to values and driving powers there appeared to be a third category of what we called other characteristics. In this category most quoted were characteristics qualified as values referring to the company. Almost all managers had a specific notion that their work and personal functioning was to a major extent directed by values located in their company. The restaurant managers referred to company related values as: it is very important that I am not the only person who thinks that we are responsible for exceeding the experience of the guest, but the whole team, 12 people, should have this attitude and should feel the same about this, so thats why the others have to have the same kind of spirit in order to deliver the same service. All our waiters should be optimistic, and this is also where we look at when we hire new staff, I really want to be proud of my team, but they have to deserve it, and therefore they have to be willing themselves, I think the standards that we set are very important. These are standards from Sofitel, but also the standards that the management of The Grand makes itself and we are working here with a young team, we do a lot with training and coaching, which I really like. Another characteristic that influences the thinking and acting of the restaurant managers in this sample is their potential ambitions. There is no unity in their ambitions and answers ranged from the traditional line promotion scenario expressed as maybe in a later stage a function in the management of a hotel will attract me as well or within this company I would like to go one step, or two steps, higher, as long as my position still got to do with guests, the sky is the limit. But it is not in my planning to still carry plates when I am fifty years old. Another respondent added the dimension of family life and the restrictions connected to that also in my private life things changed, I have two children of 1,5 and 4 years old, and I think, its also one of the reasons I am more happy with my job, I am not going to switch jobs as fast if I did when I was younger. Obviously some respondents have the ultimate goal of starting some business of their own there is a dream to start a small Bed & breakfast in a foreign country together with my girlfriend which another respondent formulated like I have some goals left, I am not finished in this field yet. I would like to start my own business while simultaneously adding the limitation but I also have a family, my wife and three kids, and they have to be supportive on every aspect which it will take. Other less quoted characteristics were about the fact that they thought skills are not the most important element in their work, the crucial importance of a good attitude and the coincidence of particular steps in their career or even as one respondent stated the absence of real career planning.

www.ejbrm.com

44

Academic Conferences Ltd

Sjoerd A. Gehrels

Figure 4: Other characteristics connected to the respondents Interestingly the respondents related a lot to the elements that they felt are important in their profession as restaurant manager and/or the challenges that should be faced in that same profession. Elements were mentioned such as: having knowledge of people, the profession and the trade, having strong discipline and the willingness to do long hours, being calm in stressful situations and being the example for the rest of the team, directing good organisation and communication within the team, planning both long-term for the operations and financially. Then as the challenges of the profession firstly and most dominantly the number of hours (10 to 11 per day) and the stress incurred by the work were mentioned. A category of answers that we were looking forward to when starting this research was obviously connected to the research question What do you think students of hospitality management educational programmes should learn from your value systems and driving factors? In the analysis of the answers there appeared to be three subcategories that were arranged as: 1. Schools content direction 2. Schools connection to the hospitality industry 3. Schools awareness of structure and format. Then as a separate influence the respondents identified the need for schools to be aware of the individual differences among their students and particularly to acknowledge the fact that some students have some awareness of the upper segment of the industry while others have not. In the first category of schools content direction respondents indicated the importance to both communicate the pre-condition for a career in the upper-segment as well as the specifics of that segment. Positive attitude being a vital attribute was expressed by one respondent as I think it is important for students to realize that you really have to commit to your job or traineeship. A further need content-wise for schools would be to define the professions characteristics and to communicate the positive elements as well as the challenges of the profession to students. One respondents summarized the positive elements as .we are in a beautiful industry, in which we have to transmit that feeling to our guests, colleagues and other peoplethat if you are really enthusiastic about the company you are in, there is future in the hospitality industry, that there are a lot of opportunities to be promoted within a company if you work hard and have the right attitude.that you need a big social sensitivity to succeed in this industry... A colleague added students have to know better that working in this segment is not only about shining gold to emphasize the challenging side of the upper segment in the industry. The second category of answers indicated the strong necessity for schools to provide connection to the industry because people in the industry can provide important learning opportunities to students and as one respondent mentioned in a more philosophical manner I think that it is not only at the operational level, but also for what you are going to do after work and in further life, because it is always going on. If it is correct what you are doing here, you are also going to do well outside the company. You will take this further in decisions that you make in the future. Three respondents emphasized the importance of the practical

www.ejbrm.com

45

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (37-48)

experience for students in internships or through other activities for their awareness of the upper segment in the hospitality industry and of life in general. An explicit concern was raised by some respondents about the structure, format and size of the schools and in comparing the MHS colleges to the HHS degree programmes concluding that the former provided a much better foundation for the practitioners in the upper segment than the latter.

Figure 5: Influence on hospitality management curricula At the end of the interview the respondents were asked to raise any other issues they felt to be of concern for the topic. The issues raised were about the greater difficulties of restaurants in the upper segment to get good staff, emphasizing the importance of teachers in hospitality education to treat their students as guest in order to raise their awareness of hospitality and a re-confirming of the importance for students to be proactive and take the learning opportunities they get in many companies.

5. Conclusions and recommendations

Conclusions and recommendations in this chapter are interwoven, while at the end a summarizing set of recommendations is offered. When looking at the results of this study it is fair to say that the six restaurant managers, all male, were in their mid-careers and all but one between 32 and 42 years old. The range of values the six managers felt to be important was quite varied and further categorization would be useful. In the 10 value categories (Schwartz, 1992, Bardi and Schwartz, 2003: 1208) conformity (guest oriented, highly disciplined), self-direction (curiosity) and hedonism (enjoying work) would be represented although achievement (pride in work, high motivation, striving for perfection, striving to be the best as a company, passion for the profession, being respected) was also highly represented. It is difficult to offer very extensive explanation for the appearance of these value categories at this stage of the research. The high representation of value categories achievement and conformity may be related to the career paths and educational levels of the restaurant managers. They all went through a relatively long and hard working career development having educational achievement at secondary vocational level complemented with many years of working through the ranks. It seems that achieving the best in their work and conforming to the strict quality setting of their work environment (upper segment restaurants) has brought them to the position of restaurant manager. In the forward feeding of their values to hospitality management education this could be important information for future managers (now students) because it

www.ejbrm.com

46

Academic Conferences Ltd

Sjoerd A. Gehrels

explains the nature of career development in a labour intensive (10-11 hours per day) industry upper segment. It is then simultaneously interesting to observe the mentioning of enjoying the work as a hedonistic value category although this enjoying the work could also be explained as a sense of conformation (one has to like the endurance element incurred by the profession in order to sustain). In the driving powers for these professionals it is interesting to see that special experiences in a life before the profession in some cases led to the initial interest setting. The restaurant managers were sensitive to social settings in their youth, sometimes in a context with family or friends, that stimulated their interest in the world of professional hospitality offering. The driving powers of pull (feeling attracted to the specific industry or segment) and push (disliking other industries or segments) have most likely further channelled the interest of these professionals. Taking the particular appreciation for the intrinsic qualities (elements of the profession) as further driving power supports the process of further development the respondents have gone through in their careers in the upper segment. A preliminary conclusion drawn from the driving powers that the respondents talked about for hospitality management education would be to further investigate and analyze the pathway that potential students have gone through before they opt for professional development in the hospitality industry in order to see which particular personal driving powers they have. Furthermore an important responsibility would be put on schools management, curriculum designers and educators to explicitly be aware of the responsibility they have to offer a realistic but also inspiring learning experience related to the profession. This would entail having students to learn about both the possibilities as well as the constraints of the professional hospitality world. Another important conclusion on how the respondents characteristics should influence hospitality management is about the ongoing need to emphasize the schools connection to the industry. Gaining practical and real-life experience thereby facilitating the learning opportunities provided in companies and eventually understanding the connection of professional life to the world around it. Very importantly the awareness issue was raised by the respondents where they talked about their concern for the growing sizes of the schools that they knew before as their vocational and professional home grounds. Although the increase in scale may be a necessary aspect of contemporary sustainable organizational formats in education, it would be good for schools to consider how they organize and present this to the professional world. A necessity for better alignment between schools educating at the secondary vocational level and those at degree level could also be concluded from the respondents answers. Finally the other issues that were brought up by the respondents reconfirmed the suggestions about educations responsibility to properly and realistically prepare students for the hospitality industry while acknowledging the role that companies can have in that process. When looking at the concepts that were emerging from the interviews a first construction of theory can be undertaken. The way a management position in the hospitality industry i.e. in restaurant management is fulfilled and the degree of success in fulfilling it is correlated to a combination of factors. In a formula it is presented like this: Eex + Edu + Wex + Cha = Pos Eex = Early experiences relating to generating a hospitality awareness. Edu = Education Wex = Work experience Cha = Characteristics of the particular professional (value systems and driving powers) Pos = Position fulfilling and success If in further research this theory can be validated, confirmed and detailed it will provide an important instrument for educational programmes and the field of human resource management. The theory could support the assessing of choices of future professionals (student: before-, throughout- and after their educational programme). A recommendation from this first phase of the research would be for educational programmes to seek a closer relationship with the professional world i.e. the industries that they are primarily educating in order to feed in the actual nature, character, quality and challenges of that professional world. Another recommendation would to further develop an ongoing basis in research into the industry as presented in

www.ejbrm.com

47

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (37-48)

figure 1, thereby increasing the proper knowledge base connected to the hospitality industry and its influence on curriculum design.

References
Arksey, H. and Knight, P. (1999), Interviewing for Social Scientists, Thousand Oaks: Sage Publications. Bardi, A. and Schwartz, S.H. (2003), Values and Behaviour: strength and structure of relations, in: Personality and Social Psychology Bulletin, vol. 29, nr. 10, pp. 1207-1220, Society for Personality and Social Psychology, Inc. Bosko, F., Dekker, J. and Van der Hoek, K.W. (2005), Profiles in Hospitality Management, Leeuwarden: CHN University Netherlands, International Hospitality Management. Charmaz, K. (2006), Constructing Grounded Theory: a practical guide through qualitative analysis, London: Sage Publications. Dawes, F. (1999), Small Business Management: an overview, Dublin: Blackhall Publishing. Easterby-Smith, M., Thorpe, R. and Lowe, A. (2006), Management Research: an introduction, London: Sage Publications. Feather, N.T. (2002), Values and Value Dilemmas in Relation to Judgements Concerning Outcomes of an Industrial Conflict, in: Personality and Social Psychology Bulletin, vol. 28, nr. 4, pp. 446-459, Society for Personality and Social Psychology, Inc. Flick, U. (2006), An Introduction to Qualitative Research, 3rd edition, London: Sage Publications. Gehrels, S.A. (2004), An Investigation into Small Upper Segment Dutch Restaurant Owner/Managers Value Systems and their Effects on Business Actions and Performance, Assignment for EdD Programme University of Stirling, Leeuwarden. Gehrels, S.A. and Kristanto, S. (2007), Marketing of Dutch Culinary Restaurants: An Exploration from the Entrepreneurs Perspective, in: Journal of Culinary Science & Technology, Vol.5, Nr. 2/3, New York: The Haworth Hospitality Press Gehrels, S.A., Kristanto, S. and Eringa, K. (2006), Managing Word-of-Mouth Communication in Michelin Starred Restaurants in the Netherlands, in: Jurnal Manajemen Per Hotelan, Surabaya: Petra Christian University Glaser, B.G. and Strauss, A.L. (1967), The Discovery of Grounded Theory, Chicago: Aldlin. Henry, B. (2003), Quality Customer Care for the Caribbean, Kingston: Customer Service Academy of Jamaica Limited. Hou, W.C. (2001), The Inspirations of Tao Zhu-Gong: modern business lessons from an ancient past, Singapore: Prentice Hall. Johnson, C., Surlemont, B., Nicod, P. and Revaz, F. (2005), Behind the Stars: A Concise Typology of Michelin Restaurants in Europe, in: Cornell Hotel and Restaurant Administration Quarterly 2005, vol. 46, nr. 2, pp. 170-187. Lashley, C. and Morrison, A. (Ed.) (2000), In Search of Hospitality: theoretical perspectives and debates, Oxford: Butterworth-Heinemann. Legohrel, P., Callot, P, Gallopel, K. and Peters, M. (2004), Personality Characteristics, Attitude Toward Risk, and Decisional Orientation of the Small Business Entrepreneur: A Study of Hospitality Managers, in: Journal of Hospitality & Tourism Research, Vol. 28, No. 1, February 2004, pp. 109-120. Lingsma, M. and Scholten, M. (2001), Coaching for Competencies Development, Soest: Nelissen Publishers. Locke, K. (2001), Grounded Theory in Management Research, London: Sage Publications. Michelin (2007), Michelin Guide Netherlands 2007, Zellik: Michelin Travel Editions. Morrison, A. (2001), Entrepreneurs transcend time: a biographical analysis, in: Management Decision, 39/9, pp. 784-790, MCB University Press. Morrison, A., Rimmington, M. and Williams, C. (1999), Entrepreneurship in the Hospitality, Tourism and Leisure Industries, Oxford: Butterworth-Heinemann. Nijssen, G. (2007), Lekker 2007: The Restaurant Guide in The Netherlands, Amsterdam: Multi Magazines Publisher Olver, J.M. and Mooradian, T.A. (2003), Personality Traits and Personal Values: a conceptual and empirical integration, in: Personality and Individual Differences 35, pp. 109-125, Pergamon, Elsevier Science Ltd. nd Rubin, H.J. and Rubin, I.S. (2005), Qualitative Interviewing: the art of hearing data, 2 edition, Thousand Oaks: Sage Publications Schwartz, S.H. (1992), Universals in the Content and Structure of Values: theoretical advances and empirical tests in 20 countries, in: Advances in Experimental Social Psychology, vol. 25, 1992 Strauss, A. and Corbin, J. (1998), Basic of Qualitative Research: techniques and procedures for developing grounded theory, 2nd edition, Thousand Oaks: Sage Publications. Telfer, E. (1996), Food for Thought: philosophy and food, London: Routledge. Tepeci, M. and Bartlett, A.L.B. (2002), The Hospitality Industry Culture Profile: a measure of individual values, organisational culture, and person-organization fit as predictors of job satisfaction and behavioural intentions, in: International Journal of Hospitality Management 21, pp. 151-170, Pergamon, Elsevier Science Ltd. UoL - University of Lincoln (2006), Tourism: a global industry, in: Brochure Department of Tourism and Recreation, www.lincoln.ac.uk/dtr/Tourism.htm Accessed on: 25-07-2006. World Tourism Organization WTO (2006), Tourism in the 21st Century: engine of world growth, in: La Caixa, The Spanish Economy Monthly Report, www.pdf.lacaixa.comunicacions.com/im/eng/200606af1_eng.pdf Accessed on: 24-07-2006.

www.ejbrm.com

48

Academic Conferences Ltd

Research Dilemmas in Management and Business Studies

John Mendy University of Lincoln, UK jmendy@lincoln.ac.uk
Abstract: This paper examines some of the philosophical and practical dilemmas that are faced by researchers in management and business studies in the context of the epistemological and ontological assumptions introduced. The relevant methodological frameworks to be used stem from Strauss and Quinn (1997). The importance of employee language and organisational discourses are presented from the empirical data on Aspects of Organisational Culture and Change in Lincolnshire and Nottinghamshire to demonstrate the epistemological and ontological dilemmas faced by researchers in organisational analysis. Symbolic interactionism and stories are also used to highlight the importance of speech actors within an organisational change context in order to surface some of these dilemmas in business studies in general and management research in particular. These two provide alternative positions to Strauss and Quinns maintained analysis of how organisations and managers implement change interventions and employees reactions to these. Amongst some of my principal objectives is to demonstrate what can be contributed when researchers focus on what can be considered credible and valid knowledge that can be generalised in organisational and management studies. Keywords: language, methodological dilemmas, epistemology, theory and practice.

1. Introduction
This paper focuses on the exploration of some of the philosophical and practical dilemmas faced by researchers in business and management sciences. I examine Strauss and Quinns (1997) methodological approach to the study of organisational and management practices to point out some of these dilemmas especially in the context of how employees experience organisational culture change. I question the rather structurally maintained position adopted by these scholars in deepening researchers understanding of individual and collective, organisational behaviours in similar contexts. I use Magalas (2005) theorising and conceptual analysis of organisational cultural changes to critique Strauss and Quinns position. Some empirical case study data from my initial findings on four organisations situated in two English counties have been used to demonstrate some of the epistemological and ontological dilemmas that such structural approaches to studying dynamic and complex phenomena as organisational cultural change could pose for researchers. This paper also explores what other research approaches such as stories and symbolic interactionism could contribute in resolving some of these dilemmas of research choices.

2. Strauss and Quinns (1997) approach

In this section, I present the properties of Strauss and Quinns theorising to see what these can contribute to the papers focus. The second part of the discussions concentrates on Magalas work as the basis to criticise Strauss and Quinn and to bring out the richness and complexity in organisational cultural change studies as well as the dilemmas posed in their exploration.

2.1 Properties of Strauss and Quinns theorising

Strauss and Quinn (1997) postulate that cultural meaning can be studied through an exploration of the extra-personal and the individual intra-personal positions. This seems to suggest what happens outside of organisations as well as what goes on in between personal boundaries can account for peoples cultural values. In my current research into Aspects of Organisational Culture and Change I focus on how employees react when their organisations change their working practices and how these can be used to explore some of the dilemmas posed to researchers in business and management studies. To identify employees experiences of cultural change I use the notion of presence to examine peoples contributions to the workings of their organisations-in my case, the extent to which employees are willing to participate in organisational cultural change interventions. I am interested in exploring the possible shift in employees presence in terms of how the dividing line between the personal and organisational may be moved to the advantage or disadvantage of one of the participating parties such that the experiences of the personal or the organisational may be enriched or impoverished by this movement. This exploration will be compared with Strauss and Quinns theorising especially on the level of maintained cultural values between managers and staff to determine to what extent the epistemological and ontological dilemmas in business and management research may be expanded using the collective experiences of employees in 2 regions in the UK faced with similar or varying cultural change issues from the ones Strauss and Quinn theorised on.
ISSN 1477-7029 49 Academic Conferences Ltd Reference this paper as: Mendy, J. (2007) Research Dilemmas in Management and Business Studies. The Electronic Journal of Business Research Methods Volume 5 Issue 2, pp. 49 - 60 , available online at www.ejbrm.com

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (49-60)

What I aim to do is look mainly at the dynamics of how the personal and organisational are distinguished and played out. The initial data capture forms of behavioural changes, new ways of organising and how the deep symbolic interactions between organisational members have brought about a wide range of employees reactions and experiences especially when managers impose changes on other employees. In order to surface a wide range of these employee interactions inside and outside organisations and some of the experiences that these relationships could generate, I asked the following guiding research question: What are the consequences on employees and others when organisations change adopting totalising tendencies in their efforts to improve efficiency and performance? In this regard, not only are the internal and external relationships that shape the interactions between organisational members crucial as Strauss and Quinn maintain, but employees choices of language to describe such interactions also becomes vital, an aspect that has not been fully recognised in Strauss and Quinns theorising. Different levels of employees languages have been identified in the chosen responses highlighted in this paper so as to demonstrate some of the dilemmas that different levels of employee and organisational discourses pose in researchers claims to gaining, what could be considered by those who do not use such an approach, as credible knowledge of organisational cultural change. The dilemma that Strauss et als position poses me is to focus more on how the intra-personal and extrapersonal interplay has brought about employees preferences of engaging in certain interactions, certain cultural changes and making use of certain forms of language and symbols and not others. Both Strauss and Quinn (ibid) went on to postulate that such interactions have been created and maintained to produce organisational culture. To my mind, such maintenance could be regarded as transient relationships that evolve over time, dependent on which network of relationships and, by implication, which world of change employees wish to engage in. Organisational culture as managerial theory (such as Morgans (1986) and Scheins (1985) shared perspective to studying organisational culture and change) seeks to enshrine the need for all employees to work towards the attainment of collective organisational goals. This blurring of the boundary between the personal and organisational is similar to Strauss and Quinns maintained position. In management literature this blurring signifies that employee and organisational presence (in the form of contributions) could bring about an enhanced organisational performance in which everybody is willing to participate in the fulfilment of tasks and functions. It is being proposed in this perspective that the presence of all employees guarantees organisational effectiveness. Such theory seems to deny that the boundary between the individual and organisational presence can be out of balance for any considerable period of time, and to propose that managers and employees eventually are bound to see each other as fellow human beings whose self respect, self esteem and personal identity need to be safeguarded. No special form of transcending dominance of either the personal or the organisational is deemed necessary, therefore. The theory explains such dominance as temporary only.

2.2 Studys initial findings

The four organisations contacted to participate in the study were Longhurst Housing Association (LHA), Laurens Patisserie, Prospects and Eden Supported Housing Ltd. At the time of writing the paper the names of some of these organisations may have changed. The 51 interviews took place between January 2004 and July 2005. The interviews were conducted on the basis of 18 questions. The questions were designed to provide data that might demonstrate the dynamic nature of the relation between different roles in organisations that experience stressful changes. The categories of interview questions concentrated on management traits, and highlighted vision/mission/purpose and meaning, cultural and moral beliefs and values which also featured dimensions of empowerment and the extent of employee participation within the change processes. The questions were open. To facilitate the analysis, the responses of the interviewees were transcribed and coded such that comparisons might be facilitated. In the analysis I re-constructed the experiences of managers by way of the notion of a platform, or world, in which managers attempt to colonise employees. Next I re-constructed a platform from which employees react to the colonising or imposing managers. The constructions are based on the responses of the interviewees. The interviewees were asked to validate the re-constructions, and modify them where necessary. The results were used in the analysis. The analysis of the findings raises the importance of employee presence (i.e. an employees contributions to the functioning of an organisation) either through experiences and reactions of maintaining it, increasing it

www.ejbrm.com

50

Academic Conferences Ltd

John Mendy

or reducing it by resisting against change. Therefore, this study has highlighted that the consequences of organisational cultural changes on employee behaviours could not have been maintained and sustained by top-down managerially imposed approaches as the behaviours of such colonising actors have been resisted by employees through language and other less-overt reactions discussed in the paper. For example, an employee had this to say about his/her organisation: Management have some moral values but I have little to do with higher management. They do care about whether I am doing my job properly. I think senior managers do care but this gets diluted a bit further down the line, say at middle management level. Its a different ball game at middle management level. However, I must point out that junior support staff do not feel valued. They confide to us trainers and do not want this taken further up the management line. Support staff feel overworked and underpaid. If their line manager is not treating them as they wish then you understand how they feel. From the above respondent, it does appear that the employees choice of language that bears resonance to the individuals cognitive representations of personal and collective experiences mark a significant departure from Strauss and Quinns (1997) body of theorising. To overcome and identify such emerging employee resistance it was attempted to manipulate the middle managers, and change their values from the group values to the managerial values such that middle managers would reinforce some of the growingly mechanised forms of managerial colonisation and communication with employees. Methods to do so included for example having staff members constantly report to line managers and supervisors (in some cases) about whether they were meeting deadlines, production and delivery targets. Adherence to the managerial type of organisational culture was reinforced through constant control and supervision thus starting to change them into total institutions (maintained according to management command and control mechanisms) such that employee presence in the boundary between the personal and organisational became blurred and impoverished (through employees withdrawal of presence) during the shift in behavioural emphasis. Such control procedures exacerbated employee stress levels. To meet customers demands employees had to work longer hours, which curtailed their work/life balance and quality of life. The resulting outbursts of emotions (manifested in the use of some unsavoury language) demonstrated that staff started to become demotivated and frustrated. Reactions included a reduction of presence (people started to fall ill or take days off) or withdrawal (staff became disengaged from participating) as well as an increase (employees started to spend time creating alternative communication channels resulting in culture jamming (Dery, 1993) and pockets of resistance, based on sub-cultural values). In one of the organisations the tendency to limit presence to small groups was strengthened by the immigration of relatively large groups of mainly Eastern Europeans. They introduced different work ethics by continuously striving to work long, shift hours and a heightened perception of (sub)-cultural and group identity. As part of this development, the cultural divide between managerially maintained and non-managerial positions also started to grow as was the case between the small emerging groups and other pockets of identity formations and consolidations. Such a social and organisational phenomenon differs from Strauss and Quinns maintained position. Managers were being drawn into a strengthened blame culture, and employees started to change their expectations and reduce their interest in the overall development of their organisation and, hence, in proper and sufficient communication between groups with different roles and tasks. Rumour mongering and a recluse to group attachment became the norm in some instances. Various sub-cultural identities developed in a process of isolation, department and job site alienation and breakdown of the central chain of communication. A clear indication of the emerging process of fragmentation was the fact that many employees no longer appeared to know their mission statement. For example, staff in the care industry (one of the four organisations I studied) reduced their presence by increasingly becoming more interested in their pay packages and other personally remunerating mechanisms than in the caring values management and shareholders were expecting them to espouse prior to and during the introduction of change interventions. Little or no effort appeared to be spent to develop such values which Strauss and Quinn purport to be based on maintained interactions. A culture of resentment, disengagement and outright resistance was allowed to grow, in defiance of the official, apparently integrative organisational culture of team-working, collegiality, care for staff and the desire to meet the quality standards through employee commitment and contributions. Expectations of future personal development disappeared. Staff even showed themselves ignorant of

www.ejbrm.com

51

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (49-60)

management plans which were expected to foster an integrative approach to cultural change. Their world (presence) clearly had shrunk considerably in scope and participation and contributions to organisational cultural change had become reduced or withdrawn with some employees referring to a them and us dichotomy. Managers also changed their values, and reduced their presence although in a different form and apparently with less regret than non-management staff. They started to think in terms of power and control and of manifesting their authority such that they became more and more colonising of employees contributions ushering in what was similar to a totalising institution. New disciplinary and intermittent punishment procedures served to distance managers from other employees, even though the new organisation was intended to be based on cooperation and integration of cultural value systems. In this sense, the managerial world seems to become mechanised and robot-like, and to increasingly lack communication structures beyond command, control and authoritative power frameworks. Managers lose presence, respect from staff and individuality and no longer appear able to behave according to longer term strategic orientations. Unfairness creeps in with some employees referring to what has evolved to be the upper group who enjoy special privileges such as longer holidays, friendlier shift-work patterns and so on. Proposals to create an employment tribunal may be seen as attempts to re-establish secondary communication processes with distinctive sets of language and value orientations, be it only in a formal sense. Other attempts may include the introduction of new computer-based technology, clocking/signing-in and out of work and other routine-like reporting mechanisms. These technological developments did not appear to stop the fragmentation of communication, or prevent the increasing absence of secondary, noncommand communication channels such as back or outside-office talk and other informal modes of sharing employee expectations about work. Employees tended to underestimate the speed with which the latter needed to be re-established. Secondary communication processes may include information about the task of an organisation and its engagements with other organisations within the area but will also include exchanges about better wages, friendlier hours of work and more favourable rota systems. If this type of communication is missing, other content will develop, such as rumour-mongering, not turning up for training sessions, absences and other forms of letting management know of grievances and a reduction or withdrawal of employee presence. It appears that their introduction has to be minimally part of the change process itself. The changes in the internal communication obviously affected the outside communication. Where previously there appeared to have been a relatively warm relationship with shareholders such as Job Centre Plus, Business Link, Learn Direct, Trade Unions, Learning and Skills Council and Connexions, this changed with the increasing need to sell the services of some of the organisations to their external communities. Relationships became colder, more formal and even conflicting in some cases as some of these organisations were not able to adapt to changes in their local environment, but also because managements were seen to assume an air of superiority, resulting in conflicts that were new to the organisations and their communities. Such phenomena contradict to a great extent Strauss and Quinns theorising on how organisations culturally reinvent themselves and how the revolving aspects and issues can be studied. Strauss and Quinns theoretical analysis of organisations and society have, on the whole, been premised on the understanding that societies and organisations can be perceived and studied as stable and integrated entities, which they are not. This realisation problematises the maintained perspective developed by Strauss and Quinn and seeks to explore a deeper and richer understanding of organisational studies through employees lived experiences and life-worlds (Schutz, 1967). These platforms or worlds range from the imposition of managements integrative values to the friendliness, cooperation and resistance that this has brought about in employees. In order to try to deepen my understanding of the complex range of employee experiences and possibly resolve some of the dilemmas identified in this study, I propose to examine Magalas (2005) analysis of organisational cultural changes and the importance of language, among other things, to see what these can contribute to business and management studies research.

3. Magalas (2005) approach

Since, according to Magala (2005), we do not manage dispassionately I question the rationality of Strauss and Quinns (1997) body of theorising. The former recognises the crucial importance of emotional intelligence (EQ) and other approaches as complements of the cognitive processes through which employees languages are expressed. The human cognitive process suggests that all beings (members of staff within an organisation) are capable of rationalising. Hofstede (1980), like Magala, highlights the

www.ejbrm.com

52

Academic Conferences Ltd

John Mendy

importance of the mind and cognition. Living (and, by implication organisational) entities can therefore be deemed to be subject to mental software programming. Reactions of individuals to organisational cultural change and to the instructions of managers are due to enshrined social patterns, to the norms, values and deep-seated beliefs that result from their enshrinement. This suggests that the differences between managers and staff during organisational change express such norms and values through various forms of language, stories and symbols. These communication forms are not necessarily maintained and mutually agreed upon by all participating organisational members. Magala (2005) examines various forms of communication carriers such as the media (television, video, computer games, mobile phones), our use of languages, accounts and coded behaviours as embodied in bowing, the shaking of hands and so on. All of these symbolic communication processes could account for the tacit curricula through which members of an organisation are socialised thus underscoring the multiple processes and cultural, sub-cultural and behavioural differences between organisational members. Such differences cannot be simply identified within an integrative and maintained framework as proposed by Strauss and Quinn but through a recognition framework that recognises the myriad of employees cultural norms and values that constitute their identity and give meaning to their interactions with one another. The influential work conducted by McLuhan (2001) argues that new communication styles such as print, television, i-pod, eye-bites, mobile-forms of working and so on will, of themselves, alter our socialisation and cultural processes and thus create more organisational cultural differences and shifts than Strauss and Quinns (1997) maintained values perspective. Magalas (2005) exposition of the cultural revolution from the oral to the written forms, the coding and recoding of writing from Platos era to Gabriels (2000) notion of narrative deskilling emphasise cultural transformations and shifts over time and the massive accompanying implications on peoples cultural values and behaviours. Such richness in the diversity of theoretical perspectives, and some of the practical implications thereof, is more comprehensive and exposes the complex dilemmas than Strauss and Quinns (ibid) rather rule-like generalisations of employees behaviours and cultural norms and values. The richness of the theoretical discussions on how organisations culturally redesign themselves is further enhanced by the empirical data on how employees experience such changes. However, these propositions to theorising and understanding organisational cultural changes pose renewed dilemmas for organisations and managements. For example, Hofstedes (ibid) cultures consequences and the emphasis on language can be regarded as uniting networked and intermingled employee and organisational experiences and the memory that this produces with a fast growing multi-media network such as text messaging, emailing, cell-phoning and so on. This intermingling results in a new medium of communication between different individuals interacting within specific contexts, some of which are nonstationary. The full impacts of these evolving forms of interaction, ways of organising work and the communication processes through which these are surfaced, are increasingly computer-aided and the cultural values that they inadvertently create have begun to reflect on changing employee behaviours thus creating paradigmatic schools of thought (Kuhn, 1970) with massive ontological influences on how project teams, task and role cultures (Handy, 1993) operate in a growing number of organisations and businesses. This suggests that members of an organisation performing different roles and functions view and present their experiences according to how these have been influenced by their social, cultural and ethnic backgrounds and relationships with those inside as well as outside of work, amongst others. Magala (2005) proposes that the cultural consequences will be as massive as those that occurred when oral communication gave way to the written word during Platos era. Magala (ibid) states We are only just beginning to understand the increased risk accompanying outsourcing and offshoring and the general retreat from large bureaucratic structures with clear chains of command and responsibility and tested roles in social and political systems (Magala, 2005). Therefore, reality does not exist out there, rather it is the result of what has been co-created between speech actors, experiences (of people on whom changes are being imposed upon by colonising managers) and according to Magala the mass media that brings it to the fore and reports it. This is by way of identifying and reporting what happens in the interactions between organisational members in terms of how employees innate cultural value systems may be influenced by what happens inside as well as outside of organisations. Such a position recognises parallel networks of relationships that are happening simultaneously. Thus media events and personal experiences that are reported in them should not simply be regarded as objects in daily life such as moving the world into cyberspace but simultaneously co-created between various individual speech actors with varying experiences and cultural values. Such differences in reality can also account for

www.ejbrm.com

53

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (49-60)

part of the epistemological dilemmas that researchers who conduct snap-shot case-study analysis of organisations face. Part of this problem has to do with attempting to generalise their findings to other organisational or social contexts. However, such researchers as myself, do not necessarily claim that their findings can be generalised. I consider the richness of the responses with regard to employees complex behaviours as crucial in my study. Such richness is captured in the variety of employee responses that portray the dynamic interactions between organisational members experiencing cultural reorientation. Language is also a central part in accounting for cultural differences because it is through employees uses of language and the extent to which they maintain, increase or reduce their presence during organisational change that different spectacles for perceiving the world are created. Presence provides the study with the variety of behaviours (at the personal and organisational level) in respondents statements and stories of experiences of organisational culture changes such that an increase in employees contributions brings about enriched lives or, alternatively, employees may choose to reduce their organisational contributions such that their experiences of cultural change have been impoverished. This theoretical and methodological orientation differs from the maintained position adopted by Strauss and Quinn. The importance of this visioning and imaging of the world (in terms of how employees view their world) to facilitate our communication processes and cultural identities makes some forms of communication possible while rendering others more complex and varied. Such varieties and complexities of employee behaviours and perceptions which I have identified and analysed in employees languages and presence can be used to understand the idiographic nature of organisational culture and change. I also question Strauss and Quinns (1997) claims of maintained employee behaviours especially within organisations experiencing changes in cultural values. This position would appear to deny employees abilities to not comply with managements maintained corporate value sets and behaviours that they expect staff to emulate. This position also seems to deny employees resistance to managements imposed values in efforts to change. Such resistance manifested the overt and more subtle forms of some management and non-management staff not complying with the decision-making process. The study has identified employees who fall within this categorisation in the data analysis. Strauss and Quinns (1997) structural, functionalist approach could be further critiqued in the sense that the variety of subjective employee opinions, emotional experiences, cultural backgrounds and different ways of seeing and representing the world present inherent conflicting demands on managements, organisations and, possibly, individuals desire for order, stability and integration. Such conflicts and tensions emerge when these are least desired by managers who wish to see speedy and cost-efficient implementation of change interventions in their desire to maintain organisational performance, quality standards and profits, amongst other objectives. As employees interact with one another, there could arise what Durkheim (1938) proposed as collective consciousness which, could consequently restrict the sustainability of the argument regarding individually maintained cultural norms and values. Within this scenario individual and group culture should feature prominently in organisational analysis. Such a combination and the issues that arise from their interaction is what I identified in the notion of presence as reflected in the contributions people make within the realms of the personal and organisational. This critique poses the question about the problem of cultural order and the stability of employee behaviours in the sense that what might be regarded as cultural consensus, the acceptable status quo, the integrated way we do things around here (Deal and Kennedy, 1982) by a group of individuals such as management could be perceived by other individuals such as employees, for example, as totally unacceptable. This therefore raises the notion of conflict of interests and mismatch of values. These emerging themes have been highlighted from respondents accounts and observation statements (depending on which platform they choose to engage in) during the interviews. Furthermore, the maintenance, or at times, the changing of cultural norms and values could be shaped by what has been negotiated and mutually agreed by and between employees. This metaphorical concept implies that organisational members are in a constant process of organising and reorganising with one another (thus changing), interpreting and reinterpreting organisational values and norms so as to review what would appear to be their shared cultural expectations and actions over time. Culture can therefore be viewed as an emerging phenomenon, not by any means a static given awaiting study. Tendencies to impose as well as to resist or be friendly are part of a conversation, a discourse or negotiated exchange that allows for proper functioning in daily life, but also for the possibility of more fundamental changes, e.g. in case of market forces or other possible external drivers to organisational cultural change. Negative effects, such as subversion, jamming and imposition, have to be seen as part of an emerging discourse, therefore: they signal through the emerging platforms that something needs to be changed or redesigned.

www.ejbrm.com

54

Academic Conferences Ltd

John Mendy

The language of imposition and totalising as well as that of resistance, conflict, tension and cooperation could therefore be perceived within a transient context which depicts how employees view their definition of the situation at a given point in time. This element of change denotes that the expected roles being performed within the organisations studied could evolve depending on what has been mutually and implicitly agreed within and between groups and statuses as the appropriate ways of doing things such as complying with or resisting against management demands to change working practices. It seems to me that this element of change counteracts the argument that organisations are concretely stable, maintained and coherent entities and could be theorised and analysed as such without identifying and appreciating some of the dilemmas that such a nave position poses.

4. Alternative research approaches

In this section, I present 2 approaches to see what they can contribute to unravel part of the dilemmas faced by researchers in business, management and organisation studies. These are symbolic interactionism and stories. The latter is presented sometimes in the form of accounts and, at other times, in the form of observation reports from the interviews that I conducted with 51 employees in 4 organisations. The organisations were selected because they were going through tremendous alterations in their working practices at the time of the research.

4.1 Meads (1964) symbolic interactionism

Symbolic interactionism refers to a perspective in sociology (Mead, 1964; Blumer, 1969) that inspires researchers to focus on how individuals experience their environment, rather than on the general conditions that shape the latter. It suggests exploring peoples subjective definition of the situation, and the way this definition is negotiated through different interactions and to link this definition to a collective characterisation of the interaction. I use the notion of account as it is used by symbolic interactionists as explained by (Gergen, 1991): Accounts of the world...take place within shared systems of intelligibility -usually a spoken or written language. These accounts are not viewed as the external expression of the speakers internal processes (such as cognition, intention), but as an expression of relationships among persons (Gergen, 1991:78). I also find these accounts of employees experiences highlighted the numerous sets of possible meanings (from employees languages) that were being created in the interactions between different employment statuses. These included how top management, middle management and employees view their worlds. These accounts were also facilitated because of the interaction between the interviewees and myself in the course of the data collection process. From the findings, managers and employees offer these accounts as a way of identifying with their organisations, their sub-cultural groups, choosing which changes to cooperate with or not and so on. Viewed alternatively, they regard themselves as principal parties in comprehending, but, at times contesting (if needs be) the inequalities expressed in the lay theories of organisational culture and change as those of Schein (1985) and Handy (1993). Managers presented themselves to the researcher as responsible and caring leaders that were interested in their employees welfare. In their perception, such characteristics of management styles make them appear more fashionable. Thus, the accounts they offered of their management behaviour could be influenced by their anticipation of what their roles were supposed to be or what they intended I might have assumed their roles to have been. They appeared to denote an integrative approach to implementing organisational changes (mainly through imposition) in which everyone appeared to cooperate unquestioningly with management domination. For example this manager said: I think we have a very tolerant culture. We are very equal opportunities drivenThere is no them and us and no blame culture. All the problems get sorted out at our team meetings which can be a bit heated. There is a lot of praise culture. I think the T&D Manager is partly responsible for the praise and encouraging culture. We have worked in other sections where a lot of praise was not the norm. You were undervalued [in the past] not necessarily with pay. We value staff.

www.ejbrm.com

55

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (49-60)

Although knowledge acquisition refers to a process that is essentially open (one cannot be sure that testing has been sufficient, that no adaptation of knowledge may be necessary), it does appear possible to achieve plateaus periods where the need for further adaptation and survive market forces and external change drivers appears at least temporarily diminished as the above respondent seems to suggest. In such periods employee presence will appear easily negotiable, and its establishment may be considered friendly and desirable by managers and shareholders. It happens when all parties proceed (in an enriching and nondominating or totalising manner) within the constraints of overall adaptation to change requirements. The above-cited employee response totally contrasts with the language being used by another employee of a different status along these lines: There is a mixture of praise and blame culture in my organisation. There is also a them and us culture especially reflected in the friendships between those in management. Ordinary employees do not enjoy some of these privileges. At very few times you are praised for doing a good job especially when the orders are not delayed but when these targets are not met on time then the production lines are blamed as a whole. Conflicts have a tendency to arise when they are least wanted, i.e. when there are other more pressing conflicts that need the attention of both managers and employees alike. This appeared to be exemplified by the periods of stressful change. Managers and employees no longer supported each others presence, but tried to manipulate the latter to what they considered their own personal survival advantage. Managers tended to reduce their own presence (by mechanising processes of communication), while other employees at times tried to increase it for example by bringing conflicts and other forms of resistant manifestations to the fore at what managers will consider the wrong time. A clear dichotomy in the concept of them and us (different from Strauss and Quinns maintained position) signals the growing rift and tensions between management and employees. One such conflict concerned frequent cigarette breaks. It was intensified due to mutually incompatible linguistic systems between managers and employees. On the one hand such breaks were deemed to express the attitude of lazy staff, on the other they were seen as rightful ways to cope with disenchantment and disengagement from the change process. Such differences in interpretation will not disappear when one language is made to dominate, e.g. to lead to measures to push staff to become less lazy and more integrated with cultural change interventions. Differences in the linguistic systems being used by employees may create a sphere of mystery and alienation between managers and non-management staff. This sometimes is supported intentionally, as in the case of political figures using two such systems to communicate with those higher up and their electorate, those lower down. Unfortunately, the resulting form of command communication tends to be inefficient in high-level tasks precisely the type of situation that organisational change is intended to improve. In the interview process, I may have appeared to ask managers and employees to present accounts of what they might or might not have done through different levels of language. The latter option may have been wished for by employees. Such potential preference of interview issues could have skewed the data presented in this study to reflect some platforms of interactions more than others. Viewed alternatively, the aspects that respondents highlighted also showed the dynamic range of interactive experiences when employees undergo stressful cultural changes.

4.2 Stories
My interest in using interviews as a research platform to collect employee stories is linked to my principal research question What are the consequences on employees and others when organisations change? After collecting these stories I interpreted the sets of meanings (the other platforms) that emerged from the issues raised and validated by employees. In this approach, meanings are inter-subjectively shared between the respondents and myself. The world is understood and meaning is constructed through social artefacts, (stories, myths, artefacts and symbols) products of historically situated interchanges among people (Gergen, 1985). Instead of a straightforward emergence of themes from the interview data there were some unexpected subthemes such as implicit and explicit conflict, overt and covert resistance and so on that emerged from the main themes and issues evoked in the interviews, not just denoting the richness in research but part of human existence as well. It was such non-rule-like nature in peoples cultural values that I tried to surface

www.ejbrm.com

56

Academic Conferences Ltd

John Mendy

from employees experiences of organisational changes which are at odds with the Strauss and Quinn position. It was also by way of identifying employees contributions within the framework of maintained, increased or reduced presence that I tried to highlight the extent to which employees experiences have been impoverished or enriched within their interactions. Some employees tend to relate their individual expectations to what should be happening in their organisations and how managers need to behave in order to facilitate a smooth cultural change process. Sometimes, these expectations were at variance with organisational practices, expectations, policies, procedures and managers actions, thus the sub-themes mentioned earlier. Such varieties of interactions and networks of relationships provide the study with its rich employee experiences which are not necessarily maintained in as integrative a manner as presented in Strauss and Quinns theorising. This highlights some of the dilemmas that face researchers interested in complex management and organisation issues such as culture and change, among others. These dilemmas can be further broken down and categorised into 2 sections: ontological and epistemological.

4.2.1 Ontological dilemmas

I now wish to examine some core assumptions and dilemmas about ontology based on the studys findings. These are intended to provide an inevitably sketchy map for thinking about the different perspectives that various organisational and social scientists hold about human beings, their behaviours and the world around them. Each of these social thoughts has evolved in recognition of the existence and proliferation of other points of view, and, to some extent, developed as a reaction to competing methodological and theoretical viewpoints. The social world and what could be considered as reality is a projection of individual consciousness. This paper has also recognised that the consequences of organisational changes on employee behaviours are created through the inter-subjective sets of meanings between respondents and researcher. I compared such individual rationalising using employee language and symbols as they related to the quality of the sameness of the statements and experiences before being classified under collective experiences in the form of platforms. The varieties of experiences denote a combined collection of sets of cognitive recollections of peoples reactions and the sense they make of their organisational changes. It was this body of knowledge that I tried to represent in the emerging platforms that demonstrated the interactions between the personal and organisational domains. Social reality and meaning are found in the nature and use of these cultural modes of interactions that are worthy of interpretations. These provided the basis of employee experiences from which I question Strauss and Quinns theorising. Capturing symbolic modes of meanings in employees daily lived experiences of cultural change may be shared with managers, but, at times, multiple assumptions and cultural differences (that account for the conflicts, tensions and resistances) have also been recognised in this study. This poses yet another dilemma for the social scientist who intends to generalise his/her findings in an integrative way. Thus to discuss organisational culture within an entirely integrative framework appears to me too idealistic and utopian as it fails to fully account for the existence and recognition of other fragmentary and sub-cultural ways of doing things. Social and organisational reality rests not on following rules but in what the system of interactions mean to respondents that presents itself to the researcher as following a pattern. Piecing out a pattern of the interactions between respondents stories and statements would appear rule-like. This seems to render the phenomenological researcher rather structuralist thereby defeating the very notion of depicting employees lived experiences, adding to the dilemmas of researchers.

4.2.2 Epistemological dilemmas

These different ontological assumptions also pose epistemological dilemmas. The various world-views expressed imply different forms of knowledge for the social and organisational worlds we inhabit in and try to construct and analyse. From one assumption to the other one what could be regarded as the subjective/objective continuum, the nature and form of what really makes up knowledge varies. For example, the objectivist perspective of what constitutes the world as a concrete structure would encourage an epistemological position that stresses the importance of studying relationships that are rendered concrete, static and, in Strauss and Quinns (1997) view maintained. Such knowledge implies an understanding of research phenomena with much emphasis on the empirical analysis of concrete and causal relationships. It yearns for what appears to be objective form of knowledge identifying the nature of rules and relationships among phenomena. Such events such as employees behaviours are considered social facts (Pugh and

www.ejbrm.com

57

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (49-60)

Hickson, 1976a, 1976b). The famous Durkheimian (1938) dictum that we should consider social facts as things and recognise them by the way they constrain us is an early origination of this argument. At the extreme end of the epistemological continuum is the subjectivist perspective of reality as a projection of human experience and interaction. It contests the causal and rule-like basis on which knowledge of organisational phenomena such as culture and change can be claimed. It is more inclined towards a study of processes through which organisational and social actors render their interactions concrete. This phenomenological and subjective stance challenges the notion of objective knowledge that can be transmitted in any concrete form. This is because the knowledge created could merely be regarded as an expression of the manner in which the scientist (a subjective being), has subconsciously imposed a personal reference frame on the world, which is quite often, wrongly regarded as lying in an external and separate sphere (Husserl, 1960). The bases for the creation of and claims to knowledge in the assumptions stated herein are quite divergent because their viewpoints on what forms organisational and social reality (such as the dynamic interactions between members during organisational culture change) are equally polarised. This is part of an ongoing set of dilemmas in management and business studies.

5. Results of analysis
The results of the analysis may be summarised as follows. Without an understanding of the notion of presence, managers tend to show behaviour with unreasonable levels of destructive effect, i.e. behaviour to the detriment of the organisation as well as to its employees. By initially recognising and trying to manage presence increasing it, decreasing it in terms of overall performance of some organisation such behaviour may be avoided and may make companies (private-sector organisations in this case) better places for employees to work in, to learn and develop. Typically, such companies strive continuously to redesign their organisational structure to deal with internal and external challenges and increase their competence, i.e. their ability to work systematically and cooperatively (Hammer and Champy, 1995). Changes of this nature depend on suitable forms of communication in the interactions between top and middle management and employees in the organisations studied. It was noted that many attempts tend to be made to improve communication, for example through Staff Forums, newsletters, notice boards and so on, but that these remain ineffective if they are not geared to increase and balance employee presence. Making employees feel valued and dignified human beings need not imply that companies cannot change or downsize. It means that they can do so without first increasing or maintaining the damage that they try to reduce.

6. Concluding remarks
The theoretical and methodological position I have maintained in this paper is one in which the phenomenological approach to research in management and business studies makes its case as one moves from the objectivist end to the interpretive and subjectivist end of the research continuum. I have tried to sustain the argument that Strauss and Quinns (1997) culturally-maintained position is questionable by using symbolic interactionism and stories as alternative approaches. These theories and methods facilitated through the interview platforms recognise the importance of language and organisational discourses validated by speech actors through which consequences of organisational cultural changes on employees experiences and reactions can be richly understood. I have tried to distance myself from reducing the complexities and dynamic interactions entailed in this study to an objective form of measuring relationships in what would appear to be rule-like, concrete, static and regulated. However, the accompanying epistemological dilemma which renders this knowledge generated from employees experiences of cultural changes as relativistic and context-specific still remains. I also accept the thrilling complexity, ambiguity, fuzziness and unpredictability of culture and change studies and striven for a more interactive approach to understanding the consequences on employees and others when their organisations change.

References
Berger, P. and Luckmann, T. (1966) The Social Construction of Reality, Harmondsworth: Penguin. Deal, T. and Kennedy, A. (1982) Corporate Cultures; The Rites and Rituals of Corporate Life, Hammondsworth: Penguin. Dery, M. (1993) Culture Jamming, Hacking, Slashing and Sniping in the Empire of Signs, New York: Addison-Wesley. Durkheim, E. (1938) The Rules of Sociological Method, New York: Free Press. Garfinkel, H. (1967) Studies in ethnomethodology, London: Englewood Cliffs. Gergen, K. (1991) The saturated self: Dilemmas of identity in contemporary life, New York: Basic Books. Handy, C. (1993) Understanding Organisations, London: Penguin. Hofstede, G. (1980) Cultures consequences, London: Sage. Husserl, E. (1960) The idea of phenomenology, The Hague: Martinus Nijhoff.

www.ejbrm.com

58

Academic Conferences Ltd

John Mendy

Kuhn, T. (1970) The Structure of Scientific Revolutions, Chicago: University of Chicago Press. Magala, S. (2005) Cross-Cultural Competence, London: Routledge. McLuhan, M. (2001) The Medium is the message, California: Gingko Press. Morgan, D. L. (1997) Focus Groups as Qualitative Research, London: Sage Publications. Pugh, D. S. and Hickson, D. J. (1976a) Organisational structure in its context 1, Farnborough: Saxon House. Pugh, D. S. and Hickson, D. J. (1976b) Organisational structure: Extensions and replications 2, Farnborough: Saxon House. Schein, E. H. (1985) Organisational Culture and Leadership, San Francisco: Jossey-Bass. Schutz, A. (1967) Phenomenology of the Social World, Illinois: Evanston: Northwestern University Press. Strauss, C. and Quinn, N. (1997) A cognitive theory of cultural meaning, Cambridge: Cambridge University Press.

www.ejbrm.com

59

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (49-60)

www.ejbrm.com

60

Academic Conferences Ltd

Development of Variant Definitions for Stakeholder Groups with Regard to the Performance of Public Transit in the United States
Jason Keith Phillips1 and Diane M. Phillips2 1 Department of Marketing, West Chester University, Pennsylvania, USA 2 Department of Marketing, Saint Josephs University, Philadelphia, Pennsylvania, USA jphillips@wcupa.edu dphillip@sju.edu
Abstract: In the United States, the performance of public transit systems is often characterized, in both the popular press and academic literature, as being mediocre at best and growing steadily worse. Drawing on insights from the extant literature and a census of all U.S. transit systems, this research argues that multiple definitions of transit performance exist and are composed of three macro-constructs which are weighted differently by different stakeholder groups and thereby provide unique definitions for each stakeholder group. Statistical analysis of the data suggests that there are differences in both the absolute importance and relative importance placed on the three macro-constructs by U.S. transit stakeholder groups. Therefore, the examination of public transit performance from a multiple stakeholder points-of-view perspective appears warranted in order to better evaluate the performance of public transit systems. Keywords: transit performance, public transit, performance assessment, performance constructs, stakeholder

1. Introduction
The performance of public transit systems in the United States is often characterized, in both the popular press and academic literature, as being mediocre at best and growing ever worse. Statistics are often cited which show rising costs, increasing deficits, declining ridership, and other associated ills (see, for example, Lave (1991)). One thing all these publications have in common is the implicit assumption of common agreement in defining transit system performance. That is, the literature on transit system performance, both popular and academic, has generally proceeded from the assumption that there is a common, universally accepted definition of performance held by all stakeholders. The present research argues that no such definition exists (and is unlikely to ever exist) and that the definition of transit system performance is, in fact, dependent upon the point of view of the stakeholder. Specifically, rather than taking the definition of performance as a given, this research assumes that no universal definition of performance exists and argues that each stakeholder group develops its own unique definition based upon its own particular needs and desires. These multiple definitions are composed of a variety of macro and micro constructs, such as vehicle efficiency and pollution reduction, which are weighted differently by different stakeholder groups and thereby provide unique definitions for each stakeholder group. For this research, the term stakeholder will be used to refer to groups that are directly or indirectly affected by the implementation and results of social programs (Rossi and Freemen, 1993: 2).] A closely related example that supports the multiple constructs approach this research advocates is provided by Gault and Doherty (1979) who state that there are many aspects of bus service that make up a passengers total view of its performance. They state that such aspects as comfort, cleanliness, speed, and reliability are just a few of the components of transit system performance. As passengers are but one stakeholder of public transit, it is logical to infer that the other stakeholders of public transit, for example state governments, metropolitan planning organizations, and transit operators, also define performance in terms of a variety of different aspects (or constructs). This research builds upon the work of Connolly, Conlon, and Deutsch (1980) who argue for a multiconstituency approach to assess the effectiveness of an organization. Connolly, Conlon, and Deutsch (1980) state that the various constituencies of an organization will assess the effectiveness of an organization differently. Specifically, they propose: ...A view of organizational effectiveness in which several (or, potentially, many) different effectiveness statements can be made about the focal organization, reflecting the criterion sets of different individuals and groups we shall refer to as constituencies. (1980: 212)

ISSN 1477-7029 61 Academic Conferences Ltd Reference this paper as: Phillips, J. K. and. Phillips, D. M (2007) Development of Variant Definitions for Stakeholder Groups with Regard to the Performance of Public Transit in the United States. The Electronic Journal of Business Research Methods Volume 5 Issue 2, pp 61 - 70, available online at www.ejbrm.com

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (61-70)

The constituencies advocated by Connolly, Conlon, and Deutsch (1980) are rather diverse and include groups both within the organization (for example, senior management) as well as groups outside of the organization (for example, customers). The present research utilizes the term stakeholder rather than constituency because it is interested primarily in the constituencies that are directly involved with the transit system on a macro basis. That is, this research is concerned with constituencies on a firm or organizational level (i.e., the transit operator, the MPO, and the state government).

2. Contribution to business research methods

This research examines if what are purported to be commonly held beliefs with regard to the definition of public transit system performance are, in fact, actually commonly held. This paper utilizes the research on transit system goals and performance measurement to illustrate the importance of verifying the underlying assumptions which compose the foundation of a research area before conducting any new research in that area. It utilizes a survey of all transit systems in the United States to examine whether certain governmental stakeholders of public transit weigh the three constructs of transit system performance efficiency, effectiveness, and impact in the same way. The importance of this paper with regard to business research methods lies in its illustration that failing to examine the underlying assumptions which compose the foundation of a research area may undermine the validity and usefulness of a new research effort. Niccolo Machiavelli (1469 1527), the Italian dramatist, historian, and philosopher, acknowledged the importance of strong foundations in his book, The Prince: He who has not first laid his foundations may be able with great ability to lay them afterwards, but they will be laid with trouble to the architect and danger to the building. The same can be said for the need for a strong foundation when conducting research, because defining and understanding the foundation upon which a given research is based is critical before commencing any new research project.

3. Challenges faced by public transit operators

It should be noted that public transit systems face three primary challenges due to the goals and objectives of transit stakeholders. These challenges result due to conflicting goals both within and among stakeholder groups, changing stakeholder goals and objectives over time, and differences in the specific goals and objectives of stakeholders and the importance placed on those goals and objectives by different stakeholder groups. First, one difficulty faced by transit operators is the often conflicting goals which transit must serve. That is, the achievement of one goal by a given transit system may hinder, or prevent, the achievement of another goal(s). The fact that transit services must respond to a variety of needs, including those of the elderly, the handicapped, and residents of minority areas, has been noted in the literature. The performance goals of public transit can result from the societal goals advocated by governments at the federal, state, and local levels. Federal legislation (such as the Americans with Disabilities Act) and local government plans for urban revitalization and development can impact the performance goals of public transit systems and can result in performance goals conflicting with one another. For example, if a transit system discontinues service on a particular transit corridor due to low ridership, the overall cost efficiency of that system might improve. However, if the elimination of the service hinders the urban revitalization plans of the relevant MPO, the elimination would reduce the performance of the system from the point of view of the MPO (as well as to the passengers who currently use the eliminated service). This problem was identified by Fielding, Glauthier, and Lave (1978) when they noted that many goals may be suggested for transit, but transit systems can not pursue all of them simultaneously. The fact that transit systems that satisfy some of these needs may not appear to be performing well when appraised via traditional performance evaluation techniques has also been identified in the literature (see, for example, Stokes, 1979). Second, public transit systems are challenged by the dynamic political/legal environment in which they operate. Fielding (1987) identified that the evaluation of public transit is made difficult by changing stakeholder goals. Preserving the commercial advantages of central cities and the jobs of transit employees was initially the primary concern of government with regard to public transit; then social, environmental, and conservation goals were

www.ejbrm.com

62

Academic Conferences Ltd

Jason Keith Phillips and Diane M. Phillips

added in the 1970s. As a result, performance analysis for transit must embrace efficiency, effectiveness, as well as equity dimensions (Fielding, 1992). Finally, public transit systems are challenged because government objectives may differ from one level of government to another or one level of government may place greater weight on a particular objective than another level of government (Dajani and Gilbert, 1978) (Talley, 1983). As a result, the achievement of a particular goal by a public transit system may result in substantially different interpretations as to the importance of that achievement among transit stakeholders.

4. Literature review
What is transit performance? If performance is defined as the manner in which or the efficiency with which something reacts or fulfills its intended purpose (Stein, 1983: 1070), transit performance could be defined as the manner in which or the efficiency with which public transit fulfills its intended purpose. But what is the intended purpose of public transit? An examination of the literature reveals that disagreement exists with regard to what the intended purpose of public transit is and thereby what components, or constructs, of transit service should be considered when attempting to assess the performance of transit systems. Phillips (2004), in his article concerning an application of the balanced scorecard approach, provides a general discussion of the metrics of public transit as well as a rationale and methodology for categorizing transit performance measures. This current research is based on the shopping list of macro and micro constructs developed by Phillips (2004) for use by managers when developing balanced scorecards for public transit systems. Phillips (2004) also gathered together, for the first time, a comprehensive list of public transit constructs / goals and their corresponding performance measures in order to allow public transit managers to develop balanced scorecards or any other performance assessment framework in a more efficient manner.

5. General categories of performance measures

The literature, in general, has settled upon two categories of performance measurement: efficiency and effectiveness. The present research will also utilize a third category, impact, which is advocated by Dajani and Gilbert (1978). That is, for the purpose of the present research, the concept of transit productivity in the public sector encompasses efficiency, effectiveness, and impact. Efficiency indicates the extent to which the government produces a given output with the least possible use of resources. Efficiency indicators rate the processes by which transit services are produced, particularly through the relationship of inputs to outputs. That is, efficiency is concerned with "doing things right." Effectiveness has been defined as the comparison of produced output (provided service) to intended output or objectives. That is, measures of effectiveness are concerned with the extent to which the service is provided - in terms of quantity, location, and character - corresponds to the goals and objectives established for it by government and the needs of citizens. Thus, effectiveness is concerned with "doing the right things." The third category, impact describes the macro effects of public transit and reflects the efficiency and effectiveness of transit, as well as external and indirect effects on social well-being, economic development, and environmental quality. That is, impact includes externalities and indirect effects both beneficial and adverse, intended and unintended (Dajani and Gilbert, 1979).

6. Methodology
6.1 Respondents
Respondents were U.S. transit systems identified by the Federal Transit Administration as well as their respective MPOs and state governments. Further, it was believed that it was critical to identify a highly placed individual at each of these stakeholder entities in order to ensure that the person filling out the survey was in a position that was senior enough to be able to provide an informed opinion about the organizations goals and priorities. [However, the use of the key informant method to collect information (Campbell, 1955)(Phillips, 1981) was also one of the primary limitations of this research. As a result, the opinion of the chief executive officer of each stakeholder entity was used to approximate the overall importance placed on the constructs of transit performance by a given stakeholder group. The key informant method uses informants that are selected based on their appropriateness to answer the questions being asked as

www.ejbrm.com

63

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (61-70)

opposed to random selection. For the purpose of this research, the importance placed on various transit performance constructs is a question best answered by individuals involved in stakeholder strategy formulation: the chief executive officer.] This procedure resulted in a total of 467 transit system contacts, 341 MPO contacts, and 53 state contacts (including Washington DC and Puerto Rico).

6.2 Development of survey materials

Previous research on governmental bodies has been conducted using survey methodology. The current research also develops a survey which was designed to assess the importance placed, on the basis of stakeholder group, on each of the three constructs of transit performance which were identified via the literature review. Each contact person received a small packet of materials by mail which contained a cover letter explaining the basic purpose of the survey and requesting their participation in this research effort. In addition, each packet contained a copy of the survey and a return envelope which was stamped and addressed to the author. The survey was 5 pages long (front and back) and took respondents approximately 15 minutes to fill out. Due to limited financial resources, only one mailing was possible. An informal pretest of this study was conducted by mailing the survey to one transit system that expressed a willingness to work with researchers in the past. After this manager had completed the questionnaire, the questionnaire was discussed with him by telephone. This individual was asked to criticize the questionnaire and was asked if any of the questions it contained were unclear or confusing. Based on this respondent's recommendations, the wording of some questions was changed.

7. Survey overview
The first section of the survey focused on the macro constructs of transit performance. It began by defining the three broad components of transit performance (e.g., efficiency, effectiveness, and impact), asking participants to rate each item on importance, and then asking participants to rank order the three items by importance. The next section provided brief definitions of the 25 individual elements of transit performance compiled by Phillips (2004) and asked participants to rate each item based on the importance placed on the item by their respective organization. The final section of the survey asked participants to rank order the 8 individual elements of efficiency, the 11 individual elements of effectiveness, and the 6 individual elements of impact.

8. Development of the dependent measures

Performance Component Scale. The first part of the survey asked participants to rate on a 17 scale the importance of each of the macro constructs of transit performance. Endpoints were labeled not at all important very important. Performance Component Rating. The first part of the survey asked participants to rank order the three macro constructs of transit performance. Thus, participants were required to assign rankings of 1st, 2nd, and 3rd to efficiency, effectiveness, and impact. Efficiency Composite. In the second part of the survey, participants were required to rate each of the 25 individual elements of transit performance on a 1-7 scale with endpoints labeled not at all important very important. The 8 items making up the efficiency component were then combined to create an efficiency composite. A reliability analysis indicated that this measure is internally consistent (Cronbach alpha = 0.8518). Effectiveness Composite. The 11 items making up the effectiveness component in the second part of the survey were combined to create an effectiveness composite. This measure also proved to be internally consistent (Cronbach alpha = 0.8643). Impact Composite. Finally, the 6 items making up the impact component of performance in the second part of the survey were combined to create an impact composite. Again, this measure was internally consistent (Cronbach alpha = 0.8380). When combined, the efficiency composite, effectiveness composite, and the impact composite also demonstrate fairly high internal consistency (Cronbach alpha = 0.8103). This indicates that these factors may indeed be a part of some larger construct such as transit performance, as asserted herein.

www.ejbrm.com

64

Academic Conferences Ltd

Jason Keith Phillips and Diane M. Phillips

Efficiency Ranking. The 8 individual elements of efficiency were rank ordered by the participants such that each individual element received a ranking of between 1st and 8th. No tie scores were allowed and each element had to receive one rank Effectiveness Ranking. The 11 individual elements of effectiveness were also rank ordered by participants with each element receiving a rank of between 1st and 11th. No tie scores were allowed and each element had to receive one rank. Impact Ranking. The 6 individual elements of impact were also rank ordered by participants with each element receiving a rank of between 1st and 6th. No tie scores were allowed and each element had to receive one rank.

9. Results
Of the 875 surveys sent out, 289 were returned within three weeks. This resulted in an overall return rate of 33.0% for the surveys. Of those returned, there were 137 usable surveys from transit systems (29.3% return rate), 109 usable surveys from MPOs (32.0% return rate), and 16 usable surveys from state governments (30.2% return rate). Table 1: Key to variable abbreviations
Variable Abbreviation Effic_s Effec_s Impac_s Effic_r Effec_r Impac_r Effic1s Effic8s Effec1s Effec11s Impac1s Impac6s Effic1r Effic8r Effec1r Effec11r Explanation A single item measure of efficiency, measured on a 1-7 importance scale. A single item measure of effectiveness, measured on a 1-7 importance scale. A single item measure of impact, measured on a 1-7 importance scale. A ranking of efficiency by importance, measured by assigning a 1st, 2nd, or 3rd ranking to efficiency, effectiveness, and impact. A ranking of effectiveness by importance, measured by assigning a 1st, 2nd, or rd 3 ranking to efficiency, effectiveness, and impact. A ranking of impact by importance, measured by assigning a 1st, 2nd, or 3rd ranking to efficiency, effectiveness, and impact. Eight single item measures of the micro constructs making up the efficiency construct, measured on 1-7 importance scales. Eleven single item measures of the micro constructs making up the effectiveness construct, measured on 1-7 importance scales. Six single item measures of the micro constructs making up the impact construct, measured on 1-7 importance scales. The eight items making up the rankings of the individual items of efficiency, st th measured by assigning rankings of 1 8 to each of the efficiency items. The eleven items making up the rankings of the individual items of st th effectiveness, measured by assigning rankings of 1 11 to each of the effectiveness items. The six items making up the rankings of the individual items of impact, st th measured by assigning rankings of 1 6 to each of the impact items. The 8-item composite of individual scale items of efficiency. The 11-item composite of the individual scale items of effectiveness. The 6-item composite of the individual scale items of impact.

Impac1r Impac6r Effictot Effectot Impactot

10. Macro-level constructs of transit performance

The literature review identified the three macro constructs of transit performance to be efficiency, effectiveness, and impact. In addition, the literature review identified several micro constructs associated with each of these macro constructs. Efficiency is made up of 8 individual elements, effectiveness is made up of 11 individual elements, and impact is made up of 6 individual elements. In all, 25 individual items are components of the three constructs of transit performance. These 25 individual items can be conceptualized as being micro constructs of transit performance. A factor analysis of these individual items was performed in order to confirm that these items did indeed group together as efficiency, effectiveness, and impact. The 25 items were subjected to a promax rotation and the results converged after six iterations. The three factors (i.e., constructs) accounted for 53.01% of the total variance and an inspection of the scree plot indicated that a three factor solution was optimal (see Table 2). Only two of the 25 items failed to load on the proper factor (e.g., effec9s and impac3s). These two items were, however, assigned to the factor to which there is already a strong theoretical basis for these items to belong. Because this is the first study of its kind to examine the

www.ejbrm.com

65

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (61-70)

three constructs of transit performance and because the three constructs are strongly correlated with one another, it was deemed acceptable to do this. Table 2: Factor loadings for the 25 performance items
Item Effic1s Effic2s Effic3s Effic4s Effic5s Effic6s Effic7s Effic8s Effec1s Effec2s Effec3s Effec4s Effec5s Effec6s Effec7s Effec8s Effec9s Effec10s Effec11s Impac1s Impac2s Impac3s Impac4s Impac5s Impac6s 1st Component 0.141 0.048 0.116 0.134 0.179 0.231 -0.216 -0.148 0.540 * 0.850 * 0.739 * 0.342 * 0.419 * 0.689 * 0.658 * 0.736 * -0.099 * 0.647 * 0.627 * 0.022 0.148 0.816 -0.001 0.204 -0.193 2nd Component 0.730 * 0.593 * 0.399 * 0.639 * 0.672 * 0.372 * 0.782 * 0.897 * 0.003 -0.232 0.012 0.071 0.031 0.117 -0.032 0.174 0.575 0.247 0.070 -0.013 -0.254 -0.185 0.158 0.018 0.029 3rd Component -0.082 0.099 0.358 -0.052 -0.051 0.011 0.121 -0.044 0.195 0.020 -0.000 0.339 0.275 -0.170 0.106 -0.073 -0.050 -0.101 0.053 0.835 * 0.786 * -0.021 * 0.845 * 0.577 * 0.905 *

The three factors are fairly highly correlated with one another: r (1,2) = 0.601, r (1,3) = 0.542, r (2,3) = 0.471.

11. Primary research question: Q1: Do stakeholder groups view the three macroconstructs of performance differently?
In order to answer this question, three separate analyses were conducted. The first analysis assessed the extent to which the single item ratings of performance on importance would be different depending on stakeholder group. An ANOVA with effic_s as the dependent variable and level of government as the independent variable indicated that ratings of efficiency did indeed depend on the stakeholder group. The overall model was significant (F (2, 257) = 5.759, p < .004) as was the level of government (F (1) = 11.518, p < .001). A closer look at the means indicated that transit systems rate efficiency as being more important than do MPOs (6.1407 vs. 5.7890, p < .008) or state governments (6.1407 vs. 5.4375, p < .010) (see Table 3). ANOVAs were also conducted for effec_s and for impac_s, but no significant differences were found. The second analysis attempted to answer this same question of whether stakeholder groups would view the three constructs of performance differently. This analysis used the forced rating scales as the dependent variables. The ANOVAs and the analysis of means comparisons indicated no significant differences according to this measure (see Table 3). The third analysis used three composites formed from the micro constructs as dependent variables. For the first variable, effictot, the ANOVA revealed that ratings of efficiency do depend on level of government. The overall model was significant (F (2, 152) = 7.070, p < .001) as was the level of government variable (F (1) = 12.599, p < .0005). An analysis of the means indicated that transit organizations rate efficiency as significantly more important than do MPOs (5.4361 vs. 5.0330, p < .0005) or do state governments (5.4361 vs. 4.9844, p < .051). For the second macro construct, effectot, the ANOVA revealed that ratings on effectiveness do indeed depend on the level of government. The overall model was significant (F (2, 257) = 16.167, p < .0005) as was the level of government variable (F (1) = 31.084, p < .0005). A closer look at the means indicated that transit systems rated effectiveness as significantly more important than did MPOs (6.0602 vs. 5.646, p < .0005) or state governments (6.0602 vs. 5.4318, p < .0005). The analysis of the third macro construct, impactot, also found that ratings of impact depend on the level of government. The overall

www.ejbrm.com

66

Academic Conferences Ltd

Jason Keith Phillips and Diane M. Phillips

model was significant (F (2, 256) = 2.339, p < .098) as was the level of government variable (F (1) = 4.042, p < .045). Here, the analysis of means indicated that the transit systems rated impact as significantly more important than did MPOs (5.0662 vs. 4.7809, p < .038) (see Table 3). No significant differences were noted between transit systems or MPOs and state governments. Table 3: Comparison of means for the 3 macro-constructs
Efficiency Measure And Mean Effic_s 6.1407 c d 5.7890 c 5.4375 d 5.9500 260 Effic_r 1.9070 1.9608 2.1875 1.9474 247 Effictot 5.4361 a f 5.0330 a 4.9844 f 5.2402 255 Effectiveness Measure And Mean Effec_s 6.2296 6.1376 6.1875 6.1885 260 Effec_r 1.5659 1.5784 1.5625 1.5709 247 Effectot 6.0602 a b 5.6246 a 5.4318 b 5.8406 260 Impact Measure And Mean Impac_s 5.4060 5.2963 5.5625 5.3696 257 Impac_r 2.5271 2.4608 2.2500 2.4818 247 Impactot 5.0662 e 4.7809 e 4.7778 4.9305 259

Transit MPO State Total N

Transit MPO State Total N

Transit MPO State Total N

Key: for both the scale ratings and the composites of the scale ratings, higher numbers indicate greater levels of reported importance. For the rankings, however, lower numbers indicate greater levels of importance. Levels of significance are as follows: a: p < .0005, b: p < .0005, c: p < .010, d: p < .010, e: p < .050, f: p < .060. In all, these results indicate that ratings of efficiency, effectiveness, and impact do critically depend on the stakeholder group that is providing the rating. That is, the three stakeholder groups examined - transit systems, MPOs, and state governments - do indeed view the constructs of performance differently. The composite measures seem to provide the best evidence of this dependence.

12. Analysis
The primary research question was examined through the development of testable hypotheses that decompose transit system performance into its three macro constructs -- efficiency, effectiveness, and impact -- and then relate these constructs to stakeholder group. The three null hypotheses H01: The importance placed on public transit efficiency by a government entity is not related to the level of that entity. H02: The importance placed on public transit effectiveness by a government entity is not related to the level of that entity. H03: The importance placed on public transit impact by a government entity is not related to the level of that entity. and the three alternative hypotheses: Ha1: The importance placed on public transit efficiency by a government entity is negatively related to the level of that entity. Ha2: The importance placed on public transit effectiveness by a government entity is negatively related to the level of that entity. Ha3: The importance placed on public transit impact by a government entity is positively related to the level of that entity.

www.ejbrm.com

67

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (61-70)

were examined using the three stakeholder groups: transit systems, MPOs, and state governments. Given the statistical results, all of the null hypotheses were rejected. Further, of the three alternative hypotheses, one (Ha3) was not supported. Specifically, significant results were found between the level of government entity (i.e. transit systems, MPOs, and state governments) and the impact construct of transit performance but in the opposite direction of that predicted by Ha3. That is, impact is more important to transit systems than to MPOs. With regard to Ha1, significant relationships were found between the level of government entity (i.e. transit systems, MPOs, and state governments) and the efficiency construct. Specifically, as predicted, the efficiency component of transit performance is most important to transit systems, followed by MPOs and state governments. The fact that Ha1 is supported suggests that the higher the level of government, the less important efficiency aspects of transit performance (i.e., doing things right) become. This finding suggests that the assertion that efficient use of transit funds becomes more important as the amount of money involved (as a percentage of a stakeholders total budget) increases may be true. In addition, it suggests that transit systems may value efficiency more than either MPOs or state governments because inefficiency can result in additional governmental regulation. Thus, the assertion that transit systems tend to value efficiency in order to avoid the imposition of additional regulation may indeed be one possible explanation for the importance placed on efficiency by transit systems. With regard to Ha2, significant relationships were found between the level of government entity (i.e. transit systems, MPOs, and state governments) and the effectiveness construct. That is, the importance placed on effectiveness appears to decrease as the level of government increases. Further, the analysis suggests that the effectiveness of public transit (i.e. doing the right things) is more important to transit systems than to either MPOs or state governments. Although further research is needed, the assertion that as the percentage of constituents that are transit users increases, the effective provision of transit becomes more important appears to have some merit. Finally, the lack of support for Ha3 shows that additional research is needed with regard to the relationship between level of government and importance of transit system impact. It was expected that the higher the level of government, the more important the impact aspects of transit performance would become. However, the exact opposite is implied by the findings of this study. Specifically, the impact of public transit is more important at the transit system level than at the MPO level. This result implies that transit systems are also concerned with the macro effects of transit. That is, transit systems value efficiency and effectiveness of transit as well as its external and indirect effects on social well-being, economic development, and environmental quality. It is hoped that future research, which includes data from the federal level, will clarify this result as the assertion that the federal level is most concerned with social engineering and advancement of social well-being, economic development, and environmental quality is a logical one.

13. Discussion
The purpose of this research was to determine whether the definition of transit system performance is, in fact, dependent upon the point of view of the stakeholder. Further, it argued that these multiple definitions are composed of a variety of constructs, such as vehicle efficiency and pollution reduction, which are weighted differently by different stakeholder groups and thereby provide unique definitions for each stakeholder group. The literature review identified the three macro constructs of transit performance: efficiency, effectiveness, and impact. It also identified the 25 micro constructs of transit performance of which the three macro constructs are composed. This literature review was then used as the basis to explore the primary research question of this study: Q1: Do stakeholder groups view the three macro-constructs of performance differently? This question was examined through the development of testable hypotheses that decompose transit system performance into its three macro constructs -- efficiency, effectiveness, and impact -- and then relate these constructs to stakeholder group. The primary contribution of this research is the preliminary confirmation that stakeholder groups do indeed view the three macro-constructs of performance differently and as a result have stakeholder-specific definitions of transit performance. However, given that this study was exploratory in nature, future research is clearly needed to confirm its findings.

www.ejbrm.com

68

Academic Conferences Ltd

Jason Keith Phillips and Diane M. Phillips

One limitation of this research is that stakeholders with vastly differing operations and objectives were grouped together. Although a broad range of stakeholder types exist within each stakeholder group, this research does not categorize transit stakeholders except with regard to the demographics of the areas in which they operate. The assignment of transit systems, MPOs, or states into peer groups is an area of ongoing research and to date there is no generally accepted methodology for peer group formation. Thus, although the methods used in this research are based on prior research, these methodologies are valid but not necessarily optimal.

14. Summary and conclusion

This research suggests that future efforts to assess transit system performance must take into account the various stakeholders of public transit. However, as this is the first research effort to examine the relationship between transit performance and transit stakeholder group, additional research and verification is needed. It is hoped that this research will serve as a starting point for the further examination of this relationship. For example, future research on the development of transit performance measures and measurement sets may be improved if each measure is clearly related to both its applicable construct and the importance placed on that construct by the various transit stakeholders. Further, if transit system operators feel that their point of view regarding their performance is being considered by their respective governmental stakeholders, they may be more willing to support programs for transit system performance evaluation and improvement. This paper also illustrates the importance of examining the underlying assumptions of a research area when beginning a new research project. It suggests that verifying the assumptions which form the foundation upon which a given research is based is critical in order to maximize the value of new research.

References
Campbell, D. (1955) The Informant in Quantitative Research, American Journal of Sociology, Vol. 60, pp 339-342. Connolly, T., Conlon, E., and Deutsch, S. (1980) Organizational Effectiveness: A Multiple Constituency Approach, Academy of Management Review, Vol. 5, pp 211-217. Dajani, J. and Gilbert, G. (1978) Measuring the Performance of Transit Systems, Transportation Planning and Technology, Vol. 4, pp 97-103. Fielding, G. (1987) Managing Public Transit Strategically, Jossey-Bass, San Francisco. Fielding, G. (1992) Transit Performance Evaluation in the U.S.A., Transportation Research, Vol. 26A, No. 6, pp 483491. Fielding, G., Glauthier, R. and Lave, C. (1978) Performance Indicators for Transit Management, Transportation, Vol. 7, No. 4, pp 365-379. Gault, H. and Doherty, P. (1979) Measurement of the Performance of Bus Services, University of Newcastle-upon-Tyne: Transport Operations Research Group, Research Report 28. Lave, C. (1991) Measuring the Decline in Transit Productivity in the U.S., Transportation Planning and Technology, Vol. 15, pp 115-124. Machiavelli, N. (1950) The Prince, Random House, New York. Phillips, J. (2004) An Application of the Balanced Scorecard to Public Transit System Performance Assessment, Transportation Journal, Vol. 43, No. 1, pp 26-55. Phillips, L. (1981) Assessing Measurement Error in Key Informant Reports: A Methodological Note on Organizational Analysis in Marketing, Journal of Marketing Research, Vol. 18, pp 395-415. Rossi, P. and Freeman, H. (1993) Evaluation: A Systematic Approach, Sage Publications, Inc., Newbury Park (CA). Stein, J. (ed.) (1983) The Random House Dictionary of the English Language, Random House, New York. Stokes, B. (1979) The Need For and Use of Performance Indicators in Urban Transit, Transit Journal, Winter 1979, pp 3-10. Talley, W. (1983) Objectives for Public Transit Firms, Oxford University: Transport Studies Unit, Working Paper No. 225.

www.ejbrm.com

69

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (61-70)

www.ejbrm.com

70

Academic Conferences Ltd

Applying Grounded Theory to Study the Implementation of an Inter-Organizational Information System

Joan Rodon1 and Joan A. Pastor2 Universitat Ramon Llull, Barcelona, Spain 2 Universitat Oberta de Catalunya, Barcelona, Spain joan.rodon@esade.edu jpastorc@uoc.edu
1

Abstract: This paper shows the application of Grounded Theory (GT) method, particularly the Straussian approach to GT, in a research project that studies the role of managers during and after the implementation of an inter-organizational information system (IOIS). We present the steps followed sampling, data collection, analysis, and literature comparison paying special attention to the intricacies that arose during the research process, and we reflect on the lessons learned from using GT in an interpretive case study. The paper shows: first, the application of the coding paradigm proposed by Strauss and Corbin to analyse process data; second, how action diagrams can help structure and report on process data; and, third, the importance of flexibility, creativity, and keeping an open mind when using GT analytical tools, given that various avenues may be apparent before a plausible theory starts to emerge. We consider the paper illustrates some experiences that may inform others in their GT research process. Keywords: grounded theory, Straussian approach, coding paradigm, action diagrams, inter-organizational IS implementation

1. Introduction
Grounded Theory (GT) can be traced to the seminal work of Glaser and Strauss (1967), The Discovery of Grounded Theory. In that book, both authors were critical of what they perceived to be an approach to research that drew upon an existing grand theory (Mills 1959), and that was satisfied with testing hypotheses built on this underlying theory. In contrast to this hypothetic-deductive approach, GT starts with observations, which are made not to test existing theories, but to discover and generate theories that are as close as possible to the reality observed. While there are no major differences between the respective views of Glaser and Strauss towards key elements like theoretical sampling and constant comparison, the two founders of GT took somewhat different paths. Strauss continued to refine the coding technique by incorporating more analytical techniques, and gave a more active role to the researcher (Strauss et al. 1990). Glaser, on the other hand, argued that rather than putting more emphasis on methods and forcing structure onto data, the researcher should take a passive stance free from preconceptions, trusting that theory will emerge (Glaser 1992). This essay adds to the existing literature, which discusses the use of the two GT approaches (Allan 2002; Fernandez 2004; Hunter et al. 2005), by showing how the Straussian approach to GT was applied in a case study (Rodon et al. 2007a) that inquires into the role of managers during the implementation of an interorganizational information system (IOIS). The purpose of the present paper is not to focus on the results of the GT case study but rather to explain the process followed in conducting such empirical work, paying special attention to the choices and decisions made throughout the research process. The structure of the paper is as follows. First, we present the initial research problem and justify the choice of GT. The next section presents the research process we followed: sampling, data collection, data analysis, data presentation, and theoretical sampling. Finally, the paper highlights several points of the research process and presents the conclusions.

2. The research problem and the choice of GT

In early 2005, the first author had the opportunity to conduct a research study on the implementation of an IOIS in the seaport of Barcelona. That IOIS implementation had generated many discrepancies and taken more time and resources than the promoters had expected. Our purpose in the study was to inquire into the difficulties that companies had faced and were still facing in the integration of their pre-existing information systems with the IOIS and in the further use of the IOIS. The early research question that guided our research was: how does the integration of pre-existing systems with an IOIS unfold? Finding the answer to this research question involved adopting a process perspective to study the implementation of an IOIS.
ISSN 1477-7029 71 Academic Conferences Ltd Reference this paper as: Rodon, J. and Pastor, A. (2007) Applying Grounded Theory to Study the Implementation of an Inter-Organizational Information System. The Electronic Journal of Business Research Methods Volume 5 Issue 2, pp 71 - 82, available online at www.ejbrm.com

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (71-82)

Then we found from existing IOIS literature critical reviews (Elgarah et al. 2005; Lyytinen et al. 2001; Ramanathan et al. 2003) that there were very few empirical papers proposing process models. Most of prior IOIS studies proposed factor-based models or developed descriptive case studies. Given the lack of process-based IOIS research and our interest in generating new insights on the existing literature, we rejected a purely deductive, hypothesis-testing approach for our study. Moreover, in this study we conceptualized the IOIS as a social-technical system, where the dichotomy between social context and technical artefacts dissolves in the complex intertwining of socio-technical actors (Latour 1987). Given the exploratory nature of the research question, our conceptualization of the IOIS, and our concern with describing and analyzing the emerging process of IOIS implementation, rather than explaining it through cause-effect relations between a set of constructs, we opted for an inductive and interpretive approach, and we chose GT as the research methodology. We proposed conducting an in-depth case study following the principles of GT. According to Eisenhardt (1989), the combination of case study with GT has three major strengths: (1) it is likely to produce novel theory (p. 546), (2) the emergent theory is likely to be testable (p. 547), and (3) the resultant theory is likely to be empirically valid (p. 547). On the one hand, a case study would allow us to investigate IOIS implementation within its real-life context, and with multiple sources of evidence. On the other hand, the use of GT was justified for the following reasons. First, GT allowed us to focus on contextual, process-based descriptions of the implementation (Myers 1997). We were concerned with discovering process in data, more precisely, in patterns of action and interaction between the people in response to the problems and situations in which they find themselves (Strauss et al. 1994). Second, GTs consistency with interpretive case studies: ours is interpretive work and interpretations must include the perspectives and voices of the people whom we study. Interpretations are sought for understanding the actions of individual or collective actors being studied. (Strauss et al. 1994, p. 274). Third, GT provided a set of established guidelines both for conducting data collection and analysis (Goulding 2002), which gave us a sense of security when exploring the unknown territory of IOIS management. Fourth, GT has its roots in symbolic interactionism (Blumer 1969). Symbolic interactionism set out three basic premises: (1) "Human beings act toward things on the basis of the meanings that the things have for them."; (2) The meaning of such things is derived from, or arises out of, the social interaction that one has with one's fellows."; and (3) "These meanings are handled in, and modified through, an interpretive process used by the person in dealing with the things he encounters" (Blumer 1969). Thus GT was well suited to capture the interpretive experience of adopters of an IOIS. Finally, GT had been a widely used and recognized research method in the IS field (Orlikowski 1993; Urquhart 2001).

3. The grounded theory building process

3.1 Entering the field and conducting a literature review
Even though the research project described in this paper started in 2005, the first author was already familiar with the research setting the seaport of Barcelona and the phenomenon under research the implementation of an IOIS. From 2000 until 2004 the first author had analyzed diverse aspects of the implementation of an IOIS in this research setting. Some of the outcomes were a teaching case and a research paper (Rodon 2003; Rodon et al. 2007b). On the other hand, we reviewed existing literature on IOIS in seaports (Applegate et al. 1995; Applegate et al. 2001; Baalen et al. 2000; Damsgaard et al. 2001; Hock-Hai et al. 1997; Wrigley et al. 1994). Consequently, we started this study with some knowledge about the phenomenon and the research setting, which helped us to delimit our research problem. Next, as we chose to use GT method, we also analyzed papers in the IS field that had used GT. We then discovered that there were the Glaserian and the Straussian approaches to GT, and proceeded to read the seminal work of Glaser and Strauss (1967). We also complemented these authors seminal work with the reading of Goulding (2002), and other papers that compared and contrasted the Glaserian and the Straussian approaches. Finally, we opted for the Straussian version because it provides a more structured and linear approach to the methodology. Given that we were uninitiated in GT, we considered that the Straussian version would help more in guiding the data analysis. We then carefully read Strauss and Corbin (1990; 1998). In parallel, we conducted an initial literature review on IOIS implementation (Rodon 2006), which confirmed that that there was a lack of studies adopting a process-based approach to the study of IOIS implementation an issue that merited further investigation. An additional reason for choosing the Straussian version of GT was its approach to the use of the existing literature, much more in line with what we had already done in this respect. In fact, Glaser (1992) states that there is a need not to review any of the literature in the

www.ejbrm.com

72

Academic Conferences Ltd

Joan Rodon and Joan A. Pastor

substantive area under study. This dictum is brought about by the desire not to contaminateit is vital to be reading and studying from the outset of the research, but in unrelated fields (Glaser 1992, p. 32). On the other hand, Strauss and Corbin (1990) are more open to the role of existing literature, maintaining that all kinds of literature can be used before a research study is begun (Strauss et al. 1990, p.56). In our case, however, given the results we obtained from the literature review (Rodon 2006), we were not contaminated by existing theory as it did not suggest any hypotheses. We moved on to the next step in the research process without a preconceived theory or model in mind. Thus, we started with an area of study IOIS implementation, a focus on the implementation process, immersion in the research setting, and allowed the theory to emerge from the data. Figure 1 shows the stages followed in the GT building process.
3.2 Sampling

3.3 Data Collection 3.1 Entering the field and Literature review 3.4.1 Open Coding
if no theoretical saturation

if no theoretical saturation

3.4.3 Axial & Selective Coding

3.4.2 Memoing

3.5 Diagramming

3.6 Theoretical matching & generation

Figure 1: Grounded Theory building process

3.2 Sampling
Sampling is an ongoing part of the process of data collection and consists of selecting a sample according to the emerging theory. On the one hand, the collection of data is guided by the sample. On the other hand, the sample is redefined by the emerging theory and it is therefore impossible to predict the size of the sample prior to starting the study. We started with an open sample, which consisted of 11 companies operating in the Seaport of Barcelona. We selected those companies based on the following criteria: (1) companies playing different roles (e.g., freight forwarders, haulers, shipping agents, inland terminals); and (2) companies that had been successfully using the IOIS as well as companies that had failed or rejected to use the IOIS. These companies were selected after we interviewed managers of the IOIS and the Port Authority of Barcelona. We then wrote to the companies in the sample, inviting them to participate in the study. The letters were signed by the first author and the manager of the IOIS. We stated the aim of the study as: to enhance the understanding of the use of PortIC [the name of the IOIS] in order to define new information services and applications that would increase the efficiency and effectiveness of transactions, as well as to enhance the use of PortIC among port agents. The eleven companies all decided to take part in the study.

3.3 Data collection

We started interviewing the general managers of five of the companies in the sample. We asked these managers to give their views of the IOIS, their reasons for adopting (or rejecting) it, the expected benefits, and the problems they were facing in integrating with and using the IOIS. Once we had interviewed these managers, sampling became more focused. We then interviewed real users of the system (operational managers and clerks) and information systems personnel of these companies (managers, analysts, and developers). We asked them to explain their experience in using and integrating with the IOIS, paying special attention to relevant events and problems that shaped the way their companies were integrating their information systems with the IOIS or the way they were using the IOIS. From a conceptual standpoint, our research involved understanding human behaviour and action from the informants perspective. We prepared some general questions that served as guidance during the interviews, but these never determined the questions we finally asked.

www.ejbrm.com

73

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (71-82)

All the interviewees accepted being recorded. That facilitated our work during the interviews as we could exclusively focus our attention to listening and understanding informants. Interviews were conducted either in Spanish or Catalan. All the interviews were conducted and transcribed by the first author. Each interview was transcribed immediately after it was conducted and always within three days of the interview. This enabled us to add our field notes regarding our impression of interviewees body language, tone of voice, attitude, etc. The interviews were transcribed verbatim. Where the interviewee was less than articulate, we added our interpretation but left his words intact. Likewise, when during transcription we found any jargon that we did not understand, we called the interviewees and asked them for the meaning of those terms. Thus we adopted a constructivist approach to GT as we added our perception of how the interview went. In addition to interviews, we also collected data from other sources: the minutes of meetings, internal documents, company visits, and attending meetings. These data sources complemented interview data and helped guide the sampling.

3.4 Data analysis

We followed the principle of continuous interplay between data collection and analysis. During the whole analysis, we used the computer software QSR NVivo 2.0 to organize the vast amount of information collected, and to support our coding.

3.4.1 Open coding

In GT, analysis involves the assignment of concepts and themes to the data gathered. This process, called coding, consists of fracturing, conceptualizing and integrating data to form theory. A concept is an abstract representation of an event, object, or action/interaction that a researcher identifies as being significant in the data (Strauss et al. 1998, p.103). The analysis of the data began with a microscopic (sentence-by-sentence) examination of each interview (Strauss et al. 1990). The microscopic examination was the first step in the open coding process used to create initial codes for comparisons. During open coding, "data are broken down into discrete parts, closely examined, and compared for similarities and differences" (Strauss et al. 1998, p.102). We started codifying without predetermined ideas or a preconceived model. Initially, we used codes based on the terms used by informants (in vivo codes). The iterative process of data collection, coding, and analysis gave new insights into the research, helped us to formulate new questions in subsequent interviews, and helped indicate the most appropriate informants. Codes emerged through constant comparison (Glaser et al. 1967) of instances of data when we saw that they fitted with each other. Likewise, there was a continuous feedback with informants, which allowed us to look for new informants as well as to check whether emerging concepts fitted reality. After conducting about 15 interviews, we had already obtained 241 codes, which were grouped and organized into trees (Table 1). At this stage, we also started to conceptualize some of the open codes in the light if prior literature. For instance, in codifying the semantic interoperability conflicts that firms faced when they integrated their systems with the IOIS we used Park and Rams (2004) classification; or in codifying the extent of use of EDI we used Massetti and Zmuds (1996) facets for EDI usage. Coding aims to arrive at focal codes. However, we obtained many codes, which led our analysis in different directions. The reason why we obtained so many codes was that even though we compared the different instances of data, we did so in a descriptive rather than an analytical sense. We have two explanations for this: (1) our the lack of extensive and in-depth experience with GT, particularly in doing comparative analysis, and (2) the use of computer software (which on the one hand helped in organizing the codes but on the other led the researchers to focus more on descriptive coding than analytical coding). We could say that as the software greatly facilitates the management of a vast number of codes, researchers may have less incentive to work with a smaller number of abstract codes.

www.ejbrm.com

74

Academic Conferences Ltd

Joan Rodon and Joan A. Pastor

Table 1: Concepts that arose during open coding organized into trees
Concepts Adoption Open Codes Adoption reasons (mimesis within the port community, mimesis outside the port community, sense of community, to set an example, follow the clients, have good relations with the port authority); Expected benefits (agility, simplicity, speed, less work, less time, better service, better quality of work, better planning, reliability, improve productivity, no queuing, extend working hours) ; Non-adoption reasons (scope of standard, lack of preparation); Beginning; Sunk costs; Pressure; Critical mass; Readiness; Sense of responsibility Personal-Impersonal; Electronic channels; Fax exchange; Communication problems; Multiple channels; Paper exchange; Asynchronous Structure; Size; Scope of operations; Business; Commitment; Internal process; Customer focus; Dependence on headquarters; Location; Planning process; Relations with trading partners Dependence; More work; Improve service; Impact on business units; Perceived benefits; Partner relation; Return; Effects on non-integration; Side effects; Effects from bad operation; Consequences from channel duplication; Lack of coordination; Interdependent benefits; Changes of individuals work; Less data entry; Agility; Logistics; Container control; Spring effect Period; Post-implementation; Implementation problems; Adaptation; Testing; Training; Power Industry competitiveness factors; Other port communities; Barriers to competitiveness; Other IOIS; Relevance; Diversity of interests; Industry association Semantic interoperability (data representation conflict, schema isomorphism conflict, schematic discrepancy, data unit conflict, entity identifier conflict, generalization conflict); Problems with integration; Next steps with integration; Network integration; Meaning of integration; Internal integration; Interface integration; Evolution of integration; Database changes; Changes in applications; Manual integration; Changes in processes; Syntactic interoperability; Dedication; Particularities; Changes in ICT infrastructure; Automatic Integration; Path dependency; Pragmatics Value of IOIS; Future services of IOIS; IOIS role; IOIS perception; Problems of IOIS; Processing capacity; Strategy; Services; Pricing; Involvement in standardization; Ownership; Business model; Marketing Message analysis; Check content; Message format; Message generation; Message persistence; Message reception; Message translation; Error messages; Diversity of messages; Acknowledgement of receipt; Acknowledgement of processing; Check flow of messages; Message processing Responsibility for the process; Check status; Prior process; Physical process Participation; Adoption of the standard; Perception; Discrepancy; External influence; Standard committee; Visibility of information Technological change; Standardization of infrastructure; Database location; Software application; Connector; System in-place Problems with use; Perception form the user; Volume; Previous situation; New requirements; Aligning use; Non-use; Trading partners; Interdependence Security; Attitude towards the system; Knowledge about the system; Pressure from the customer; Uncertainty; Number of users; Lack of information; Mistrust; Resistance to change; Work overload

Communication Company

Consequences

Implementation Industry Integration

IOIS

Message

Process Standardization Technical level Use User

Another important element of this GT research process has been the use of memos, which are defined as the researchers record of analysis, thoughts, interpretation, questions and directions for further data collection (Strauss et al. 1998, p.110). Throughout the open coding, we wrote memos as a way to sketch and note our ideas, reflections, and concepts in parallel to data collection and open coding. The focus of our reflections was often the actual wording or formulation used by interviewees, which we then interpreted during the analysis. Whilst memoing, we immersed ourselves in the data so that we embedded the narrative of the participants in the research outcome.

3.4.2 Axial and selective coding

Once we had the codes organized into trees (Table 1), we moved into axial coding in our quest for a higher level of conceptual abstraction but in the process got lost in the data. Given the amount of codes we obtained during open coding, as well as their different levels of abstraction, we found difficulties in (1) reassembling these codes, (2) looking for code properties, (3) specifying relationships between those codes, and (4) finding an underlying story in them. We therefore decided to change our coding strategy. We repeated the analysis of all the interviews and other sources but this time abstaining from generating new codes, and focusing instead on writing memos in order to develop a picture of what the data meant in a

www.ejbrm.com

75

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (71-82)

broader sense. Two general questions guided this analysis: (1) what is happening in the data?, and (2) what patterns are occurring in the data?. Reviewing the memos from the previous stage proved particularly useful in answering them. During the axial coding we organized and combined the initial memos with the new ones and we wrote a paper (Rodon et al. 2006), which served as a preliminary presentation and validation of the findings of the research. At the same time, we adopted a new perspective towards the analysis of data: the paradigm model (Strauss et al. 1990), which is a tool to help contextualise the phenomenon by modelling the action and interaction strategies of the actors. Strauss and Corbin (1990) suggest using a coding family that consists of causal conditions, the phenomenon, the contextual conditions, the intervening conditions, the interactional strategies, and the consequences of these. We applied the paradigm model to our data and obtained the model in Figure 2. However, we considered that this model was a snapshot of the IOIS implementation rather than one which revealed the dynamics of the process. We then examined how other IS papers (Crook et al. 1998; Esteves et al. 2003) had applied the paradigm model. However, these authors had also used it to give a static picture of the phenomenon. For instance, Esteves et al. (2003) used it to present the factors that affected the implementation of an ERP, and Crook et al. (1998) used it to develop a theoretical model of EDI use.

Figure 2: Paradigm model As we were interested in a more dynamic view of the implementation of the IOIS, we decided to adopt a simplified version of the paradigm model developed by Strauss and Corbin (1998), which consists of conditions, action/interaction, and consequences. Strauss and Corbin (1998) regard a process as a series of evolving sequences of action/interaction that occur over time and space, changing or sometimes remaining the same in response to the situation or context (Strauss et al. 1998, p.165). Actions, which occur in response to changes in the context, may be strategic when they are taken in response to problematic situations, or routine when they are carried out without much thought (Strauss et al. 1998, p.165). That is, the authors conceptualize the process in terms of sequences or shifts in the nature of action and interaction between actors. Accordingly, rather than looking for properties of each code, which is what we initially tried to do in the axial coding, we purposely looked at action and noted movement, sequence, and change as well as how it evolvesin response to changes in context or conditions (Strauss et al. 1998, p. 167). As our analysis of the empirical data on the basis of conditions actions consequences moved forward, our initial research question also changed and became more focused. We realized that the data gathered were telling stories concerning not only the problems that firms faced when integrating their systems with the IOIS or when using the IOIS in their daily work, but also the IOIS managers actions in helping users adopt the

www.ejbrm.com

76

Academic Conferences Ltd

Joan Rodon and Joan A. Pastor

IOIS. Thus our research problem shifted from how the integration of pre-existing systems with the IOIS unfolded, to how IOIS management acted during the implementation process. We realized that the IOIS management acted to modify existing context e.g. lack of use of the IOIS or the consequences of their previous actions to adjust the IOIS to its adopters environment. Then we found five categories in our data which we interpreted as manoeuvres that managers of the IOIS made to support the integration of the adopting firms systems with the IOIS and to enhance the use of the IOIS. Following the analysis, we collected further data in order to validate these results and to reveal new types of managerial actions. The five categories that emerged were: Maintaining adopters autonomy, Accommodating to unintended uses, Managing the coexistence of exchange channels, Agreeing on the operational use of the system, and Balancing the degree of integration. We ended data collection when analysis revealed similar incidences and events occurred over and over again. Further data collection would therefore have added nothing to these five categories. In the end, 27 interviews were conducted over a 9month period (March 2005-November 2005). The five categories were subsumed into a core category: Managerial action in the implementation of an IOIS, which was the basis for the emergent theory. This core category concerns the set of found actions that managers performed to support the integration of adopters pre-existing systems with the IOIS and to enhance the use of the IOIS.

3.5 Diagramming: Presenting the findings

In GT, the illustration of the theory is done mainly during axial and selective coding, when categories are created and related. Besides memos, Strauss and Corbin (1998) also suggest the use of diagrams as a tool to gain analytical distance from materials and to present the results. Diagrams are useful to sort out the relationships between categories that arise during axial and selective coding. Strauss and Corbin (1998), however, do not propose a systematic way of presenting diagrams or of integrating them in GT, suggesting instead, that the analyst has to develop his or her own style and techniques (Strauss et al. 1998, p.223). In order to make up for GTs lack of illustration techniques, we used action diagrams (Axelsson et al. 2004) to structure and report on the five salient categories that emerged from the coding process. Action diagrams helped us graphically represent the categories that emerged from applying the simplified version of the paradigm model (Strauss et al. 1998). In the action diagrams (see Figure 3 and Figure 4 for an example) the diverse components of each category are related to each other as causal-pragmatic relationships. This means that the links between the different components in the diagrams are not deterministic, rather they are the result of interpretations of the actors (Axelsson et al. 2004). In the diagrams, we use different labels to indicate the role of each component within the diagram, such as conditions, actions performed by the IOIS management and consequences. Conditions refer to the set of circumstances, situations or assumptions, in which the implementation is embedded. We focus on the conditions that managers consider lead up to a problem or opportunity for the development and use of the IOIS. An action (or agency) refers to the stream of actual causal interventions of managers in the ongoing process of events-in-the-world (Giddens 1976). Conditions and actions are related as the formers may explain why and how persons or groups respond in certain ways (Strauss et al. 1998,p .130). Finally, consequences, which refer to the results from action, may be intended or unintended, and primary or secondary. An unintended consequence arises when an action that is performed with the intention of producing one consequence produces a different one, which can be conflicting, negative or positive. A primary consequence is the immediate intended result of an action. A secondary consequence is the result of a primary consequence, and can be either intended or unintended.

www.ejbrm.com

77

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (71-82)

Figure 3: Action diagram for the category Maintaining adopters autonomy (Rodon et al. 2007a)
1, 2 and 3. Conditions. There was a common belief among the promoters of the IOIS that adopters would introduce organizational changes to take advantage of the IOIS when they integrated with it. However, most of the firms in the study merely saw the IOIS as a glorified mailman, which received messages and forwarded them on to the right target. From the outset, the organizations in the study did not introduce changes in their organizational structures, and avoided making changes in their internal physical and documentary processes. They saw the IOIS as a tool that simply replaced the fax or former EDI systems they had, and they were not willing to implement any change to their business processes. Moreover, some firms complained about the flows and content of some of the messages initially defined in the standard and implemented by the IOIS. They considered those business processes differed from their in-house business processes. 4. Action. Then the standardization committee and the company managing the IOIS decided that the standard and the IOIS would adapt to the demands of adopters. For instance, in the case of exports, the standardization committee members agreed that truck drivers would not need to show any paper-based documentation to enter the inland terminal providing that the hauler had previously specified the driver in electronically submitted pre-arrival notification. However, once the IOIS had implemented the procedure, inland terminals objected to it arguing they had never worked that way. Then the IOIS made some changes, which lay outside standards scope, to persuade inland terminals. 5. Primary Consequence. This accommodation of the IOIS to adopters needs allowed the latter to avoid making changes to their business processes and databases. Interoperability was mainly accomplished through conversion tables but never by changing their data models. 6. Unintended consequence. Once firms started interfacing with the IOIS, they realized they required information to complete messages that was not always stored in their databases. However, the design logic behind the IOIS had been that of a virtual clearing house. The IOIS stored data from incoming messages and forwarded them to the specified target. That meant that, in some cases, firms were forced to store data from incoming messages that they had never used before but that they required in order to complete a message they had to generate later. These firms considered that although from the outset they had agreed with the design logic of the IOIS, they later realized it should have been different. In such case, the IOIS could add some value if it avoided firms retyping some data that they did not have at that moment or which they had already previously sent (action 3). Later, the IOIS accommodated to this requirement. 7 and 8. Secondary (intended) consequences. A secondary consequence of the action was that some firms, which had agreed to integrate with the IOIS several years before but had never been active users, boosted the use of the system. There was an increase in the number of messages exchanged. On the other hand, the maintenance costs of the IOIS increased as more customizations were implemented. 9. Unintended consequence. As some of the changes made by the managers of the IOIS had not been approved by the standardization committee, the IOIS implemented some procedures that deviated from what the standard laid down. Moreover, the customization of the IOIS to current customers has diminished the capacity of the IOIS to adapt to future changes. In addition, maintenance costs of the IOIS have increased (secondary consequences 7) given that any new measure approved by the standardization committee requires customization of the IOIS support.

Figure 4: Description of the action diagram Maintaining adopters autonomy (adapted from Rodon et al. (2007a))

3.6 Theoretical matching and generation

Once the five categories emerged, we began scrutinizing the IS and management literature for models, frameworks, or theories that might be relevant to our findings and thus enhance the theoretical sensitivity of

www.ejbrm.com

78

Academic Conferences Ltd

Joan Rodon and Joan A. Pastor

our study. On the one hand, based on our initial IOIS literature review, we were aware that the role of management in the process of IOIS implementation was an unexplored area in the IOIS literature. On the other hand, we presented our findings in workshops and research seminars, from which we got feedback that helped us in our quest for theoretical sensitivity. We then confronted the five categories that emerged from the analysis with three other streams of literature: misalignments (Leonard-Barton 1988), organizational change (Orlikowski et al. 1997), and emergent strategy (Mintzberg 1994). Organizational change literature (Orlikowski et al. 1997) tells us that when it comes to daily operations, IOIS management has to deal with emergent changes i.e. unexpected user appropriations of the system. Such emergent changes create unforeseen conditions and reveal misalignments (Leonard-Barton 1988), which trigger managers to perform adaptive responses (opportunity-based changes) in order to reinforce or attenuate them; in turn, as managers respond, new unintended outcomes and changes may emerge. We group the adaptive responses that arose from the case into five managerial manoeuvres (our five categories). Finally, we observe that these manoeuvres converge into two strategies in action (also referred to as emergent strategies (Mintzberg 1994)): attract users to bootstrap the IOIS, and keep the IOIS adaptable. As a result we obtained the model in Figure 5.

Figure 5: Process model that shows the emerging theory (adapted from Rodon et al. (2007a)) Through theoretical matching and generation we were able to (1) adapt the results of the GT study, and (2) provide a model that complements prior IOIS literature on whether the implementation process goes according to a plan and use of the IOIS meets expectations. Our model, which is grounded in the behaviour, words and actions of a set of professionals in a specific context, offers a plausible explication of the phenomenon under study. It is a process model that emphasizes the role of misalignments, and emergent changes during IOIS implementation, and depicts how management can cope with them.

4. Discussion and conclusion

Having conducted the aforementioned GT case study, we should like to highlight the following points, which emerged from our experience: Having some knowledge about the research setting and the research problem before a GT research starts is advisable. By conducting several empirical studies before we started the GT study, we were able to quickly immerse ourselves in the research context, which helped us define the initial sampling and minimize the jargon problems during the analysis. In addition, although we conducted a literature review prior to starting the GT study, we did not formulate any hypothesis or theoretical models. The reason was that one of the outcomes of the review was that our research problem had not been tackled by prior studies. The literature review also proved useful in the later stages of the research process by helping us enhance the theoretical sensitivity and grasp novel findings when they emerged. At the outset of open coding, analysis was unfocused. We did not have a clear idea of what they were looking for so they focused more on code generation than analysis. We obtained a large amount of open codes, which greatly increased the complexity of the data analysis. Likewise, we found difficulties in scaling up the codes into more abstract codes, in finding properties and categories, and in giving names to codes. Sometimes the names we used were too abstract, thus they lacked precision; other times, the names were too tied to the data and they thus lacked the status of a concept (Strauss et al. 1998). We consider that a researchers overemphasis on identifying codes without relating them and developing theoretical codes (Allan 2002) is a normal and sometimes inevitable pitfall in the initial stages of the analysis, especially when the research

www.ejbrm.com

79

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (71-82)

question and objectives are very open, and the researcher is fairly unfamiliar with GT. Doing open coding at the right level of abstraction requires certain experience and practice with the research setting and with GT. As our case shows, we had some knowledge about the research setting, and we were able to overcome our lack of practice in the later stages of the coding process. The researcher has to approach coding with an open mind, flexibility, and creativity. First, coding has to be performed as much as possible without predetermined ideas. The researcher has to keep an open mind when making sense of the data. He has to avoid looking for confirmation, in the data gathered, of any pre-conceived ideas he may have had before. Moreover, the researcher usually starts coding with a vague idea of the research problem and question. Later on, as the research process moves forward, the researcher is able to pin down the research question. Second, although the process of creating categories is mainly creative, the categories have to be grounded in data. In our case, even though during axial coding we coded all the data gathered again from scratch, the open codes and especially the memos written during open coding helped during axial coding by facilitating our abstraction process. Memos and diagrams became a relevant vehicle for our creativity. Third, researchers should approach coding with flexibility. As Strauss and Corbin (1990) recognize, researchers have to be flexible in the sense that while we [Strauss and Corbin] set these procedures and techniques before you, we do not wish to imply rigid adherence to them" (p. 59). It is the interpretation and flexibility of the researcher what really matters. Therefore, although GT provides with a set of procedures for coding, comparing, categorizing, etc, which may seem quite mechanical, the analytical process is highly interpretive by nature and fairly flexible to use. Theoretical sensitivity and matching occurs at two levels. At one level, theoretical sensitivity is enhanced by constantly reading in the same and other areas of research. At the other level, the development of concepts directs researchers attention to specific literature. For the latter to occur, it is useful to share ones findings with colleagues from ones own field or others. Workshops and conferences are just two settings in which findings might be shared. In our case, we started incorporating external literature once the five categories started emerging. Finally, a GT study is very time and resource-consuming, especially the processes of transcription, codification and constant comparison, thus any tool that supports the research is advisable. For instance, computer software may ease the process of cross-checking code generation; that is, they reduce the clerical work. Likewise, in our case the software facilitated the writing process as it was easy to browse the large number of codes we obtained. However, these tools are no substitute for the researcher making sense of the data given that abstractions are mental activities which cannot be formalized (Goulding 2002). As we have mentioned earlier, the use of computer software led us to focus on descriptive coding rather than analytical coding (Strauss et al. 1998). Because this type of software is designed to facilitate data handling, inexperienced researchers tend to overemphasize code generation rather than abstraction. Although we consider computer software to be very useful throughout the research process, it cannot replace intuition and abstraction of the researchers or their need to make judgement (Hunter et al. 2005). This paper has presented our personal experiences in conducting a GT case study that follows the Straussian approach. We describe the research process by focusing on the problems, decisions and paths taken throughout the process. We consider this paper provides a practical understanding of how the Straussian version of GT can be applied and which may serve as guidance for novice researchers. The paper also shows how we applied the paradigm model (Strauss et al. 1998) in order to develop a process model of IOIS implementation, and gives some novel ideas on how to report on and present the results of GT studies that adopt the paradigm model.

References
Allan, G. (2002) "A critique of using grounded theory as a research method," Electronic Journal of Business Research Methods, Vol. 2, No. 1, pp. 1-10. Applegate, L.M., Neo, B.-S., and King, J. "Singapore TradeNet : Beyond TradeNet to the Intelligent Island," Harvard Business School Publishing, 1995, pp. 1-14. Applegate, L.M., Siong, N.B., Bartlett, N., and Chang-Leow, D. "PSA: The World's Port of Call," Harvard Business School Publishing, 2001, pp. 1-34.

www.ejbrm.com

80

Academic Conferences Ltd

Joan Rodon and Joan A. Pastor

Axelsson, K., and Goldkuhl, G. (2004) "Theory Modelling - Action Focus when Building a Multi-Grounded Theory," 3rd European Conference on Research Methods in Business and Management, Reading, UK, pp. 1-11. Baalen, P.v., Oosterhout, M.v., Tan, Y.H., and Heck, E.v. (2000) Dynamics in Setting Up an EDI Community, Eburon Publishers, Delft, The Netherlands. Blumer, H. (1969) Symbolic Interactionism: Perspective and Method, University of California Press, Berkeley. Crook, C.W., and Kumar, R.L. (1998) "Electronic data interchange: a multi-industry investigation using grounded theory," Information & Management, Vol. 34, No. 2, pp. 75-89. Damsgaard, J., and Lyytinen, K. (2001) "Building Electronic Trading Infrastructures : A Public or Private Responsability?," Journal of Organizational Computing & Electronic Commerce, Vol. 11, No. 2, pp. 131-151. Eisenhardt, K.M. (1989) "Building theories from case study research," Academy of Management Review, Vol. 14, No. 4, pp. 532-550. Elgarah, W., Falaleeva, N., Saunders, C., Ilie, V., Shim, J.T., and Courtney, J.F. (2005) "Data exchange in interorganizational relationships: review through multiple conceptual lenses," The DATA BASE for Advances in Information Systems, Vol. 36, No. 1, pp. 8-29. Esteves, J., Pastor, J.A., and Carvalho, J. (2003) "Organizational and National issues of an ERP Implementation in a Portuguese Company," IFIP 8.2+9.4 conference, Athens, pp. 139-154. Fernandez, W.D. (2004) "Using the Glaserian Approach in Grounded Studies of Emerging Business Practices," Electronic Journal of Business Research Methods, Vol. 2, No. 2, pp. 83-94. Giddens, A. (1976) New Rules of Sociological Method, Stanford University Press, Stanford. Glaser, B.G. (1992) Emergence vs. Forcing: Basics of Grounded Theory, Sociology Press, Mill Valley, CA. Glaser, B.G., and Strauss, A.L. (1967) The Discovery of Grounded Theory: Strategies for Qualitative Research, Aldine Publications Company, New York. Goulding, C. (2002) Grounded Theory: A Practical Guide for Management, Business and Market Researchers, SAGE Publications Ltd, London. Hock-Hai, T., Bernard, C.Y.T., and Kwok-kee, W. (1997) "Organizational Transformation Using Electronic Data Interchange : The Case of TradeNet in Singapore," Journal of Management Information Systems, Vol. 13, No. 4, pp. 27. Hunter, K., Hari, S., Egbu, C., and Kelly, J. (2005) "Grounded Theory: Its Diversification and Application Through two Examples From Research Studies on Knowledge and Value Management," Electronic Journal of Business Research Methods, Vol. 3, No. 1, pp. 57-68. Latour, B. (1987) Science in action: how to follow scientists and engineers through society, Harvard University Press, Cambridge, MA. Leonard-Barton, D. (1988) "Implementation as mutual adaptation of technology and organization," Research Policy, Vol. 17, No. 5, pp. 251-267. Lyytinen, K., and Damsgaard, J. "What's wrong with the Diffusion of Innovation Theory?," in: Diffusing Software Products and Process Innovations, M.A. Ardis and B.L. Marcolin (eds.), Kluwer, B.V., Deventer, The Netherlands, 2001, pp. 173-190. Massetti, B., and Zmud, R.W. (1996) "Measuring the extent of EDI usage in complex organizations: Strategies and illustrative examples," MISQ Quarterly, Vol. 20, No. 3, pp. 331-345. Mills, C.W. (1959) The Sociological Imagination, Oxford University Press, New York. Mintzberg, H. (1994) The Rise and Fall of Strategic Planning, Prentice Hall, Hertfordshire. Myers, M.D. (1997) "Qualitative research in information systems," MISQ Quarterly, Vol. 21, No. 2, pp. 241-242. Orlikowski, W.J. (1993) "CASE Tools as Organizational Change : Investigating Incremental and Radical Changes in Systems Development," MIS Quarterly, Vol. 17, No. 3, pp. 32 (309-340). Orlikowski, W.J., and Hofman, J.D. (1997) "An Improvisational Model for Change Management: The Case of Groupware Technologies," Sloan Management Review, Vol. 38, No. 2, pp. 11-24. Park, J., and Ram, S. (2004) "Information Systems Interoperability: What Lies Beneath?," ACM Transactions on Information Systems, Vol. 22, No. 4, pp. 595-632. Ramanathan, S., and Rose, J. (2003) "Rationalizing, Probing, Understanding: the Evolution of the Inter-Organizational Systems Adoption Field," Hawaii International Conference on Systems Science, IEEE Computer Society, Big Island, pp. 254-263. Rodon, J. (2003) "PortIC: The Electronic Trading Infrastructure of The Port Community of Barcelona," North American Case Research Association (NACRA), Tampa, Florida. Rodon, J. (2006) "Methodological and Conceptual Review of Inter-Organizational Information Systems Integration," 14th European Conference on Information Systems, Gteborg, pp. 1-12. Rodon, J., Pastor, J.A., and Sese, F. (2007a) "The Role of Emergent Strategies in Managing the Implementation of an Industry IOIS," 67th Academy of Management Annual Meeting, Philadephia. Rodon, J., and Ramis-Pujol, J. (2006) "Exploring the Intricacies of Integrating with a Port Community System," Bled eConference, Bled, Slovenia, pp. 1-15. Rodon, J., Ramis-Pujol, J., and Christiaanse, E. (2007b) "A process-stakeholder analysis of B2B industry standardisation," Journal of Enterprise Information Management, Vol. 17, No. 1, pp. 83-95. Strauss, A.L., and Corbin, J. (1990) Basics of Qualitative Research. Grounded Theory Procedures and Techniques, Sage, Newbury Park, CA. Strauss, A.L., and Corbin, J. "Grounded Theory Methodology," in: Handbook of Qualitative Research, a.Y.S.L. Norman K. Denzin (ed.), Sage Publications, 1994, pp. 273-285. Strauss, A.L., and Corbin, J. (1998) Basics of qualitative research : techniques and procedures for developing grounded theory, Thousand Oaks, CA: Sage.

www.ejbrm.com

81

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (71-82)

Urquhart, C. "An Encounter with Grounded Theory: Tackling the Practical and Philosophical Issues," in: Qualitative Research in IS: Issues and Trends, E. Trauth (ed.), Idea Group Publishing, Hershey, 2001, pp. 104-140. Wrigley, C.D., Wagenaar, R.W., and Clarke, R.A. (1994) "Electronic Data Interchange in international trade : frameworks for the strategic analysis of ocean port communities," Journal of Strategic Information Systems, Vol. 3, No. 3, pp. 211-234.

www.ejbrm.com

82

Academic Conferences Ltd

Managing the Fear Factor (or how a Mini-Viva Assessment can Improve the Process of Learning for International Students)
Susan Sayce Bournemouth University, UK ssayce@bournemouth.ac.uk
Abstract: This paper is about an exploration of international business students learning through the use of mini-vivas as a form of assessment. It also includes an investigation of the meaning of a mini-viva for students who come from a wide range of nationalities. Pedagogical research has indicated that using this form of summative assessment for large cohorts of international students may be problematic (Carless 2002). However, experimentation with this model of assessment with MA business students in research methods has indicated that mini-vivas can enhance and consolidate the learning potential of international students. So in effect this paper is also about explaining why this has happened in relation to students learning. Keywords: international students, research methods, mini-viva, deep learning, assessment

1. The Mini-Viva and research methods

1.1 Introduction
An exploration of international MA students learning through using mini-vivas as a form of summative assessment is central to this research. Pedagogical research has indicated that using this form of assessment for large cohorts of international students may be problematic (Carless 2002). However, the researchers experimentation with this model of assessment with a large cohort of international business MA students has indicated that mini-vivas can enhance the learning potential of international students. Previous research into the mini-viva as a form of summative assessment indicated that students place great emphasis on gaining a mark rather than focusing on the process of learning (Greer 2001). There appears to be a shortfall in research about how a diverse group of international students perceive a spoken type of evaluation and how it influences their learning despite claims that the most influential element in promoting learning for students is the assessment (Boud et al 1999). Often it is how, and when, the student is assessed that will determine how they learn (Struyven et al 2005). However, having to aurally defend a piece of individual written work seemed to encourage more students to engage in more reading and understanding of the issues pertaining to research design. This is despite concerns over students levels of anxiety in having to defend their ideas in a secondary language and the worry that less articulate students could be disadvantaged in learning through this type of summative assessment. The mini-viva assessment in this research was linked to the teaching of research methods to postgraduate business students who are undertaking a MA courses in Administration and Marketing. These courses are designed for international students and have around sixty international students between them. However, this was the first time that many of these international students had done this type of spoken assessment so in order to understand teaching outcomes we had to understand how different international students interpret a mini-viva assessment and how it helps them to understand methodological issues. These concerns influenced the design of the research objectives which were; to conduct an investigation of the meaning that international students give to oral assessments, to explore how students overcome issues of language in verbal discussions and finally investigating how a written and verbal summative assessment impacted on students learning.

1.2 Deep and surface learning

There has been in many universities a drive towards encouraging students to become more independent learners, to encourage students to take on a more active role within learning (Leathwood 2006). Adopting a deep approach to learning can shift responsibility for learning from the institution to the student, with the student taking control of their learning (Rust 1998:71). It has been shown that in order to facilitate this situation students need to adopt a more critical and questioning stance to their studies in order to adopt a more reflective deeper approach to learning (James 2000). Deep and surface learning originated in the 1970s and identified two different levels of processing knowledge. Those students who used surface-level processing focused on the text itself, memorising as much as possible while those who utilise a deep-level approach were aiming to understand the underlying meaning of that text (Case and Marshall 2004).
ISSN 1477-7029 83 Academic Conferences Ltd Reference this paper as: Sayce, S. (2007) Managing the Fear Factor (or how a Mini-Viva Assessment can Improve the Process of Learning for International Students). The Electronic Journal of Business Research Methods Volume 5 Issue 2, pp 83 - 92, available online at www.ejbrm.com

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (83-92)

Concern has been raised with the rigid bipolar nature of these two categories in light of the increasing numbers of international students who have experienced different traditions in learning. It has been well documented that Asian students in particular are more likely to adopt a surface approach to learning. This has often been linked to their different educational experience where much teaching is by rote (Rust 1998). Trigwell and Prosser (1999: 169) address this issue, they argue that what is key to students adoption of a particular learning style is not the method -such as learning by rote- but the intention behind the method that is important. They claim that it is this intention that influences students adoption of their learning approach. For Trigwell and Prosser the key question in understanding students learning approach is an understanding of what are the students attitudes to learning as well as how they learn.

1.3 The relationship between learning and assessment

An increasing emphasis on students learning is driving research into an evaluation of the relationship between assessment and students learning. This has led to recognition that students are often more focused on attaining a grade than learning gains (Carless 2002). It has been suggested that the assessment system can drive students into coping strategies and elicit a surface approach to learning, which may impact on the way in which a student approaches future learning. Argris and Schons model of single and double loop learning outlines two approaches to learning which share some similarities with surface and deep learning. A homeostatic or single loop of learning is similar to surface learning. Students adoption of this process means that their learning entails giving the right answer for favourable results; this usually results in people usually taking a similar strategy time after time (Yorke 2003). Double loop learning shares more characteristics with deep learning because this is about individuals reflecting on their practice and experiences, challenging existing ideas and letting other approaches emerge (Brooks 2003). Boud (2000) and Struyven et al (2005) point out that assessment practices in higher education often do not equip students for learning even though it is how, and when, the student is assessed that helps determine how they learn. Boud (1995) argues that summative assessment should not be predominantly about evaluation but also should be about learning because poor assessment can result in students losing confidence in their abilities and skills thereby undermining learning. Ramsden (1992) argues that if we understood how students learnt we would recognise that uniformity of methods may not permit the student to display their knowledge and originality. He claims that what is needed is more variety in assessment in order to promote greater responsibility for self-direction in learning (1992:192). In terms of assessment, student learning is a function of the intrinsic quality of the form of assessment. Case and Marshall (2004:606) consider that how students perceive the course is an important determinant of the choice of approach. Another key component here is the students interpretation of the task and the assessments context (Boud 1995, Ramsden 1992). Learning interpretations by students are also strategically influenced by the context of learning as well as the content, so if the student has a range of assessments to complete at the same time they may have less time to focus on something they consider to be meaningful. Often this result in students appearing only to focus on the grade they achieve through their assessment than on the process of learning (Greer 2001). Often the experience of learning for many students is not reflected upon once the assessment is completed and there is no strategy beyond the task where the focus is exclusively on the words or text needed to complete the assignment (Ashcroft and Foreman-Peck 2002, James 2000).

1.4 The Mini-Viva assessment and learning

The mini-viva assessment was conceived as an experimental summative assessment. Its aim was to prepare students for conceiving, designing and planning an international MA dissertation project, to recognise how different methodological elements needed to be fitted together. In order to facilitate this, the assessment was constructed in two parts. Firstly, the students wrote a short paper outlining a potential dissertation project justifying their theoretical and methodological choices. Two weeks later they had a ten minute mini-viva with their research methods tutor. The focus of the session was on exploring what they had written and questions were asked to help the student overcome design weaknesses and methodological inconsistencies thereby increasing or decreasing (depending on their mini-viva) their grade. If in terms of assessment, student learning is a function of the intrinsic quality of the form of assessment (Boud 1995, Ramsden 1992). Then in this case the intrinsic element is talking through a written design to enhance student learning, to give the student immediate feedback on their understanding of their own individual research design. This was considered to be particularly desirable in relation to research methods as the

www.ejbrm.com

84

Academic Conferences Ltd

Susan Sayce

tutors were also trying to get the students to engage with the topic of research design so when they did their individual dissertations they felt more confident about making methodological decisions. Surface learning appears undesirable in relation to understanding research methods because students are required to apply and justify methodological knowledge in designing and implementing their own specific dissertation project. A deeper approach permits the students to reflect on their own interpretation of research methodology and permits them to extract understanding of the key concepts they need to discuss in their later proposal and dissertation. Thus they need to have an awareness of how theoretical interpretations around research methodology can be applied to different contexts including previous work and learning (Ramsden 1992, Prosser and Trigwell 1999) Encouraging deep approaches to learning is important but there was also a concern to recognise that students also have the skills and capacities to assess themselves (Boud 1995: 39). In this research, this is not so much about evaluating marks but in reflecting on, and acknowledging, their own strengths and weaknesses in planning their research design and designing their own individual research methodology. Carless (2002) reasons that when students explain and justify their ideas verbally in assessments such as a mini-viva it can help consolidate students learning. It also permitted us to explore their understanding of research methods because students needed to understand methodological issues before they engaged with writing their dissertation research proposal. While the above clearly outlines the direction the tutors want the students to move towards, the key question here is what messages are we transmitting to the students in our mini-viva assessment? Good assessment entails finding an appropriate method and using it sensibly in relation to the subject matter. However, there will always be unintended consequences in assessment that results in students choosing to reject or adopt a more conceptual approach to learning. Case and Marshall (2004:606) consider that how students perceive the course is an important determinant of the choice of approach. This suggests that we need to know what approach to learning the students are themselves choosing if we are to understand the students experience of the mini-viva. However, the interpretation and perceptions of the students are not always predictable. Indeed for international students from different background this could be problematic because learning is an act, which can result in students exploring ideas and seeing the world in new ways (Boud 1995:43). Thus for this research a key component is the students interpretation of the task and the assessments context.

2. Methodology
Interpretative philosophy is fundamentally concerned with exploring the complexity of human sense making. Thus interpretative studies focus on understanding phenomena through the meanings that people give to them (Flick 2004). In interpretative qualitative studies the aim is to reveal the embedded holistic interconnected nature of human experiences in an attempt to extract the significance of this interaction (Bryman and Bell 2003). The adoption of the methodological paradigm helps determine the research methodology, which in this instance was qualitative. A qualitative approach seemed better suited to an investigation of the different meanings and interpretations that different international students, using their own frames of reference, give to a mini-viva (Collis and Hussey 2003). The research consisted of semi-structured interviews with eighteen international MA business students as they progressed to working on their MA dissertations. Subsequent to a pilot round of exploratory interviews with three international students an agenda was drawn up for the semi-structured interviews. A performance criteria was considered at this stage but it was decided to interview the students at a later stage in the dissertation process and thereby let the student outline (or not) whether the grade they received was still influential in relation to their experience of doing a mini-viva. Semi-structured interviews were chosen as they provide a useful means of access to exploring sense-making of human events (Berg 2004). It also gave space for the researcher to follow up views that related to individual differences and experience of learning styles. Quota sampling was used to pick equal numbers from students studying International Business Administration and International Marketing Management, a third of the cohort. The gender balance of the sample reflected the gender balance of the two courses. The students chosen were also deliberately picked so that voices from Burkina Faso to Indonesia, from Venezuela to Jordan could be heard. There were no native English speakers in these cohorts so all students were speaking in a secondary language. The sample included students from Europe, Asia, Africa and South America. However, because of the variety of nationalities and the smaller sample it was decided not to group the students according to national

www.ejbrm.com

85

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (83-92)

characteristics about students abilities and learning. For example, if we had had a European group this would have included one Norwegian, two Turkish students, one Cypriot student, one Russian and Spanish student who all could have had a very different experience of teaching and learning. Thus each student was treated as an individual case study to explore the creation of meaning that students give to spoken assessments. Thus the research was about investigating students viewpoints for similarities and differences between them as individual international students rather than confirming or denying national stereotypes about learning (Neuman 2003). The students Programme leader was also included in the interviews to draw in a wider range of student feedback. The interviews were then subject to content analysis in which the researcher strove to listen to the text, the manner in which they stated their views to improve understanding of how the students constructed their perspective on assessment and learning in relation to the mini-viva.

3. Mini-Viva research findings: The fear factor

Anxiety is an important component in this type of spoken marked assessment. Ashcroft and Foreman-Peck (2002) make the point that in order to move towards a deeper learning approach anxiety needs to be contained. At the beginning their was the concern that different categories of nationalities would find this easier than others that those from a European background would be less apprehensive than those from South Asian countries that were more used to learning by rote. However, these divisions were not supported by the students responses, in all the interviews all nationalities made the point that they were nervous about speaking in a secondary language, they were all conscious of the shortfalls in their English language ability whether they were Thai, Indian, Spanish or Jordanian. All of the students regardless of nationality commented that they had little experience of this type of summative assessment in relation to explaining and justifying their own decision-making processes in research design and interestingly enough it was a European student albeit from Uzbekistan who found this aspect the most difficult. However, two students from South Asia, one from Thailand and one from Taiwan did comment that grammatically that this type of assignment was easier for those from Europe but they also considered that developmentally it was important that all students went through this experience. This viewpoint was shared by all the students (apart from one Turkish student whose different views are discussed later). It is summarized below by one South American and one South Asian student: It is very difficult to express the ideas when I speak compared to when I write down my own idea because you can change it, you can express things in different ways. When you are talking you think Oh God, I am using the incorrect verbs to say something. I hope that the person understands me but I am not sure. But afterwards it made me feel really good. If I can do something like that interview in the future, Ive got some feedback from that experience and I gained some experience as well. I am learning at the same time and that is very important for me and my future development (Female Columbian student) I would tell other students not to be too nervous and express what you are thinking, It doesnt matter how good your English is just express with any words you know. They want to know what you are thinking about so just talk (Female Korean Student) What became significant from analyzing the students responses was an apparent willingness to take on responsibility for the language issue. Several of them argued that at the MA level they should be able to converse on their topic to an interested other. They also argued that while the answers may not be in perfect English they should be able to understand questions based on what they had written and be able to express their ideas for their topic The students considered that they should be able to do this type of exercise, because they were expected to have a certain level of language before becoming MA students although that didnt mean that did not find it nerve-racking. The Programme Leader considered this point, while doing a spoken assessment like a mini-viva was challenging he felt that business MA students needed challenging because you could see an equivalence in work situations so this made the experience worthwhile, a viewpoint that several of the students supported in their comments.

4. The meaning of a Mini-Viva

Students views of what a mini-viva meant were diverse but basically fitted into two categories, a spoken exam and a formal interview, the formality of these interpretations were suggested by the summative nature of the assignment. The group of students who saw the mini-viva as more of an interview (while nervous) were less concerned about the grade and more concerned with the exercise. Several of them alluded to the

www.ejbrm.com

86

Academic Conferences Ltd

Susan Sayce

role-play they had seen and based their interpretation on that or quizzed other students who had done a viva. They saw the exercise more about expressing ideas and keeping talking. This slightly less formal interpretation of the meaning of the assessment maybe influential in student approaches to learning because this group were less anxious compared to those who saw it as an exam (Boud 1995). A mini-viva was just another way of doing an assessment, but doing it deeper because you can ask me what I am doing, I was nervous about 50 percent but all the questions were about what Ive written. When I am talking about it I am thinking about it and it gets me further and further and I realize I only know a bit of this and I have to do more. (Female Indonesian student) However, those students who perceived a mini-viva more as a spoken exam had different implications for their learning experience. A female Turkish student claimed that that many of her Turkish friends worried too much because they thought it was like an examination. They became defensive and stubborn about their ideas when the tutor suggested that they needed to change elements of their research design because they considered that you were threatening their grade. A male Turkish student was interviewed and his views seemed to fit into this category as he outlined how he thought it was unnecessary to ask what was on his mind and he just repeated what he had written rather than give another opinion. Here there seems to be clear lack of understanding of the learning outcomes of the mini-viva and this has impacted negatively on the students experience. The misinterpretation has resulted in this student becoming defensive and not seeing the mini-viva as an opportunity to improve on his written work. However, this was the only student interviewed who reacted negatively to the mini-viva despite achieving a respectable grade. One of the objectives of this research was to examine whether there were other language aspects in relation in a spoken assessment that need to be considered when experimenting with a mini-viva. One student linked this to the open expression of ideas. From Uzbekistan he outlined how people from his country were not used to openly discussing their views and justifying their decisions. This viewpoint he related to his experience of the former Communist regime where open discussion was discouraged. While he was disappointed in his mark because of this difficulty he did recognize that this was perhaps part of the learning process of doing a higher Western degree. Interestingly other students who too had experience of more authoritian regimes such as the Korean student did not raise this issue. While the Uzbek student was very committed to learning he was disappointed with his performance in the mini-viva. He pointed out that in Uzbekistan the teaching is very different with very few presentations and very little direct questioning. However, he reflected on why (for him) he had not performed so well and then acted to improve his knowledge of research methodology. He worked hard to address his lack of knowledge in his proposal and consequently when reflecting on his development subsequent to the proposal was able to track his learning development in research methodology in relation to speaking about his ideas and was proud of the progress he had made in being more proactive and taking responsibility for his ideas. The above students process of learning bears similarities with Argyriss concept of single loop and double loop learning (Brooks 2003). This connects to his reflection on the mini-viva experience, highlighting particular problems and then spending time resolving them, using an alternative approach if necessary (reading more material, contacting the tutor as much as possible). This indicates that he was adopting a double loop learning strategy in order to facilitate deep learning. However, one point to consider in relation to this adoption is as the student outlined he needed the time to do this, for him this was a couple of months, this included time to reflect and then take steps to put right deficiencies and then again reflect on progress. This also indicates the value of doing the interviews when the students are at a later stage of the research process and can examine their experience without focusing exclusively on their grade. Contrary to expectations those interviewed from more collective South Asian countries such as Taiwan, Indonesia and Thailand did not consider it difficult to explain their idea. In the main they saw speaking about their research design more as an opportunity to advance their specific skills, particularly their business skills. However, a Thai student raises a point about language and politeness that had not been considered and underlines the importance of body language and presenting a clear interest in the students work when doing an assessment such as the mini-viva to help address cultural distance: Actually I feel a little bit nervous because it never happened in my country before. I worry that if I do not speak friendly, because I am not fluent that is not friendly, but I try to speak clear and make the interviewer understand what I am saying. The interviewer did understand because I could see that

www.ejbrm.com

87

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (83-92)

from their reaction. After the viva I feel that more is clear and I feel more confident. I think during the viva the questions made me think deeper and deeper. For me this is good. (Male Thailand student) This is not necessarily a negative because outcomes can also be affected by the students intention to learning. Different experiences need not necessarily preclude the students learning if they wish to learn (Trigwell et al 1999). The above statement seems to suggest that the Thai student is able to value a different way of learning through speaking and gains confidence from that even through he was initially very concerned about this process. The above illustrates how one can not discount cultural influences when experimenting with mini-viva assessments but also at the same time how one can not necessarily rely on stereotypical groupings of nationalities to explain cultural influences. Interestingly this did not seem to have been a distinct problem in the mini-viva because as the students commented the teaching team were very used to teaching a variety of international students and clearly outlined how the viva would be more of a discussion about their paper than a gladiatorial contest. This approach was made explicit in the mock mini-viva which was conducted in an earlier lecture and is discussed by a Norwegian student: I had never done anything like a viva before. I got what I expected because we had this viva role play before. But it was kind of scary because it is assessed and it is part of the research design mark and thats scary but you always get nervous when you are assessed. I think a lot of the students found it scary. To me the exercise was a great help. I was concerned about getting difficult students you cant answer. For example, I was asked about what samples of questions I was going to ask in my interviews. I hadnt really though about it so when you get those type of questions you start sweating and try to find the right answer. I think the question side helped me to focus in on what were the key elements. I think it opened my eyes to what is possible for the proposal. I found out where it was lacking (Male Norwegian student) Several students referred to the mock-viva when talking about their lack of experience of this type of assessment and how it helped to alleviate some of their nerves. This allied with the students willingness to take on responsibility for dealing with language issues overcame most cultural issues in connection to doing a spoken assessment. There also had be a lot of reiteration about what we expected from the mini-viva to reassure the students about participating in a spoken assignment about research design. It was encouraging in the research that despite the varying cultural backgrounds many of the students, including ones from countries such as Taiwan, Cyprus, India, Jordan and even Uzbekistan, saw the mini-viva as being part of a developmental process that students wishing to gain a British MA degree should undergo.

5. Writing, speaking or both?

A mix of assessment styles has the potential to provide a more accurate picture of students knowledge and students achievement (Ramsden 1992). The Programme leader also considered that a mini-viva was a good fit in helping to establish individual students' degree of understanding in relation to research methods as they need to apply these skills later. However, what was interesting was how the students began to split into two camps in relation to learning preferences. What was significant was that this division did not correlate with national cultural stereotypes in that some students from a more authoritarian collective background such as the Indonesian and Korean students claimed that they fitted into the preferred talking group while our Taiwanese and Thai students considered that they preferred writing as discussed by our Indian student: When it came to talking about what I had written I wasnt sure how I would perform I was a bit jittery since what I was going to talk about was assessed. Compared to writing things it is more difficult because you have to be very sure of what you are saying because once you have said it, it is all done. I was looking at it from a different perspective but when the questions were put to me it was like OK. I can do it that way, that I should look at it because I knew the right things but I was not putting them in the correct places. For example, the relevance of the literature review to the research project, after the viva I could understand that it is the soul of the research project. (Male Indian student). These views contrasted with those students who preferred to speak including one of the African students:

www.ejbrm.com

88

Academic Conferences Ltd

Susan Sayce

In the viva you need to make sure you know what you are talking about maybe writing is different from talking and explaining so the viva is a chance to make sure youre on the right track because you hear yourself talking and you hear the question and you can say to yourself thats right, Ive justified my choice. Maybe I am more comfortable with talking than writing so it was quite good (Female Burkina Faso student). There were also some students who did not mind either approach. The gender split of those whose preference of talking as against writing was also considered. However, the split was fairly even between the two genders in this respect and as such did not evidence a gender difference. It is possible to connect the split between those students who preferred to write down their research design and those who relished the opportunity to speak about it as being more related to individuals different learning styles (Entwhistle and Entwhistle 1991, Greer 2001) For example, the latter research grouping, those who liked speaking, saw the mini-viva as an opportunity to improve the mark they had gained from the written design while the others saw the mini-viva as an exercise to be gone through. But as the students comments as a whole indicate there was something about both writing and speaking that helped to the release the students learning potential regardless of nationality, gender and cultural background

6. Discussion: The Mini-Viva assessment, experience and learning

It seems possible to suggest from this research that there is an intention to learn from doing this type of spoken exercise. If as we outlined earlier a key component of students taking a more conceptual, deep learning approach is the students interpretation of the task and the assessments context (Boud 1995, Ramsden 1992). You could speculate that the students saw the language issue as a barrier to overcome but once the mini-viva was completed they could use this experience to bolster their confidence when talking to their supervisors, and interested others about their research design, some took it further and thought it could be useful when doing interviews in the wider business world. So on the whole the students were, as Trigwell et al (1999) suggests, taking on responsibility for their learning in relation to their language skills. However, this does not mean that the students do not need to be supported in relation to doing a spoken assessment otherwise you run the risk of the anxiety levels being too high to facilitate learning (Ashcroft and Foreman-Peck 2002). Consideration must be given by the tutors of the stressful nature of doing this and certainly from the comments, such as relax it is not an exam or she made me feel comfortable so I could relax seems to suggest that this was happening. A role-play in relation to the mini-viva helped to take away some of the fear factor for many of the students. However, the findings indicated that some students still struggled with what a mini-viva meant and saw it as a serious event. But apart from the Turkish male students all the students saw the outcome more positively despite sometimes having received a lower grade. So how can the difference be explained? The male Turkishs students response seems on the surface to support Bouds (1995) claim that students who do poorly in assessments can lose confidence (Yorke 2003). But I would argue that what is key here is what the students do in response to not doing well in the viva and this is relational to how they perceive learning. If we examine the following quotation from the male Turkish student we can begin to assess his learning approach: After the viva everything goes wrong and suddenly you start worrying about your topic because I dont want to fail the course. I dont think as the proposal as being connected to the viva, I take them separately and in each subject I try to do my best. I dont want to fail. I dont think about the future because there is no guarantee to doing well in the future. (Male Turkish Student) The above quotation suggests that this student is employing a surface approach to learning. He is doing the task to his best endeavour and then moving on without reflecting on that experience and linking it to improving his learning. The danger of taking a surface approach is that this takes the learning element out of context and isolates it to one piece of work and links it only to the grade. Thus the student is less likely to appreciate the purpose of the mini-viva exercise (to explore the students own understanding of research methods and design). Therefore he is less likely to reflect on what he needs to do to improve his understanding of research design, and indeed this was the only student who thought that the viva was unnecessary. So for me the key question here is Ramsdens (1992) question what is this students intentions when it comes to learning and how does this influence how he learns?

www.ejbrm.com

89

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (83-92)

If approaches to learning are relational, and adoption of a deep learning approach is dependent on the students interpretation of teaching, it is relative to their independence in making choices about learning and a clear awareness of the goals and standards required in the subject-research methods (Ramsden 1992, Trigwell et al 1999), then you could argue that in connection to the male Turkish student this has not occurred. However, a Taiwanese student who failed the summative assessment seems to negate Bouds (1995) claim that students who do poorly in assessment can lose confidence. He has reacted much more positively to take on responsibility for learning and to reaching the necessary standards to enable him to progress: The viva was a positive thing. I was quite disappointed with the viva but it meant that I worked hard for the proposal and passed the proposal. The major advantage was that it made me think, research methods for international students like us is new. So we need to think about subjects like methodology, I had done projects in Taiwan but not methodology. Before the viva I was really confused afterwards I found more books about methodology whereas before I didnt think about it. (Male Taiwanese student) Failing a student in a summative assessment maybe problematic because some students may feel this reflects adversely on their ability, they could lose confidence and so shift toward a position of learned helplessness rather than an independent learner (Yorke 2003). However, I would argue that the above quotation illustrates how the Taiwanese student is shifting towards a deeper approach to learning. He is making independent choices about learning and putting these in place in order to reach the required methodological standards. If we return to Ramsdens key question; what is the students intentions and how does this influence how a students learns? Here you could claim the student has recognised that research methods is not just about the mini-viva assessment but is about learning and understanding research methods. For him the mini-viva is positive because he has realised that he took on responsibility for developing his own understanding and knowledge and gone to the library to facilitate his understanding and thereby become more of an independent learner. Argris and Schon (cited in Brooks 2003) suggest that most individuals (and organisations) are good at single-loop learning where people operate within accepted limits set objectives and monitor their performance and then keep repeating the same process. They argue that this can limit learning progression. For them double-loop learning is more desirable but this only occurs when individuals reflect on practice and recognise that some situations require an alternative solution than current practice as indicated by the Taiwanese and Uzbek student who could trace their development and had gained confidence from doing this. While it was considered that this type of assignment would probably favour the more confident articulate student it was interesting that doing a mini-viva could assist other less articulate students. There was something about combining two learning styles that has helped to release the students' learning potential. For those students who struggled with this exercise there was the recognition that they had to do more work. It made them go deeper into the subject than they would have necessarily chosen to go. They had to shift from surface learning to deep learning because they had to fill in their gaps of knowledge, which had become very evident from the mini-viva. They had to really understand what they had written if they wanted to do a successful proposal for the MA dissertation. All the interviewees thought that combining both a written and spoken exercise about their research design reinforced their learning apart from the male Turkish student. This suggests to the researcher that these students are intending to adopt a deeper approach to learning because strategically they recognize that this topic is not going to go away (Case and Marshall 2004, Prosser and Trigwell 1999).

7. Conclusion
The aim of this paper was to explore the meaning that international students gave to mini-vivas. As indicated this was split between those who considered it was a spoken exam and those who thought it was more like a formal interview regardless of nationality. An outcome of these perceptions was that those students who considered it was more like a spoken exam suffered increased anxiety than the other students. However, the students intentions about learning were key to understanding how the students interpreted the results of their mini-viva regardless of their interpretation of what a mini-viva was (Case and Marshall 2004). As a result of the mini-viva some international students have begun to move towards a deeper level of learning, to enhance their learning of research methods.

www.ejbrm.com

90

Academic Conferences Ltd

Susan Sayce

The adoption of a deep learning approach is dependent on the student recognizing the value of research methodology goes beyond the original context of the assessment (Boud 1995). The students need to recognize that a surface approach may not give them all the methodological skills to justify choices not only in research design but for their later proposals and dissertation. This takes us back to theorists who argue that in designing assessment the focus is not on transmitting information but understanding why you individually need to know this information (Trigwell et al 1999). The mini-viva has to be about changing students conceptions (Prosser and Trigwell 1999). This has been achieved for some students who recognized that there was a shortfall in their knowledge and they had to improve if they wanted to progress. It is possible to argue here that by engaging in a spoken assessment where the students expand on their understanding of research methods for their own individual design highlights vividly strength and weaknesses. Just the act of speaking about their decisions can help them to internalize concepts that they have written about as well as generate new ideas (Ramsden 1992). When this research began it was thought that issues of language would be predominant but as the above indicates although this is an issue, it has not overset the value of doing a mini-viva. Indeed the combination of both a written and spoken exercise may help students in different ways to release their learning potential, help them to realize that they are independent learners, and as such they can overcome language problems despite their anxiety.

Acknowledgement
The researcher received a teaching and learning grant from the Higher Education Academy of Business, Management, Accounting and Finance based at Oxford Brookes University in the UK.

References
Ashcroft K and Foreman-Peck 2002 Managing Teaching and Learning in Further and Higher Education Routledge London Berg BL 2004 Qualitative Research Methods Pearson Education Inc Boston Boud D 1995 'Assessment and Learning: Contradictory or Complementary chapter in Knight P Assessment for Learning in Higher Education Kogan Page London Brooks I 2003 Organisational Behaviour Prentice Hall London Bryman A and Bell E 2003 Social Research Methods, Oxford University Press Oxford Carless DR 2002 The mini-viva as a tool to enhance assessment for learning Assessment and Evaluation in Higher Education 27:4 pp353-363 Case J and Marshall D 2004 Between deep and surface: procedural approaches to learning in engineering education contexts Studies in Higher Education 29:5 pp605-615 Collis J and Hussey R 2003 Business Research Palgrave Macmillan Basingstoke Entwhistle NJ and Entwhistle A 1991 Contrasting forms of understanding for degree examinations and its implications Higher Education 22 pp205-227 Flick U 2002 An Introduction to Qualitative Research Sage London Greer L 2001 Does changing the method of assessment of a module improve the performance of a student? Assessment and Evaluation in Higher Education 26:2 pp127-138 James D 2000 Making the graduate: perspectives on student experience of assessment in higher education chapter in Filer A ed. Assessment: social practice and social product Routledge London Leathwood C 2006 Gender, Equity and the Discourse of the Independent Learner in Higher Education Higher Education 52 pp611-633 Neuman WL 2003 Social Research Methods Allyn and Bacon London Prosser M and Trigwell K 1999 Understanding Learning and Teaching Open University Press Buckingham Ramsden P 1992 Learning to Teach in Higher Education Routledge London Rust C ed 1998 Improving Student Learning: Improving Students as Learners, The Oxford Centre or Staff and Learning Development, Oxford Struyven K , Dochy F and Janssens S 2005 Students perceptions about evaluation and assessment in higher education: a review Assessment and Evaluation in Higher Education 30:4 pp325-341 Trigwell K, Prosser M and Waterhouse F 1999 Relations between teachers approaches to teaching and students approaches to learning Higher Education 37 pp.57-70 Yorke M 2003 Formative assessment in higher education: Moves towards theory and the enhancement of pedagogic practice Higher Education 45 pp.477-501

www.ejbrm.com

91

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (83-92)

www.ejbrm.com

92

Academic Conferences Ltd

The Catalog Usability Questionnaire Adoption and Validation of a Usability Scale for Print-Catalogs
Sebastian Schulz, Gunnar Mau, and Gnter Silberer Institute of Marketing and Retailing, University of Gttingen, Germany sschulz8@uni-goettingen.de gmau@uni-goettingen.de gsilber@uni-goettingen.de
Abstract: The usability of interfaces is mainly tested via standardized-questionnaires or via less standardized processoriented verbal protocols. Standardized questionnaires have got the advantage of generating reasonable indices with quite low effort. In contrast, process oriented verbal protocols offer the opportunity of gaining more thorough insights into the interaction and the problems that might come up. While efficient standardized methods for the measurement of online media with their non-linear structure already exist, there are no such methods to measure the usability of linear structured offline media. However, it can be assumed that usability of offline print media plays an important role for a brands image and the buying behavior of customers. The objective of our study was to adapt a scale to measure usability of online shops to print catalogs in order to measure their usability. A field study (n = 60) and a laboratory experiment (n = 92) have been made to test the reliability and validity of the adapted scale. Participants were asked to evaluate the catalogs of two travel companies by using the adapted questionnaire. Concerning the reliability, in both studies the scale showed good internal consistency. The values of Cronbachs alpha were comparable to results in studies using the original scale with online shops. Concerning the validity, results show correlations between usability and buyers intention as well as recommendation. Furthermore, the usability scores of both catalogs differed as experts (n = 20) expected. Moreover, the questionnaire was validated using the results of verbal protocols that were collected simultaneously in the laboratory study. In example, the correlation between usability scores and problems that came up during the interaction was proven. Overall, the results support the usefulness of the adapted scale to measure the usability of print-catalogs. Implications for research and management will be discussed. Keywords: usability scale implementation, verbal protocols, print-catalogs

1. Introduction
If potential customers decide to purchase a trip or holidays, they turn mostly to print catalogs as a source of information (cf. Fodness & Murray 1999). If the decision to buy is based to a large extent upon information from the catalog, its design and, above all, usability appear to be significant. After all, only the information the customer can find in the catalog can be considered in the purchase decision. In the study presented here, we concentrated on travel catalogs as the catalog plays a particularly significant role in this industry. The catalog serves as a permanent reference book for customers. Furthermore, it is significant for online commerce as nearly 40 percent of online buyers first use the dealers catalog prior to ordering on the Internet (cf. Bhv 2005). Anyone who is searching for instruments to measure the usability of print catalogs in scientific publications will be disappointed. Hodgson (1993: 51) pointed out that the scientific research on usability in the print field primarily concentrates on newspapers, advertisements, magazines and pocket books (similarly Schmeier et al. 2005). Only a few publications have addressed the topic of catalogs or travel catalogs at all and only a small proportion of these the effect of texts and other elements in catalogs and tourist brochures (e.g. Hodgson 1991, 1993; Goossens 1994). The referral to in-depth interviews and focus groups as evaluation instruments, as are often used, appears to be as time-consuming as it is costly. Evaluation methods which work with verbal protocols seem far more promising. The low level of standardization may cause various problems (e.g. in the coding and evaluation of data) but it offers the opportunity of gaining more thorough insights into the interaction. A standardized interview instrument is not yet available. In contrast, there are many studies on the topic of the usability of web sites and online catalogs and, furthermore, also on tourist web sites. Meanwhile, science and management can fall back upon validated
ISSN 1477-7029 93 Academic Conferences Ltd Reference this paper as: Schulz S., Mau G., and Silberer G. (2007) The Catalog Usability Questionnaire Adoption and Validation of a Usability Scale for Print-Catalogs. The Electronic Journal of Business Research Methods Volume 5 Issue 2, pp 93 - 104, available online at www.ejbrm.com

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (93-104)

instruments for the measurement of usability in this field. The use of standardized usability questionnaires has established itself as both a simple and inexpensive evaluation method (e.g. Konradt et al. 2003). After years of the transfer of instruments and methods of offline market research in an online context, a validated online usability questionnaire (UFOS Scale, Konradt et al. 2003) is tranferred to the evaulation of the usability of two travel catalogs in the scope of this empirical study. In doing so, the use of such a survey instrument will be presented and, furthermore, the quality of the instrument and its impact on final factors will form the focus of this article. In addition, there will be a comparison between verbal protocols and the scale and implications for a combined use.

2. Usability and studies on catalogs and brochures

2.1 Results of the usability research
In recent years, the topic of usability has been addressed in many scientific and management publications. Usability primarily shifted into the focus of science and practice through the spread of modern information technology and questions on the operability of computer applications and web sites. However, usability refers to more than merely software and web site applications, as Nielsens (1993) definition suggests: usability is the measure of the quality of the user experience when interacting with something - whether a Web site, a traditional software application, or any other device the user can operate in some way or another. In addition, usability does not represent a one-dimensional construct. In science and practice, the usability understanding of the international norm ISO 9241(ISO 1998) has prevailed. According to this understanding, an object or application is user-friendly if its aims can be achieved effectively, efficiently and satisfying. The three criteria effectiveness, efficiency and satisfaction should also be regarded for the usability of a travel catalog as dimensions of usability. Table 1: Categorization of usability evaluation methods (Wilhelm & Yom 2004; Stoessel 2002; Sweeney et al. 1999; Mack & Nielsen 1994; Nielsen 1994)
Theory-based methods Expert-based methods (analytical) Heuristic Evaluation Cognitive Walkthrough Check lists User-based methods (empirical) Surveys Verbal Protocols Concurrent verbalization Retrospective verbal reports Eye Tracking

Cognitive Models & Architecture GOMS (Goals, Operators, Methods and Selection Rules)

In the field of usability evaluation methods, different categorizations have developed. One which applies the most is the breakdown into theory-based, expert-based (analytical) and user-based (empirical) evaluation (cf. Sweeney et al. 1993; cf. Mack & Nielsen) (see table 1). For example, expert-orientated processes would be the cognitive walkthrough or the heuristic evaluation; all forms of survey represent user-based processes. However, Hodgson (1991) only distinguishes qualitative (e.g. in-depth interviews, face-to-face interviews) and quantitative (e.g. standardized survey) methods to investigate usability. In this study, two methods of the user-based processes are used: firstly, survey and, secondly, retrospective verbal protocols.

2.2 State of the research on catalogs, travel catalogs and brochures

Fodness and Murray (1999) examined whether the travel catalog plays a role at all in the search for information. In their study, they develop a Tourist Information Search Behavior model and test it empirically. In the process, seven different groups, each with different information search behavior, are identified. The results show that catalogs and brochures are almost referred to as an information source in a decision. Five out of seven groups refer to catalogs in the information search. An initial study on catalog purchasing behavior and the influence of situative determinants was published in 1974 in the Journal of Marketing (cf. Reynolds 1974). In an experimental study, the main influencial factors were examined for purchasing with catalogs. For this purpose, the author investigated nine factors using a factor analysis from 107 items, which he used as a basis for comparison. The results show that regular catalog purchases are motivated by the need for simplicity and convenience. Summarizing, the author states that it appears that catalog buying is more a result of catalog offerings, as revealed in the negative attitudes toward local shopping conditions expressed by young, affluent risk takers who comprised the frequent buying group. (ibid.). Even if the results of this study reveal important findings regarding the

www.ejbrm.com

94

Academic Conferences Ltd

Sebastian Schulz, Gunnar Mau, and Gnter Silberer

situative determinants of catalog behavior, they also demonstrate the singular role of the catalog content. In this respect, it goes without saying that a sensible presentation and simple usability are of great importance but this was not considered in Reynolds study. Goossens (1994) goes into the effect of elements in the catalog more specifically and examines the effect of texts and pictures in travel catalogs on the external information behavior in an experimental study. In doing so, the author varies the size of the pictures within a 2x2 factorial design and the emotional character of the texts in a travel catalog. In each case, 5,000 copies were sent to a total of 20,000 households which had been chosen at random. The author does not determine any differences dependent on the stimuli. He attributes this to the fact that the variations of the catalog are not perceived as different (cf. Goossens 1994). Recording the usability and the the effect of the elements through a standardized questionnaire would also have been possible here but was not used. Most notably, Hodgson dealt with Tour Brochure Design from the beginning of to the mid-90s. In the process, the author reports that only a small proportion of companies make the sufficient effort and utilize enough financial resources to design their catalogs in a user-friendly fashion (cf. Hodgson 1993). As, in his opinion, usability is a central factor in the success of catalogs, evaluation should be a permanent feature of catalog development. Therefore, Hodgson suggests first and foremost qualitative methods (focus groups, indepth interviews), which he regards as more sensitive (cf. Hodgson 1991). However, as the cost and efforts are often very high in qualitative methods, he also quite rightly addresses quantitative methods, such as standardized questionnaires. These are admittedly less explorative and sensitive but with the appropriate effort can provide useful indications as to the problem areas. The literature research carried out shows that the scientific research deals with a wide variety of problems from the catalog sector. However, qualitative methods dominate the evaluation methods. A standardized tool for the effective and efficient examination of the usability in due consideration of the cost-use relationship has not yet been developed.

3. Empirical Study
Within the scope of our study, the adoption of a standardized online usability questionnaire (UFOS scale) to print catalogs was tested. Furthermore, the combination with verbal protocols was examined. This study was carried out in collaboration with a large German travel agent. From a methodical point of view, the following objectives and research questions underlie the study: How can the standardized questionnaire for online evaluation be adapted for print catalogs? Does the scale fulfill the quality requirement of reliability, i.e. does it produce reliable values? Does the questionnaire facilitate valid measurements? Do the questionnaires and the verbal protocols produce the same result? What other results can be obtained from the verbal protocols in the evaluation of a print catalog? Firstly, the UFOS scale, which represents the basis for the instrument developed here, will be presented below and the adaptation to the travel catalog sector elucidated. The presentation of verbal protocols and their application in the field of catalog evaluation ensues in section 3.2. This is then followed by the design of the study in section 3.3 and the presentation of the central results in section 3.4 and 3.5. The section closes with a summary.

3.1 A questionnaire to measure the usability of print catalogs

The UFOS scale developed by Konradt, Wandke, Balazs and Christophersen (2003) forms the basis for the questionnaire used here. The underlying definition of usability comprises the three dimensions, effectiveness, efficiency and satisfaction, addressed in section two (Konradt et al. 2003; ISO 9241). In the development of the questionnaire, the scientists conducted extensive research using existing usability questionnaires (e.g., SUMI, PUTQ; Kirakowski 1996) and considered these and the results obtained. Furthermore, they conducted an extensive validation of the instrument in an online context. Primarily due to the integration of different questionnaires, the systematic derivation and the available validation, we opted for the UFOS scale as a basis for the questionnaire to measure the usability of print catalogs. The original questionnaire consists of 47 questions, which are structured into seven categories (general usability impression: 14 questions, information on the basic conditions: 7 questions, product search: 7 questions, cart handling: seven questions, product overview: 4 questions, self-description ability: 4

www.ejbrm.com

95

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (93-104)

questions, product description: 4 questions). The original items were adjusted to the evaluation in the print catalog field. Items which did not include relevant aspects for catalog usability were eliminated. After the modification and adaptation of the original questionnaire, a total of 23 questions in five categories remained. According to expert opinion, these cover the relevant aspects of the usability of travel catalogs. The questions used and the corresponding categories are illustrated in table 2 (Catalog-Usability-Questionnaire, CUQ). The emphasis is on the category general usability impression, from which 13 of the 17 questions were adapted and adopted, and the category club overviews, which largely consists of items from the former category product overview. The category cart handling was dispensed with in the modified questionnaire. Table 2: Catalog-Usability-Questionnaire (CUQ)
Scale: In each case, 5-Point Rating Scales from strongly agree to strongly disagree 1. General Usability Impact (13 Items) - This catalog makes the search for a club vacation / hotel easier - I find using the catalog too complicated - You can quickly gain an overview of the catalog - Dealing with the catalog is easy to learn - The catalog encourages me to explore it further - The search for a trip / holiday can be completed swiftly - The catalog offers useful advice on its use - It is easy to find your way around the catalog - The catalog is constructed consistently - The catalog is well-structured - I can always recognize important information easily - I can always find the information which is important to me easily - The design helps in the use of the catalog 2. Additional Information on the Travel Conditions (1 Item) - I always have an overview of the costs which arise 3. Search for Clubs / Hotels (1 Item) - It is easy to search for a suitable club / hotel for me 4. Club / Hotel Overviews (7 Items) - The overview pages of the hotels / clubs are clear - The categories of hotel / club overviews offered are selected sensibly - The symbols on the overview pages are clear - I always have an overview about the price of the club / hotel - I can verify whether the club corresponds to my expectations - I am satisfied with the descriptions of the club / hotel - I find the pictures used helpful 5. Self-description (1 Item) - I do not require any explanations to use the catalog
a

Items original in German language

3.2 Verbal protocols to measure the usability of print catalogs

Besides the modified questionnaires, a further user-based method of evaluation was used within the scope of the evaluation in the form of verbal protocols (cf, section 2). Verbal protocols offer the possibility of a process-associated assessment of cognitions (Ericson & Simon 1993). The concurrent verbalization can be differentiated from the retrospective verbal protocols (ibid.). In the case of the concurrent verbalization, the participant expresses all thoughts on usability during the interaction with the catalog; for the retrospective verbal protocols, the participant is asked to comment on his thought process in hindsight (Bttner & Silberer 2007). The retrospective video-cued thought protocols represent a development on retrospective verbal protocols. In the process, the subsequent verbalization is supported by a video of the interaction process which was recorded during the actual activity (Silberer 2006). The methods can be distinguished through their proximity to the process, without binding cognitive capacities through verbalization during the interaction.

www.ejbrm.com

96

Academic Conferences Ltd

Sebastian Schulz, Gunnar Mau, and Gnter Silberer

Figure 1: Test procedure in the retrospective video-cued thought protocols In the assessment of the cognitions on usability, first the participant is filmed during the interaction with the catalog (see fig. 1 A). Afterwards, the video is shown to the participant on the PC and all thoughts which he/she expresses on usability are recorded (see fig. 1 B). A combined file is subsequently put together from the video and the commentary. The videos produced facilitate both a qualitative and quantitative assessment of the usability problems expressed. For the quantitative assessment, the participants comments have to be classified in a category system. The category system used here was established from literature and expert discussions with practitioners (Nielsen 1993). For an objective classification of the positive and negative statements, it should comprise a manageable number of selective categories. The following category system was used: Statements on: Category 1: impression and expressiveness of the pictures and photos Category 2: clarity and design of the colors Category 3: general color scheme Category 4: information content of the text Category 5: overall composition of the catalog

3.3 Design of the study and operationalization

The test design comprised a preliminary and a main study, which in turn consisted of three sub-studies. Within the context of the preliminary study, both catalogs were evaluated by experts with regard to different usability dimensions in order to be able to validate later results. In doing so, the experts had to gage the overall usability, effectiveness, efficiency, satisfaction and the 5 main-categories of the usability questionnaire (general usability impression, additional information, search for clubs, club overviews, selfdescription ability) on 5-point rating scales. The main study comprised a laboratory study with n1 = 92 test participants, a field study with n2 = 60 participants and a final expert survey with n3 = 20 participants. The laboratory study served the questionnaire test under controlled conditions and the gathering of the retrospective video-cued thought protocols. With the second study, users of catalogs in three travel agencies were interviewed. The final expert survey served to confirm the results found on the basis of the Catalog-Usability-Questionnaire.

www.ejbrm.com

97

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (93-104)

Figure 2: Catalogs used in the study. Cover and two pages (Aldiana GmbH and Robinson-Club GmbH) The Catalog-Usability-Questionnaire was used in all three sub-studies. The procedure was also practically identical in all three sub-studies: after the measurement of different control measures, such as experiences and socio-demography, in a pre-questionnaire, the participants familiarized themselves with a travel catalog for about 12 minutes (see fig. 1a) before finally completing the usability questionnaire. In the laboratory study, the completion of the post-questionnaire was still interrupted by playing the video with the participants verbalization. The attitude towards the catalog (3 items), the operator (3 items), the club vacation (3 items), the satisfaction with the catalog (5 items) and the intention to reuse the catalog and recommend it were operationalized as independent variables through one question each and tested. Although both catalogs originate from club vacation operators in comparable price segments, they differ in both size and presentation. The Robinson catalog is a total of 251 pages in length, whereas the Aldiana catalog is 171 pages long. By way of example, important aspects in which both catalogs differ greatly and which could lead to different evaluations should be shown in the design: the Aldiana catalog works exclusively with black writing on white or light-colored backgrounds and also uses different colors within a category (not only across categories). Furthermore, it boasts an overall clear structure and the distinct differentiation of categories (e.g. club overviews) through the use of boxes. In addition, the Aldiana catalog contains several pictures per side (see fig. 2 Aldiana). In contrast, the color of the writing and background in the Robinson catalog is uniform (e.g. background light blue and writing dark blue), there are no rigid structures or boundaries in the overview pages and it focuses primarily on one picture per page (see fig. 2 Robinson).

3.4 Results: Usability scale

The results on the quality of the scale (CUQ) are presented below. In this, the reliability of the scale is examined as the initial quality criterion and the inter-correlations considered. The results are each compared to the existing findings of Konradt et al. (2003) from the extensive validation in an online context.

www.ejbrm.com

98

Academic Conferences Ltd

Sebastian Schulz, Gunnar Mau, and Gnter Silberer

3.5 Reliability of the scale

All three sub-studies in the main study attest high to very high reliability (Crombachs a .928) for the scale used for usability using all 23 items. The reliabilities found are situated in the same area (Crombachs .960) as the reliabilities found by Konradt et al. (2003) for the usability questionnaire in an online context (see table 2). Table 3: Reliability of the overall scale usability main study
Overall Scale Usability Labortory study Field study .928 .954 Experts .958 Konradt et al. (2003) .960

Crombachs

Furthermore, as in the case of Konradt et al. (2003), the inter-correlation of the scale was examined. The results found for the modified scale for usability used here also show that all scales correlate highly to extremely significantly with the overall scale (.51 r .98) and positively with each other (.34 r .85, see tables 3 and 4). The same pattern was also apparent in Konradt et al.s original scale (see table 3) and speaks in favor of the quality of the modified scale. Table 4: Inter-correlation of the overall usability scale with the single sub-dimensions
Overall Usability Value Sub-values for Usability from the item categories (Dimensions) 3 4 5 2 1
General Usability Impression Additional Information Search Club Overviews Self-Description

Laboratory study .96*** Field study .97*** Experts .98*** Konradt et al. (2003) .92*** * p < .05 ** p < .01, *** p < .001

.58*** .58*** .79*** .62***

.74*** .66*** .81*** .84***

.85*** .94*** .89*** .71***

.59*** .66*** .51*** .79***

Table 5: Inter-correlation of the single sub-dimensions (Items) in the Laboratory (n1) and Field Study (n2)
Sub-Values Usability from the Item Categories (Dimensions) 1 2 3 Gen. U. impression Add. information Search Lab | Field Lab | Field Lab | Field 2 Add.Information .60** | .50*** 3 Search .60*** | .61*** .56** | .34** 4 Club overviews .67*** | .60*** .40** | .40** .57*** | .59*** 5 Self-description .57*** | .85*** .56** | .56*** .51*** | .41** * p < .05 ** p < .01, *** p < .001 ; 4 Club overviews Lab | Field

.49*** | .53***

3.6 Construct validity

Construct validity signifies how capable an instrument is of illustrating the existing differences. These differences were determined on the basis of the independent expert survey prior to the actual main study with various instruments of usability evaluation. Consultants in travel agencies served as experts here. We can assume that they are capable of estimating how good a catalog is based upon their years of experience. Furthermore, the expert survey also represents a recognized instrument of measuring usability. A total of 17 experts (travel agents, average experience M = 12.8 years) evaluated the two catalogs in the dimensions effectiveness, efficiency, satisfaction, overall usability and the five main-categories of the questionnaire used here independently of the main study. In the process, the Aldiana catalog was evaluated as user-friendly in all categories except Additional Information on Travel Conditions. Both catalogs fall into the good to satisfactory categories and the experts quantitative estimations lie close together. This raises the question as to whether the scale used here is capable of illustrating the differences in the two catalogs found by the experts in the preliminary study. This is clearly the case (cf. table 5). The Aldiana catalog is also evaluated to be considerably better than the Robinson catalog by the consumers in the sub-studies 1 and 2. Furthermore, the scale is capable of recording significant or potentially significant differences in the catalogs in all categories. Although the evaluations of the catalogs by the experts are similar in all categories in the preliminary study, the questionnaire is capable of differentiating between both catalogs (cf. table 5). Significant differences could primarily be found in the areas in which the catalogs differ greatly due to the design, the club overviews and the possibility of finding information quickly (cf. table 5).

www.ejbrm.com

99

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (93-104)

The catalog which emerged from this study as user-friendly was evaluated again by experts within the context of a third study. This served the repeated examination of the results obtained by experts. This time, the complete usability scale was used (identical to that of sub-studies one and two). The results obtained support both the estimations of the consumers and the global estimations of the pre-study. Table 6: Significant differences between the catalogs found by the CUQ; samples n1 and n2
Significant differences between the catalogs df, T Usability Overall Value (all 23 Items) General Usability Impression (13 Items) This catalog makes searching for a club vacation / hotel easy You can quickly gain an overview of the catalog The catalog is well structured I can always recognize important information quickly I can always find information which is important to me quickly Additional Information on the Travel Conditions (1 Item) Search for Clubs / Hotels (1 Item) Club / Hotel Overviews (7 Items) The club / hotel overview pages are clear The club / hotel overview categories offered are selected sensibly The symbols on the club / hotel overview pages are comprehensible I always have an overview about the price of the club / hotel I find the pictures used helpful Self-description (1 Item) 130, 2.38 139, 1.82 148, 1.96 148, 1.87 147, 1.83 147, 1.92 148, 2.49 142, 1.82 147, 1.70 137, 2.75 146, 2.91 146, 1.67 148, 2.59 143, 2.95 147, 1.87 148, 1.68 p .019 .070 .049 .063 .070 .063 .014 .071 .062 .007 .004 .098 .011 .004 .063 .095

On the basis of verbal protocols, the amount of the usability expressed for each catalog was determined for each participant. Furthermore, the positive statements on usability were also counted. As we can gather from table 7, the overall usability value of the catalogs (recorded via the questionnaire) correlates positively with the number of usability problems expressed from the verbal protocol (r = .489). This shows that the participants usability problems identified in the questionnaire are also to be found in the verbal protocols. The participants who assessed the usability as worse also expressed more problems in the verbal protocols. Furthermore, the number of positive statements correlates negatively with the overall usability value, as was to be expected, but the correlation is not significant. Table 7: Correlation between the overall usability value and the number of the negative and positive statements on usability on the basis of verbal protocols (n= 92)
Negative Statement on Usability (Usability Problems) r p Overall Usability Value .489*** < .001 n.s. not significant, *** p < .001 ; Positive Statements on Usability r -.153 n.s. p .184

3.7 Criteria validity

Subsequently, the results of the correlation analysis between the usability and effects are presented. These effects, like reuse and attitudes, are classed as an indicator for the criteria validity. The two intentions and the attitudes were selected as criteria, because other studies proved their correlations with usability (Lee and Lee 2004; Kohavi and Round 2004; Bttner, Schulz, and Silberer 2006). Furthermore, users satisfaction with the catalog usage was considered because it is a partial aspect of usability according to the ISO norm and, consequently, strong correlations are indicators of scales criteria validity (Schweibenz & Thissen 2003).

www.ejbrm.com

100

Academic Conferences Ltd

Sebastian Schulz, Gunnar Mau, and Gnter Silberer

Table 8: Correlations of the usability with effects of the usability

Usability Laboratory study Field study r r Reuse .477*** .576*** Recommendation .546*** .617*** Satisfaction .697*** .893*** .665*** .842*** Attitude to catalog .331** .530*** Attitude to operator * p < .05, ** p < .01, *** p < .001; a) Variable not surveyed in the study Construct / Item Experts r n/aa) n/a .721** .849*** .497*

The results support the validity of the scale. As we can see in table 8, all the correlations found are highly to very highly significant and are situated on a medium to high level. With r .697, the satisfaction correlates highly significantly with the usability, which was to be expected on the basis of the definitional proximity (cf. ISO 9241). Furthermore, the correlation between both attitude factors and the usability is significant. In this, like in a study on trust, risk and usability by Bttner, et al. (2006), the attitude towards the catalog correlated highly significantly with the usability at a high level with r .665. The expected influence of the usability on the attitude towards the operator can also be confirmed, whereby this is less striking. If we assume that the catalog is the first and only source of information on the operator for many participants, the results obtained are explicable. In addition, the fact that a significant correlation between intention and usability could be found in the studies by Lee and Lee (2004), Konradt et al. (2003) and the study by Bttner et al. (2006) in an online context and can also be confirmed here in the context of print catalogs supports the validity of the scale (see table 5).

3.8 Results: Verbal protocols

Besides the measurement on the basis of the questionnaire, we will briefly address the results which were obtained with the aid of the verbal protocols. A total of 380 expressed usability problems were recorded in the context of the coding of the verbal protocols of the 92 participants. On average, every participant thus expressed M= 4.30 (SD= 2.98) usability problems. However, how the number of expressed usability problems is distributed across the categories in (see category system 3.2) is interesting (see table 9). Category 1 impression and validity of the pictures occupies first place with 113 negative statements (29.7%) and thus appears to have the highest proportion of a negative usability assessment. The high number of namings also indicates a significant need for action in the corresponding category. Category 4, information content of the texts (25.0%), and category 5, overall composition of the catalog (23.9%), subsequently follow. Similarly, the previous results can also be measured using the questionnaire scale. At this point, the verbal protocols nevertheless offer advanced possibilities for analysis. It is possible to address relevant categories in more detail through a qualitative evaluation of the statements. Consequently, the problems can be identified concretely and changes implemented more easily on the basis of the statements. Not only do we know that there is a problem with the pictures; we also know the reason for the problem. Corresponding statements of the participants are e.g. the pictures only show small sections of the beach..., the resort has been photographed from unfavorable positions... or the photos seem old.... Table 9: Number of Negative Statements on Usability
Category 1: Impression and expressiveness of the pictures & photos 2: Clarity and design of the colors 3: General color scheme 4: Information content of the texts 5: Overall composition of the catalog Percent % (Number of namings) 29.7 (113) 18.7 (71) 2.6 (10) 25.0 (95) 23.9 (91)

Finally, it can be noted that the conclusions and evaluation possibilities of the verbal statements are very big. Above all, a qualitative analysis of the statements or a process-related analysis (e.g. sequence analysis) of the interaction process promises substantial information.

www.ejbrm.com

101

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (93-104)

4. Conclusion
Both the reliability and the validity support the scale developed here to measure the usability of print catalogs. In the process, the results on reliability coincide with those of the previous studies (Konradt et al. 2003). Furthermore, the scale succeeds in illustrating the expected differences in the main study on the basis of an expert survey and analysis of the design. The results based on the verbal protocols also coincide with measurements of the usability recorded through the questionnaire. In addition, the results of the correlation analysis between the overall value of the usability and effects also coincide with known results from the field of usability research. The highly significant correlations found between the intentions and the usability are primarily interesting for industry and especially marketing management. Consequently, this result underlines the importance of a user-friendly catalog design and the necessity for testing. After all, a user-friendly catalog encourages the customer to recommend it, reuse it and ultimately book a vacation. The use of verbal protocols facilitates a deeper insight into the actual interaction process of the user. So it can help identify more detailed usability problems than an overall evaluation via a usability scale. However, the effort required for the data coding is above all immense.

5. Implications for the process of catalog development and control

The aim of the study presented here was the development of an inexpensive and handy instrument to test the usability of print catalogs. With this scale, it is possible to evaluate catalogs in a standardized process and to get useful information on problem areas with quite low effort. Consequently, it is possible to reveal reliable and valid differences of catalogs and assess their quality. The scale offers the possibility of benchmarking different catalogs, a rapid comparison of different prototypes or also the possibility of preparing for advanced and final user tests. Furthermore, the evaluation by a large number of users without too much effort is also conceivable. One possibility of implementing this would be the online presentation of the catalog pages with a corresponding link to the questionnaire. After consideration of the pages, the corresponding participants data would be allocated directly to a data bank for evaluation. Further experiences with the scale would be preferable and necessary for the future. Its possible to transfer the instrument to other catalog forms. However, a modification or adaptation of different items might be necessary depending on the aim of the catalog. The use of the scale in the evaluation of other communication instruments, such as personalized letters or instructions for use, would be equally as conceivable. Limits to the interpretability of the usability problems revealed are also set through the given dimensions of the scale. Concrete insights into the behavior and experience in the actual interaction with the print catalog cannot be obtained in this manner. In each case, it is a question of global judgment on the overall use process. If concrete insights are required, the use of verbal protocols can provide a more detailed view. As we have seen, the method provides the opportunity to record the behavior and experience accompanying the process and gain further insights. However, the effort which the verbal protocols require should not be underestimated. A use in a small part of the overall sample seems sensible and should provide sufficient information. A combination of the two methods consequently appears useful. Therefore, the scale developed represents a useful addition to qualitative methods. In this way, it should be possible to design print catalogs from different industries in an even more user-friendly manner in future and consequently attract customers.

References
Bttner, O. B., Schulz, S. and Silberer, G. (2006) Perceived Risk and Deliberation in Retailer Choice: Consumer Behavior towards Online Pharmacies, Advances in Consumer Research, Vol. 33, ed. by C. Pechmann and L. Price, Duluth, MN: Association for Consumer Research, pp197-202. Bttner, O. B. and Silberer, G. (2007) Video-cued Thought Protocols A Method for Tracing Cognitive Processes at the Point of Purchase, in: Lowrey, T. M. (Ed.), Brick & Mortar Shopping in the 21st Century, Lawrence Erlbaum Associates, Mahwah, NJ Bhv (2005) Pressemitteilungen des Bundesverbandes des Deutschen Versandhandels e.V. date: 22.11.2005 & 9.12.2005; [online], http://www.versandhandel.org/Presse.89.0.html ISO (1998) ISO 9241-11 Ergonomic requirements for office work with visual display terminals (VDTs) -- Part 11: Guidance on usability, [online], http://www.iso.org/iso/en/CatalogueDetailPage.CatalogueDetail?CSNUMBER=16883&ICS1=13&ICS2=180&ICS3= Ericsson, K. A. and Simon, H. A. (1993) Protocol Analysis Verbal Reports as Data, Cambridge, The MIT Press

www.ejbrm.com

102

Academic Conferences Ltd

Sebastian Schulz, Gunnar Mau, and Gnter Silberer

Fodness, D. and Murray, B. (1999) A model of tourist information search behaviour, Journal of Travel Research, Vol. 37, No. 1, pp220-230 Goossens, C. (1994) External information search: Effects of tour brochures with experiential information, Journal of Travel and Tourism Marketing, Vol. 3, No. 3, pp89-107 Hodgson, P. (1993) Tour operator brochure design research revisited, Journal of Travel Research, Vol. 32, No. 1, pp50-52 Hodgson, P. (1991) Qualitive research on tour brochure design, Journal of Market Research Society, Vol. 33, No. 1, pp51-55 Kirakowski, J. (1996) The software usability measurement inventory, in: Jordan, P. W., Thomas, B., Weerdmeester, B. A. & McCelland, I. L. (Ed.). Usability evaluation in Industry, Taylor & Francis, London, pp169-177 Kohavi, R and Round, M. (2004) Emetrics Summit 2004: Front Line Internet Analyticsat Amazon.com, [online], http://ai.stanford.edu/~ronnyk/emetricsAmazon.pdf Konradt, U., Wandke, H., Balazs, B. and Christophersen, T. (2003) Usability in online shops: scale construction, validation and influence on the buyers intention and decision, Behaviour & Information Technology, Vol. 22, No. 3, pp61-73 Krmker, H. (1999) Die Welt der Benutzerfreundlichkeit, in: Henning, Jrg & Tjarks-Sobhani, Marita (Ed.). Verstndlichkeit und Nutzungsfreundlichkeit von technischer Dokumentation, Lbeck, Schmidt-Rmhild, pp22-33 Lee, B.-K. and Lee, W.N. (2004) The effect of information overloadon Consumer Choice Quality in an On-Line Environment, Psychology & Marketing, Vol. 21, No. 3, pp159-183 Mack, R. L. and Nielsen J. (1994) Executive Summary, in: Nielsen, J. & Mack, R. L. (Ed.). Usability Inspection Methods, Wiley, New York, pp1-23 Nielsen, J. (1994) Guerilla HCI: Using Discount Usability Engineering to Penetrate the Intimidation Barrier, [online], http://www.useit.com/papers/guerrilla_hci.html Nielsen, J. (1993) Usability Engineering, Academic Press, Bosten, San Diego, New York etc. Reynolds, F. D. (1974) An Analysis of Catalog Behavior, Journal of Marketing, Vol. 38, No. 3, pp47-51 Schmeier, D. R., Behrendt, C. and Singer, B. (2005) Werbewirkung messen und optimieren Wegweiser und Methoden am Beispiel von Print-Anzeigen, Planung & Analyse, No. 6/2005, pp49-56 Schweibenz, W. and Thissen, F. (2003) Qualitt im Web Benutzerfreundliche Webseiten durch Usability Evaluation, Berlin, Heidelberg, New York, Springer Verlag Silberer, G. (2006) A New Approach to Assessing Cognitive Processes at the Point of Sale Video-Cued Thought Protocols, European Advances in Consumer Research, Vol. 7, ed. By K. M. Ekstrm and H. Brembeck, pp555-556 Stoessel, S. (2002) Methoden des Testings im Usability Engineering, in: Beier, Markus (Ed.). Usability: Nutzerfreundliches Web-Design, Springer, Berlin, pp75-96 Sweeney, M., Maguire, M. and Shackel, B. (1993) Evaluating User-Computer interaction: A Framework, International Journal of Market Man-Machine, Vol. 38, No. 4, pp689-711 Yom, M. and Wilhelm, T. H. (2004) Methoden und Erkenntnisse der Web-Usability-Forschung, in: Zerfa, A. & Zimmermann, H. (Ed.). Usability von Internet-Angeboten. Grundlagen und Fallstudien, Baden-Wrtenberg, Medien und Filmgesellschaft

www.ejbrm.com

103

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (93-104)

www.ejbrm.com

104

Academic Conferences Ltd

The use of Grounded Theory and of Arenas/Social Worlds Theory in Discourse Studies: A Case Study on the Discursive Adaptation of Information Systems
Ana C. Vasconcelos University of Sheffield, United Kingdom a.c.vasconcelos@sheffield.ac.uk
Abstract: This paper exemplifies the combined use of Grounded Theory and of the Arenas/Social Worlds Theory in a study of the discursive interaction amongst middle managers at a UK University administration and academic computing services. This study aimed at exploring the role of discursive interaction and negotiation in the organisational adaptation of information systems, by defining the premises upon which discourses were constructed and deployed on the basis of particular worldviews and how in turn they informed back different worldviews. It presents key lessons learned from this approach in relationship to the roles of codification, of relationships bewtween conceptual categories and between between theoretical influences and empirical work, as well as those emerging from the lived experience of research analysts. Keywords: grounded theory, arenas/social worlds theory, discourse analysis, case study, information systems adaptation

1. Introduction
This paper presents an exemplification of the combined use of Grounded Theory and of the Arenas/Social Worlds Theory in discourse studies, based on a case study exploring the role of professional discourses in the organisational adaptation of information systems. The idea of using Grounded Theory or the Arenas/Social Worlds Theory in discourse studies is not new and Clarke (2005) presents recent examples of this The methodological approach of Grounded Theory and the conceptual framework provided by the Arenas/Social Worlds theory appear, in effect, particularly congruent with each other, as suggested by Clarke (2005). Both originate within the Chicago Symbolic Interactionism School of sociology and have Barney Glaser and Anselm Strauss, amongst others, as early proponents. Grounded Theory was the methodological approach that was developed and proposed as a result of the studies in medical institutions that were amongst those that led to the development of the Negotiated Arenas/Social Worlds Theory, as a framework for explaining social interaction in complex large professional organisations, such as hospital environments, where groups of clinicians espousing different ideologies and deploying different professional rhetorics coexist. The combination of both seems therefore appropriate and of usefulness in discourse based studies that take a focus on the constitutive (rather than just merely representational) role of discourse and on negotiated interaction as the vehicle for the constitution and reproduction of discursive behaviour. In this paper it is argued that this approach is particularly powerful for: 1. analysing different perspectives on issues through the discursive practices of involved organisational actors, 2. defining the premises around which these discourses are constructed and deployed and, simultaneously, 3. exemplifying how, in turn, discourses inform back different worldviews in dynamic social contexts. However, whereas both Grounded Theory and the Arenas/Social Worlds Theory have been used extensively in various fields, they have both been subject to various re-interpretations and re-elaborations, as will be discussed ahead. This paper presents an example of an interpretation of both methodological approach and conceptual framework that is perhaps closer to their original formulation than further reformulations. An exploratory case study using this framework was carried out at a University in the United Kingdom during the implementation and post-implementation phases of a University-wide Management Information System. This study set out to explore what was perceived as a relatively neglected theme in the information systems research literature the role of discourse in the organisational adaptation of information systems within a constructivist perspective.

2. The conceptual framework: some (loose) connections between the negotiated arena/social worlds and discourse studies
The Arenas/Social Worlds framework implies exploring the social world as different arenas, where multiple worldviews coexist within processes of negotiated interaction (Strauss et al., 1964, 1981). Clarke (1991: 128)
ISSN 1477-7029 125 Academic Conferences Ltd Reference this paper as: Vasconcelos A. C. (2007) The use of Grounded Theory and of Arenas/Social Worlds Theory in Discourse Studies: A Case Study on the Discursive Adaptation of Information Systems. The Electronic Journal of Business Research Methods Volume 5 Issue 2, pp 125 - 136, available online at www.ejbrm.com

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (125 136)

defines an arena as [] a field of action and interaction among a potentially wide variety of collective entities. Social worlds were explicitly referred to only in Strausss later work (1978), although Clarke (1991) argues that they were implicit in his original study of psychiatric institutions (Strauss et al., 1964, 1981), and are defined as universes of discourse (Mead, 1938:518, referenced in Clarke, 1991: 130) and as [] groups with shared commitments to certain activities, sharing resources of many kinds to achieve their goals, and building shared ideologies about how to go about building their business (Clarke, 1991: 131). Social worlds are composed by individual actors who also bring personal interests and commitments to action, although, as Clarke (1991) notes, in the arena framework, they are essentially seen as representatives of their social groups. A development of the original arenas framework that is particularly relevant to this study is the establishment by Clarke (2005), after Castellani (1999), of an explicit focus on issues of power and of a link between the work of the Chicago Symbolic Interactionism school and that of Foucauldian studies on discourse and power. Clarke (2005:52), in effect, states that If action is at the heart of Strausss project and power at the heart of Foucaults work, they meet in related conceptualizations of practices as fundamental processes of action and change []. She stresses that concepts of practice, which include on Foucaults part, discourse/discipline and regimes of practice, and on Strausss part, social worlds/arenas and negotiated orders, are however not equivalent, but, rather, related. In effect, while both authors emphasized discourse as being constituted through interaction (Clarke, 2005), Strauss seeing social worlds as universes of discourse, Foucault placed discourse far more explicitly in frames of power and characterised discourses as [] effected by disciplining practices that produce subjects/subjectivities through surveillance, examination, and various technologies of the self [] often through the influence or the imposition of social groups (Clarke, 2005: 54). A further influence in the perspective of discourse that is herein adopted is provided by the work of Bakhtin (1984, 1986). Bakhtin was an early proponent of the view of language as rooted in social interaction, rather than being an abstract and politically neutral system of signs. In contrast, language is often framed, in his view, within social struggles. This was represented, in his work, through the tension, expressed in the form of a conflict, between centripetal forces, focused upon the production of standardised and codified meanings expressed in dogmas and accepted views of universal truth, and centrifugal forces that promote diversity and variation informed in different and alternative discursive genres. The tension between centrifugalism and centripetalism, between standardisation and diversity, is especially significant to the discussion of the findings of the case study explored in this paper, as will be discussed ahead. It has interesting parallelisms with the proposition by Ellis (1986:116) that the widespread use of IT leading to the proliferation of computer based information systems in organisations has led to the concurrent development of two opposite effects in organisations: "the centrifugal effect of the rapid, but often uncoordinated growth in the use of" computer based information systems and [] centripetal efforts to coordinate and control the information handling function [] It was thought that these theoretical influences form a particularly interesting vehicle to address how the complexity of the information environment was represented and explored at the case study organisation, how this in turn affected the relationship between organisational tensions and the management of information and, finally, how the discursive practices of organisational actors were formed around these perspectives and how these perspectives were, in turn, informed by those discourses.

3. The methodological approach and design: an interpretation of the grounded theory approach
This study focuses on discourse and on how the interplay between discourses plays a critical role in the organisational adaptation of information systems. The emphasis taken in this study is that discourse plays a constructive and constitutive role, rather than just merely a representational role. Therefore, a naturalistic inquiry approach seemed more appropriate to the research aims, where minimising the manipulation of situations was sought, in order to study comprehensively the complexity of issues involved and to avoid prior constraints to the outcomes of the study. The emergent trajectory of naturalistic inquiry is often, as proposed by Lincoln and Guba (1985: 203), [] played by ear; it must unfold, cascade, roll, emerge. The design of the research adopted a case study approach and was largely based upon some of the general principles of Grounded Theory (Glaser and Strauss, 1967). It should be stressed that it was guided by generic principles of Grounded Theory, rather than following a step by step adoption a particular version of the Grounded Theory methodology. In effect, although, as referred by Brown (1990:9), grounded theory can

www.ejbrm.com

126

Academic Conferences Ltd

Ana C. Vasconcelos

be shaped and used as 'a fully fledged methodology replete with a method of analysis', it is above all a 'particular style' of research, founded in generic principles, such as theoretical sampling and the use of constant comparisons as a basis for analysis. It can also be said, as noted by Bryman (2001:391) that, although in effect Grounded Theory is by far one of the most cited methodological approach to qualitative research, it [] may have been honoured more in breach than in observance and has been deployed in many different ways. In its original formulation (Glaser and Strauss, 1967), it was intended as a flexible approach to generate theory from data, based upon the three generic principles of theoretical saturation, constant comparison method of analysis and theoretical saturation. The subsequent reformulation of Grounded Theory by Strauss and Corbin (1992) placed strong emphasis on its proceduralisation and formalisation into a series of techniques, arguing the need to [] spell out the procedures and techniques [] in greatest detail [] in step-by-step fashion (Strauss and Corbin, 1990:8). Their approach placed greater emphasis in the codification of data. Core to this version of Grounded Theory is what Strauss and Corbin (1990) refer to as the paradigm model, defined as a systematized cause and effect schema to explain the inter-relationships between broader categories and sub-categories. The publication of this version of Grounded Theory by Strauss and Corbin led to a divergence of perspectives regarding the nature of the methodology between its two original authors, as Glaser (1992) considered that the new version was too prescriptive and emphasized too much the role of coding, whereas the original essence and intention of Grounded Theory was to focus on theory development: Anselms methodology is one full of conceptual description and mine is grounded theory. They are very different, the first focusing on forcing and the second on emergence. The first keeping all of the problems of forcing data, the second giving them up in favour of emergence, discovery, and inductive theory generation (Glaser, 1992:122). The approach adopted in this study is more aligned with its original formulation and is influenced by the three key principles that guided the original formulation of Grounded Theory of theoretical sampling, inductive derivation of the key findings through constant comparison of interview data and an attention towards theoretical saturation to decide the delimitation of the key categories of findings and the inter-relationships between them, which forms the basis for the framework emerging from this study. It differs from many applications of Grounded Theory (for example, Kirk and Vasconcelos, 2003; Lnsisalmi, Peir and Kivimki, 2004; Clarke, 2005), and certainly from Strauss and Corbins version, in that, although interviews were initially marked with open concepts and constant comparison led to define the characteristics of key categories of findings, the approach to data analysis was rather more loose than those approaches that place emphasis on the formal and detailed process of data codification and did not apply the cause and effect schema included in Strauss and Corbins paradigm model to formally delimit the properties of categories, sub-categories and their inter-relationships. Glaser and Strauss (1967:31), in their original formulation of Grounded Theory state, themselves, that theory generated through its principles can be presented [] either as a well-codified set of propositions or in a running theoretical discussion.

4. The research in action: empirical environment, data collection and analysis

The diagram in Figure 1 below attempts to represent the research process that was undertaken in action. This was experienced and is seen as rather circular in the inter-relationship between different phases of research and in the use of the literature to help illuminate and discuss the implications of the findings. The circles that represent these processes are not concentric and should be seen rather more as rotating ellipses that form contacts with each other at different points in the research in action. For example, constantly revisiting the data led to reformulations of both key categories of findings and of how the more specific and detailed concepts fitted with them, as well as to reconceptualising the relationships between findings. This often led to post hoc explorations of different strands of the literature prompted by new findings and constructs. At points, this process resembled going around in circles although the important aspect of this was that, each time a circle was completed, the research arrived at a different point and advanced towards what seemed a clearer outcome in terms of the interpretation of the inter-relationships between different elements of the argument. The methodological approach that was adopted throughout the study was inherently emergent, as was the understanding of the research problem that was derived through this process, but this allowed the need for the flexibility in planning and enabled the shifts in the understanding of situations necessary to accommodate nuanced accounts inherent to the centrifugal messiness (Sullivan and McCarthy, 2004) of complex professional arenas. There were several areas, specifically in the process of data analysis, that are particular to the process undertaken in this research which will be detailed in this section.

www.ejbrm.com

127

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (125 136)

Inductive

Conceptual framework Relating categories : Constructs

Axial-coding, constant comparison

Data analysis

Categories

Open-coding Concepts

Empirical observation Research question Theoretical sampling Data collection Deductive

Figure 1: A diagrammatic representation of the use of Grounded Theory, superimposed to Rudestam and Newtons (1992) research wheel. The case study that was carried out took place in a single organisational setting and was centred on the implementation of a corporate wide management information systems at a University, part of a national initiative called the MAC (Management and Administrative Computing) Initiative, funded by the UK University Grants Committee, which had the aim of introducing common administrative software to allow data comparability across the sector. This institution was chosen because it represented what seemed to be a very interesting opportunity, as at the time interviews started it was implementing the MAC system and, simultaneously, changing its administrative and technical structures, leading to the centralisation of the control over its corporate information processing activities under a new department, the Corporate Information Department. As a large professional organisation, the University also offered a rich ground for exploring the negotiated arena concept where negotiated interaction further reproduces social structures and forms loci for learning (McAuley, 1994; Cohen, Duberley and McAuley, 1999; McAuley, Duberley and Cohen, 1999; Darwin, Johnson and McAuley, 2002). The principal vehicle for data collection was qualitative interviewing, supported by the analysis of some internal documentation, as well as official reports on the higher education sector in the United Kingdom. The interviews were carried with a group of twelve middle managers at the University Administration and at Academic and Administrative Computing Services in the University chosen for the case study. Dick (2004: 207) shares with King (2004:207) the understanding that in the constructivist style of interviewing, particularly in discourse studies, tends to typically involve a small number of interviewees: This is because the focus is on the text, not the individual and because the aim is to provide an in-depth analysis that is focused on explanation, rather than generalization. McCracken (1988: 17) goes further, when referring to sampling in in-depth interpretive approaches, The first principle is that less is more. It is more important to work longer, and with greater care, with a few people than specifically with many of them. For many research projects, eight respondents will be perfectly sufficient. The quantitatively trained social scientist reels at the thought of so small a sample, but it is important to remember that this group is not chosen to represent some part of the larger world. It offers, instead, an opportunity to glimpse the complicated character, organisation and logic of culture. This stance influenced the process of data collection in both the style of interviewing and in the selection of the interviewees. The individuals that participated in the study had in common the fact that they were middle managers in the administrative and technical services of the University, located in different departments, who were either directly involved in the implementation of the MAC system or for whom the MAC system had directly impacted on the way they carried out their work. The choice of middle managers was deliberate, as it was considered that it could lead to particular insights, due to the mediating role between the core and the periphery of organisations that middle managers often carry out (Clegg, 2003; Clegg and McAuley, 2005). The group of interviewees included, for example: a senior manager at the Registry in charge of overseeing the development of the information strategy process; the deputy director of the department in charge of implementing the new system, its project manager and two other managers responsible for particular aspects of the systems; the systems manager at the Finance Department and an IT manager at Academic Computing Services; the administrators in charge of systems at the undergraduate and postgraduate Student Offices and heads of administration and of IT services at two different academic departments (one a

www.ejbrm.com

128

Academic Conferences Ltd

Ana C. Vasconcelos

heavy user of IT, the other less so). These managers participated in different areas of the administrative and technical arenas at the University, both at the centre and at the periphery of the decision-making process and represented both consenting and dissenting voices towards the changes related to the new set of systems, which intersected and clashed with each other. The various interviews were relatively unstructured in their style, in that they did not follow particularly prescribed formulations of questions or even a formally defined guide, although there were particular issues that were set as important to explore, albeit in a non-prescriptive manner allowing, thus, the exploration of unanticipated avenues. The following themes were set as particularly relevant to explore throughout the interviews, without, as mentioned, precluding other potentially important areas: The first issue concerned the expression of the roles and of perceptions of these roles of the different interviewees; A second area for discussion related to perceived difficulties expressed by the interviewees in undertaking the activities required by their roles; Views regarding the restructure undertaken at the University were explored in particular depth; Another theme included the use of information within their professional activities and the role of the newly implemented information systems within that, as well as issues of control and ownership over different areas of information management; Finally, activities and options regarding the planning, implementation and use of these systems were also discussed. As this study was focused on discursive practices and based upon the text of the interviews, tape recording and subsequent full transcription of interviews was carried out in order to ensure the complete capture of this data. The first step of analysis involved a familiarisation with the data. For that, it was important to review interview transcripts as soon as possible after the interview and to note concepts and themes that became immediately apparent and cross-referencing them to their occurrence in the interview transcripts. Then the interview transcripts were marked and annotated with these concepts. At this stage, referred to as open coding in the constant comparative method, these were rather loose and spontaneously generated concepts they were a first approach to identifying topics without being driven by too much concern with formalization. These different concepts seemed, at first, disparate, with loose connections and, in some cases, some potential overlaps. A large number of open concepts were initially defined and examples include, for instance, resistance and buy-in [to the systems], accuracy of local data, sensitive data, access, user participation The next stage involved the comparison each of these concepts with the data that they relate to. This often led to the observation of certain patterns in the data, defining broader key categories of findings that presented internal coherence, in this case related to the identification of tensions between discursive practices. This process involved: 1. firstly the analysis and discussion of the nature of the tensions between discursive practices, which in the case of this study were articulated around information centrifugalism and information centripetalism, as competing models of the information environment, and between a focus on standardisation of process and a focus on negotiated meanings, as models for the management of the information environment; 2. further than that, explaining how these two different tensions can be in turn inter-related and how they relate to broader discourses and perceptions of learning and sense-making and uncertainty and complexity; 3. and finally, how these inter-relations can be understood within the interpretative framework of an information arena and the particular aspects of interaction and negotiation within that characterises it, as a means to explain the role of discourse in the organisational adaptation of information systems. The process of constantly revisiting the data led to reformulations of both the characteristics of key categories of findings and of how the more specific and detailed concepts fitted with them, as well as to reconceptualising the relationships between findings, which ultimately led to the framework that emerged from the study and is presented in the following section.

www.ejbrm.com

129

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (125 136)

5. The findings: discursive tensions and connections at the University information arena
The principal finding in this study is that the discursive interaction amongst social actors plays a fundamental role in the organisational adaptation of information systems through the exploration of tensions between different discursive resources and interpretive repertoires. As stated by Brown (1998: 52), [] the capacity of IT to coordinate, structure and control is contingent only, and [] actors often have much considerable discretion over their use of technology that making reasonable a priori predictions regarding the consequences of a particular implementation is often impossible. Discretion comes from the ability to explore tensions between discursive resources and negotiate meaning through the constitutive role of discourse. The following paragraphs discuss the findings of the study against each of its aims. The first aim of the study concerned the analysis and exploration of perspectives on the relationship between the management of information systems and of the wider information environments they belong to, through the discursive practices of organisational actors. The analysis of the empirical material revealed that the discursive practices of the various interviewed professionals were focused upon the context of their work practices and activities, rather than projecting information and their information behaviour as a beginning or an end in itself. The information environment in which they operated, which included a variety of sources, systems, processes, behaviours, but also attitudes and cultures, was largely referred to, often implicitly, as an enactment and a representation of their social environment, structures and interaction, in its formal and informal manifestations. We can then extend the notion of negotiated arena to consider the University administration information arena as a field of interaction between different worldviews on information environments that coexist in the same social setting. The University administration information arena was both represented and constituted by particular discursive practices that made reference to different interpretative repertoires and resources. The second aim of the study was referred to as defining the premises around which these discourses are constructed and deployed. As mentioned previously, these discourses were articulated around three major categories of interpretative repertoires and discursive resources: models of the information environment, expressed through the tension between information centripetalism and information centrifugalism; models of information management approaches, expressed through the tension between a focus on the control over processes and a focus on sense-making and negotiation of different meanings; and, underlying the previous elements, assumptions about the nature and complexity of the environment, strategies for dealing with uncertainty and correlated models of learning, expressed through exploitation as a complexity reduction strategy and exploration as a complexity absorbing strategy. The discursive tensions were identified through the analysis of critical incidents in the data. There were many examples of these incidents and the following paragraphs will exemplify how the analysis of these incidents led to the identification of the tensions. This will be illustrated by two examples of these critical incidents centred around the internal debates that occurred around the notion of devolvement and introduction of a new corporate data model as a consequence of the implementation of the new management information systems. Devolvement was a concept frequently articulated by various of the interviewees. It was seen as key to define the roles of the various actors and to define what constituted legitimate intervention in the implementation and use of the system. However, although most interviewees used this term they different meanings to it. Actors at the Centre of the decision-making process articulated these roles through the distinction between normative and functional responsibility: [] theres different levels of responsibility which I find get in the way, like people who are responsible for the functional, actually just doing , and theres normative responsibility who say this is how it ought to be. Now we confuse both of them in this place so we have some people who believe they are responsible for things who are doing it just at the lower functional level. (CI.4:25)

www.ejbrm.com

130

Academic Conferences Ltd

Ana C. Vasconcelos

In this discursive sphere the distinction between normative responsibility (as defining rules) and functional responsibility (as working within the rules and being accountable for complying with them) led to distancing from the decision making process actors at the periphery of the systems implementation process in this case, administrators at central support services and at academic departments and faculties. Devolvement was seen as a means to define accountability. I dont necessarily think that people are going to be losing responsibility because they are still responsible within their department [] responsibility is not going to be taken away because it will only be done with the approval of the central department because when it comes back to it they are responsible for the data and responsible to the Registrar - making sure that the data on the system is accurate, so I dont think there is a sense that they are going to lose that responsibility (CI.1:30) On the other hand, administrators at central support services and at academic departments and faculties interpreted devolvement as a means to ensure the preservation of local autonomy in terms of work practices and requirements. It is our responsibility as end-users of the system to find the changes/improvements we want on this Central University database and that the system operates efficiently. It is the Department of Corporate Informations responsibility to put that into practice, in the sense of technical amendments to the software and the programmes and then come back to us and say weve done this is it better for you? and then we enter into negotiations like well, thats very good but can you just make one final change and that will be fine for us []. (SO.1:25) In this discursive sphere, the periphery reclaims normative responsibility through the reinterpretation of meaning in relationship to what constitutes legitimate action in systems interventions. The debate around devolvement provides an example of a critical incident underlying the tension between models of the information environment that emphasized information centripetalism, manifested itself through a trend towards the concentration of control and coordination of information handling activities at the Centre, and those that were positioned around information centrifugalism and emphasized the diversity of local processes and practices, and correlated knowledges. Other incidents underlying discursive tensions, in this case between approaches to information management focused on the control of processes and those focused on the negotiation of meanings, were found in the debate concerning the introduction of a new corporate data model. The new corporate data model was underpinned by new funding models and formulae and had strong implications for the allocation of resources across the University. It required the standardisation of processes and procedures across the University administration. Both the Corporate Information Department and the Finance Department at the University concentrated on defining approaches to information management focused on the standardisation of processes and procedures. Approaches to information management focused on process fostered an assumption that the definition of processes and procedures was, in effect, a means to ensure adequate meaning and information accuracy. For the administrators that adopted this approach, a focus on standard processes and procedures, rather than on the variety and multiplicity of local information, allowed the establishment of a homogeneous and consistent way of making sense of the complexity of the world of the University. From my point of view as an administrator the focus is on the process because its not my responsibility to achieve targets, its my responsibility to show that we may or may not achieve targets and to show that you need processes that will give you the information and enable you to present it in a particular way. (DIS.2:13) On the other hand, at central support services, such as the Postgraduate and Undergraduate Student Offices and at academic departments, administrators argued that establishing the accuracy of what was being presented through the new information systems was vital. They asserted their positions by establishing their focus of action on assuring that the meaning of the content of the information systems was accurate. The simplification introduced by the funnel effect of centripetalism and standardisation (Boisot, 1998) was presented as detrimental to both information richness and accuracy.

www.ejbrm.com

131

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (125 136)

I think the core information is important, though. I dont think it is just a process issue, because we are dealing with individual student places, I would say. Obviously, the purpose of having procedures is to try ensure as much as possible that the information is correct []. (SO.1:28-29) Administrators at the periphery responded therefore to centripetal attempts to control processes and standardise meanings by asserting their position and emphasizing their role as key to ensure accurate meanings. In the area of student administration, which had an important impact in the allocation of funding to Universities, this was done by developing rhetorics around notions of accuracy and of its importance, establishing themselves key holders to information accuracy, in a context where accuracy was in many instances established through negotiation of multiple meanings (in different academic committees, for example). These different categories of interpretative repertoires and discursive practices that inform information arenas and the tensions that characterise them arise from the characteristics of the interaction between Centre and Periphery, where the former is driven by colonising intents (Prichard and Willmott, 1997), requiring the simplification of the complexity of the environment and the development of knowledges and discourses of controlling organisations (hence the focus on process, standardisation and singularity of meaning), and the latter is driven by a concern to maintain a local modus operandi and identity, exploring the complexity of the environment through various niches of professional expertise (hence the focus on local knowledges, and practices and on the negotiation of multiple meanings). It would be simplistic to characterise the various tensions both, on one hand, in terms of the antagonism between two opposing and exclusive poles, and, on another, as part of some inherent invisible rational logic that would inevitably draw the interaction between forces into some state of desirable equilibrium. The response to the third aim of this study, to explain how, in turn, discursive practices inform worldviews on the information environment - addresses this issue. Although there were clear inter-linkages and alignments between these tensions, there were not always necessarily automatic correspondences between them and different individuals carried out activities of organisational translation, by framing the interests of some communities in the terms of the discourse adopted by other communities (Cohen, Duberley and McAuley, 1999), as well as, when necessary, making appeal to different rhetorical strategies and orders of discourses that sometimes implied contradictory assumptions. This suggests that the relationship between different forces in information arenas can also be characterised by elements of connection, as well as by tension. In this context, the interplay between different discursive regimes in negotiated arenas allows the reconstruction of their social environment, as well as representing it and enacting it. Much of the original literature on managerialism in Higher Education has over emphasized the imperialising effect of managerial discourses over academic work (Willmott, 1995; Parker and Jary, 1995). This case study illustrates that in the Higher Education administrative information arena, the managerial discourse, clustered around information centripetalism, an emphasis on process standardisation and generic knowledges of managing (in the sense of planning and controlling) organisations, could be successfully nuanced by discourses that emphasized professionalism, centred around information centrifugalism, plurality of meanings and local practices and knowledges. The interplay between these different discourses is much more marked by mediation and negotiation than by antagonism, as is stressed in more recent work by Prichard and Willmott (1997), Trowler (1998, 2001) and Clegg and McAuley (2005). In this context, we can also conceptualise information arenas as force fields of negotiated interaction, marked by discursive tensions and contacts. The University administration information arena can then be seen as a force field where these tensions and connections between universes of discourse are both informed by and informative of action and interaction (in the Straussian sense) or practice (in the Foucauldian sense), whether it is conflicting, collaborative, disciplining, negotiating. In doing so, these different discursive regimes served, in the particular case that was explored in this study, as a vehicle for the organisational adaptation of information systems from an initial centripetal agenda to a more distributed and negotiated model.

6. Concluding remarks: lessons from the research approach

This paper argues that the combined used of Grounded Theory, as a methodological approach, and of the Negotiated Arenas/Social Worlds Theory, as a conceptual framework, is of usefulness in discourse based studies that take a focus on the constitutive (rather than just merely representational) role of discourse and on negotiated interaction as the vehicle for the constitution and reproduction of discursive behaviour. Its major strength lies in the flexibility and openness that enable to capture shifts in the understanding of dynamic situations in complex organisational arenas, as mentioned above. This is particularly the case

www.ejbrm.com

132

Academic Conferences Ltd

Ana C. Vasconcelos

where the exploration of tensions between different discourses leads to forming new understandings and shapes the interaction between different social actors. Both the methodological approach and the conceptual framework, originally developed in the field of the sociology of healthcare and of biomedicine, are sufficiently flexible, abstract and adaptable to be applicable to a wide range of studies and across different disciplines to explore evolving and complex social environments and can be geared especially to understand how social actors negotiate meaning in contexts marked by the coexistence of different social worlds. They are also particularly useful in helping distinguish nuances in the understanding of these contexts. Several learning points were derived from this study and are inter-related. The first concerns the dangers of over emphasizing the codification of data, and thus of typifying behaviours, which is at the heart of the schism between Glaser and Strauss (Glaser, 1978), through the tension between conceptual description and emergence of theory and explanation. As stated by Goulding (1998: 55), [i]t is important to recognise that most individuals engage in a type of behaviour without being typed by it; they engage in other behaviours as well and Grounded Theory categories are concerned with patterns in social practices, including discursive practices, rather than in types of persons and whether they are, as individuals, representative of particular categories of actors. It is the recognition of this stance that allows the identification of nuances in complex arenas and their social worlds. It is argued that the adoption of Negotiated Arenas/Social Worlds conceptual elements should be used as an interpretive framework to bring out complexity and fluidity in social worlds, rather than as a vehicle to further proceduralise and formalise the process of analysis through the development of rigid codification schema. A correlated aspect is that relationships between key categories of findings are as important, if not more, as the delimitation of analytical categories. In this study, it became clear that the focus should be to consider the multiple dimensions of its context, where different forces and tensions, articulated through the discursive practices and rhetorical strategies of the various interviewees, appeared to interplay and to in turn weave a more complex situation. The crucial point at this stage was a departure from considering in isolation the different discursive categories that were identified to move on to a different stage of analysis, where mutual relationships between these categories were reflected upon. This departure from analytical categories as ideal types considered in isolation to a focus on interaction, tension and contacts, allowed the focus on the interplay between different discursive regimes and practices to reflect the dynamic nature of interaction in professional arenas. It is in this area that the contribution of the Arenas/Social Worlds Theory as a theoretical framework was crucial to help reflecting and characterising complex, multidimensional and shifting environments and arenas. This leads to acknowledging the key role of sensitising influences to the meaning and significance of emerging analytical categories and their inter-relations: This is vital, for without this grounding in extant knowledge, pattern recognition would be limited to the obvious and the superficial, depriving the analyst of the conceptual leverage from which to develop theory (Goulding, 1998:52). Glaser and Strauss, themselves, state that Grounded Theory [] will tend to combine mostly concepts and hypothesis which have emerged from the data with some existing ones that are clearly useful. What it does not do is to commit research to [] one specific preconceived theory. As stated by Dey (1993:65) [] there is a difference between an open mind and an empty head. To analyse data, we need to use accumulated knowledge, not dispense with it. In other words, analysis is emergent, but not atheoretical (Goulding, 1998). As stated by Glaser (1978) and Goulding (1998), the development of Grounded Theory is based upon an understanding of the interplay between theoretical influences and empirical work. Other important influences, however, stem from the lived experience (Sullivan and McCarthy, 2004) that analysts bring to their work. While conducting this research the researcher also developed a lecturing career in information management and information systems in various institutions in the Higher Education sector. This brought a particular slant to the understanding of the focus of this study and, conversely, the study itself also brought new insights over work practices and changes in the sector. McAuley (2004) refers to the interweaving of the personal experience of researchers, which forms a pre-understanding in hermeneutic studies, with the ways in which the subjects of the research develop and present understandings of the situation under study: Lying at the heart of the hermeneutic approach is this notion of openness to the data, the artful development of the interplay between the intuition of the researcher, the data (text or whatever) of the subjects of study, the interpretative frameworks that are brought to bear on the analysis of the text and, ultimately, the reader. If this openness is undertaken in good faith then the product of the research is on the one hand truthful (authentic) to the data but is, on the other hand, not the only truth (authentic account) that could be produced (McAuley: 2004:201).

www.ejbrm.com

133

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (125 136)

References
Bakhtin, M. (1984). Problems of Dostoievskys Poetics. Emerson, C. (ed. and transl.), Manchester: Manchester University Press. Bakhtin, M. (1986). Speech genres and other late essays. Emerson, C. and Holquist, M. (eds.). Austin: University of Texas Press. Boisot, M. (1998). Knowledge assets. Oxford: Oxford University Press. Brown, A.D. (1990). Information, communication and organisational culture: a grounded theory approach. PhD. thesis, the University of Sheffield. Brown, A.D. (1998). Narrative, politics and legitimacy in an IT implementation. Journal of Management Studies, 35 (1), pp. 35-58. Bryman, A. (2001). Social research methods. Oxford: Oxford University Press. Castellani, B (1999). Michel Foucault and Symbolic Interactionism: the making of a new theory of interaction. Studies in Symbolic Interaction, 22, pp.247-272. Clarke, A. (1991). Social worlds/arenas theory as organizational theory. In Maines, D. (ed.) Social organization and socials process: essays in honor of Anselm Strauss. Hawthorne (NY): Aldine de Gruyter, pp.119-158. Clarke, A. (2005). Situational analysis: Grounded Theory after the Postmodern turn. Thousand Oaks (CA): Sage. Clegg, S. (2003). Learning and teaching policies: contradictions and mediations of practice. British Educational Research Journal, 29, pp. 803-820. Clegg, S. and McAuley, J. (2005). Conceptualising middle management in Higher Education: a multifaceted discourse. Journal of Higher Education Policy and Management, 27(1), pp.19-34. Cohen, L., Duberley, J. and McAuley, J. (1999). Fuelling discovery or monitoring productivity: research scientists hanging perceptions of management. Organization, 6 (3), pp. 473-497. Darwin, J., Johnson, P. and McAuley, J. (2002). Developing strategies for change. London: Financial Times PrenticeHall. Dey, I. (1993). Qualitative data analysis: a user-friendly guide for social scientists. London: Routledge. Dick, P. (2004). Discourse analysis. In Cassell, C. and Symon, G., eds, Essential guide to qualitative methods in organizational research. London: Sage, pp.203-213. Ellis, D. (1986). Information management and information work. International Journal for Information Management, 6, pp. 115-116. Glaser, B. (1978). Theoretical sensitivity: advances in the methodology of Grounded Theory. Mill Valley (CA): The Sociology Press. Glaser, B. (1992). Basics of Grounded Theory analysis: emergence vs. forcing. Mill Valley (CA): The Sociology Press. Glaser, B.G.; Strauss, A.L. (1967). The discovery of grounded theory: strategies for qualitative research. London: Weidenfeld and Nicolson. Goulding, C. (1998). Grounded Theory: the missing methodology on the interpretivist agenda. Qualitative Market Research, 1(1), pp. 50-57. King, N. (2004). Using interviews in qualitative research. In Cassell, C. and Symon, G., eds, Essential guide to qualitative methods in organizational research. London: Sage, pp. 11-22. Kirk, J.; Vasconcelos, A. (2003). Management consultancies and technology consultancies in a converging market: a knowledge management perspective.[online] Electronic Journal of Knowledge Management, 1(1) March 2003, available at http://www.ejkm.com/volume-1/volume1-issue1/vol-1-issue1-papers.htm [last accessed 28.11.2005] Lnsisalmi, H., Peir,J.-M. and Kivimki, M. (2004). Grounded theory in organizational research. In Cassell, C. and Symon, G., eds, Essential guide to qualitative methods in organizational research. London: Sage, pp. 242-255. Lincoln, Y. S., and Guba, E.G. (1985). Naturalistic inquiry. London: Sage. Maybin, J. (2001). Language, struggle and voice: the Bakhtin/Volosinov writings. In Wetherell, M.; Taylor, S. and Yates, S.J. (eds). Discourse theory and practice: a reader. London: Sage, pp. 65-71. McAuley, J. (1994). Exploring issues of culture and competence. Human Relations, 47(4), pp. 417-430. McAuley, J. (2004). Hermeneutic understanding. In Cassell, C. and Symon, G., eds, Essential guide to qualitative methods in organizational research. London: Sage, pp.192-202. McAuley, J., Duberley, J. and Cohen, L. (2000). The meaning professionals give to managementand strategy. Human Relations, 53 (1), pp. 87-116. McCracken, G. (1988). The long interview. Newbury Park: Sage (Qualitative Research Methods series, v.13). Mead, G. (1938). The philosophy of act. Chicago: University of Chicago Press. Parker, M. and Jary, D. (1995). The McUniversity: organizations, management and academic subjectivity. Organization, 2(2), pp.319-338. Partington, D. (2000). Building grounded theories of management action. British Journal of Management, 1, pp. 91-102. Potter, J. and Wetherell,M. (2001). Unfolding discourse analysis. In Wetherell, M.; Taylor, S. and Yates, S.J. (eds). Discourse theory and practice: a reader. London: Sage, pp. 198-209. Prichard, C. and Wilmott, H. (1997). Just how managed is the McUniversity? Organization Studies, 18(2), pp. 287-316. Rudestam, K. E. and Newton, R. R. (1992) Surviving your dissertation.London: Sage. Strauss, A., Schatzman, L. Bucher, R., Ehrlich, D. Sabshin, M. (1964). Psychiatric ideologies and institutions. Glencoe, Il: The Free Press. Strauss, A., Schatzman, L. Bucher, R., Ehrlich, D. Sabshin, M. (1981). Psychiatric ideologies and institutions. New Brunswick, NJ: Transaction Books. Strauss, A.; Corbin, J. (1990). Basics of qualitative research: Grounded Theory procedures and techniques. Newbury Park: Sage.

www.ejbrm.com

134

Academic Conferences Ltd

Ana C. Vasconcelos

Sullivan, P. and McCarthy, J. (2004). Toward a dialogical perspective on agency. Journal for the Theory of Social Behaviour, 34(3) pp. 291-309. Trowler, P. (1998). Academics responding to change: new Higher Education frameworks and academic cultures. Buckingham: SRHE and the Open University Press. Trowler, P. (2001). Captured by the discourse? The socially constitutive power of New Higher Education Discourse in the UK. Organization, 8(2), pp. 183-201. Willmott, H. (1995). Managing the academics: commodification and control in the development of University education in the UK, Human Relations, 48 (9), pp.993-1028

www.ejbrm.com

135

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (125 136)

www.ejbrm.com

136

Academic Conferences Ltd

Wasting Time: The Mission Impossible with Respect to Technology-Oriented Security Approaches
Andreas E Wagner and Carole Brooke Lincoln Business School, University of Lincoln, UK andreaswagner75@yahoo.de cbrooke@lincoln.ac.uk
Abstract: Security is still too orientated towards technology and behavioural aspects are under prioritized. Organisations focus on the ability of technology to minimise risks. This paper as a preliminary research of a PhD thesis will argue that this assumption and approach is misguided, so we will focus on how a critical approach is more useful to exposing these issues. The key to secure systems is employees' perception and the action they take in accordance with the learned and perceived need for an understanding of compliance. The paper is about critical approaches to research and it just happens to be information and communication technology (ICT) that is the subject area. Keywords: critical research, ICT security, organizational misbehaviour, learning, compliance.

1. Introduction
The growing use of information and communication technology (ICT) is affecting modern businesses. As a result, security issues have become more apparent. It is now regarded as essential that knowledge management is supported by high security processes in order to protect and deliver organisational success. The traditional brick and mortar boundaries of organisation have disappeared and are replaced by connectivity involving internal and external users. Security is still too orientated towards technology and behavioural aspects are under prioritized. Organisations focus on the ability of technology to minimise risks. This paper will argue that this assumption and approach is misguided. A large number of security breaches result from peoples' behaviour, which cannot be solved by the use of technology itself. It is about people and specifically about organisational misbehaviour. Organisations do have to manage peoples expectations, power, and fear. People are more likely to engage themselves in security to the extent that they believe in its importance for business success. People, who trust that they will benefit while engaged in the process, both by acting according to guidelines and by promoting compliance, will support organisational security. However, people will always find a way around the system. Employees need to believe that they will benefit while acting in a secure way and if this is not the case then they will not make their contribution. Awareness programmes, education and training often seem more likely to put people to sleep than to improve their behaviour. This paper will examine a gap in management practice and the existing literature. The first author has been involved for over seven years in consulting of ICT security, thus familiar with actions and developments. His current position at an international service provider and his relationships in a large network of ICT experts is used to access the thoughts and experience of colleagues and partners in the practices of ICT security. Practice shows that the role of behaviour is not considered comprehensively enough, even though its importance has been widely recognised. Current research in this area fails to address the holistic approach, especially by lacking to focus on subjective issues. More attention needs to be paid to these aspects in order to learn why misbehaviour takes place and how it can be reduced. To tackle these weaknesses, we propose a conceptual foundation for security behaviour. The key to secure systems is employees' perception and the action they take in accordance with the learned and perceived need for an understanding of compliance. This is explored in more depth and a framework of issues proposed for addressing the problems critically. There is a need for critical research. Listening to and understanding language and power issues are themes which have to be explored further. Our initial research question will focus on the identified gap. Practice and theory do not pay enough attention to the use of technology. ICT, especially security because of the weakest link phenomenon (this will be described later in-depth), is always human dependent. But organisations are focusing on technology although most of ICT professions and experts know that the human factor is key. Mostly there is a strong
ISSN 1477-7029 117 Academic Conferences Ltd Reference this paper as: Wagner A. and Brooke C. (2007) Wasting Time: The Mission Impossible with Respect to Technology-Oriented Security Approaches. The Electronic Journal of Business Research Methods Volume 5 Issue 2, pp 117 - 124, available online at www.ejbrm.com

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (117-124)

separation in organisations to solve ICT and behavioural issues. ICT administration focuses on infrastructure and the human resources or marketing department (e.g. internal communication) addresses behaviour. Similar ideas regarding an existing gap of missing out human issues in practice are found in other research (Brooke 1994, Stahl 2005, Backhouse & Dhillion 1996, Backhouse 2004, Sauer 1993). These authors are not addressing the specific field of ICT security, but they use critical research (Brooke 1994) or social constructivism (Stahl 2005, Sauer 1993) to explore the cause of the problem for not reaching the organisations goal. On one hand Brooke identified a quality gap in total quality management (TQM), Sauer why IS projects often fail, and on the other hand Stahl and Backhouse focus on the necessity to build in soft stuff drivers to minimise failures in IS. Our research question focuses on the human factor. Each individual him-/herself is the Achilles heel for security of the organisations infrastructure (Gonzales 2002, Zegers 2000). In practice we found that battling attacks is only reactive and mostly technology dependent. Our critical approach will address relationships and language as central issues to improve the existing decision making (because organisations are wasting time) and to trigger understanding (Thomson and von Solms 1998) for the importance of sensitive and responsive behaviour. Practice shows that expertise (in our context the ICT professional) tends to push the right mind set rather than pull the suitable behaviour (Clegg 2004). The rates of accidents appear to be high, and rates of compliance low. Therefore people have to be triggered to move towards compliance. We do not present solutions but bring ideas with this paper. The recommendations we will present are for managers of organisations and research fellows. We believe there is a need to stimulate further debate concerning whether and how the existing gap can be filled.

2. The extensive nature of literature

To set this interdisciplinary paper in context we wish to make three interrelated points. Firstly, there is a lot of research in the technology area, especially on ICT security and its adoption but a lot of it is best practicesbased and draws attention to the involved technologies itself or service processes supporting the organisations business. The important question is: What is good practice, and perhaps just as importantly, which lessons learned are drawn from bad practices. There are a lot of practice-oriented approaches to ICT security, but only leading vendors and service providers summarise some best practices. In general ICT security is presented as a heterogeneous landscape. Second, our research makes strong claims of the necessity to address the individual user. The effectiveness of ICT infrastructure is dependent on the individual. People enable technologies and processes. Inevitably perhaps, there are general needs, wants and demands of each individual. We recognise that behaviour is dependent on the ability to behave in a right way and the drivers which influence behaviour. Therefore it is crucial to acknowledge individuals ability in spite of perception, experience, education, and training. On the other hand we will critically review drivers such as motivation, attitude, and intention of the individuals. Finally, it is apparent that most of ICT security approaches have to balance technical issues and at the same time human-related changes. We identified some generic points of relevance, which we will bring together to fill the existing gap. Change management, knowledge creation, and communication play an important role in organisations. ICT security has to be addressed continuously to minimise threats and vulnerabilities. Why are most of the organisations wasting time? For ICT security the centre of interest should be the individual. This phenomenon calls Ghosh (1998, 2001) weakest link scenario in his two books. It is not only a buzzword to involve people, to drive participation, and to trigger real ownership of the individuals. It is a real need of organisations, which adjudicates (in the worst case) if an organisation will survive or not. Misbehaviour of only one individual regarding access to vulnerable information at a specific time (in this context high threatening time frame) could give him/her or a third person (with or without his/her knowledge) the possibility to attack, destroy or modify important information. The wave of simplification in ICT networks and the growing complexity (e.g. access anywhere and anytime) makes the tasks of ICT professionals difficult. Vulnerabilities and threats are the counterparts of business needs, which should be addressed through the right perception of the individuals and the suitable action they take. In a way ICT professionals have to deliver a secure system which adds value to the business.

www.ejbrm.com

118

Academic Conferences Ltd

Andreas Wagner and Carole Brooke

3. Identifying the gap

The extensive literature and research on one hand and our gained experience out of practice on the other hand show that traditional views are still dominating this field. The subjective nature of behaviour is rarely addressed but its influence is still increasing (which is also seen and accepted by practitioners and researchers). In this paragraph we will take a deeper look into best practises and standards of organisations, which can be summarised by three areas: technology focus, process focus, and general management. Pye and Warren (2005) are benchmarking security models and framework. They are taking a technologyoriented approach if they claim that security has to become proactive by reviewing and continuously improving security (p.80). As a lot of technology only approaches they underline the importance of measures and policies, so security can only proactively be realised through periodic revision (2005, p.80). We picked out of the mass of research, papers and books their work, because they are drawing relationships to continuous improvement principles of TQM. At this point we cannot agree that it is realistic to reach their mission by focusing on TQM principles (see our former research, especially Brooke 1994). Moreover Pye and Warren (2005) conclude their research by claiming that a security chain is only as strong as its weakest link, but highest importance have measures and policies which are designed to deliver guidance, manageability and consistency (2005, p.86). Again, we disagree because they focus on explicit knowledge enabling management to drive an input-output mechanism, so human behaviour becomes a computed machine. Pye and Warren modify their viewpoint by trying to involve soft stuff, but they see it as an added value if the technology stuff is solved (they call it as a need to encourage a culture of security awareness (2005, p.86). Schneiers book Secrets and lies (2000) pays more attention to the individual by addressing security not as a product. For him security is a process, so he claims not to find the perfect solution but working on a process to secure systems which can not be completely broken. We see the interesting fact that the author looks not for a 100 percent secure system (because it is impossible to realise) but a way to minimise threats and vulnerabilities. For Schneier (2000) the right service process is the key. Therefore he claims to design, implement and maintain suitable processes, which can be best solved by outsourcing of security processes (2000, p.387). We see that this ideology of process-oriented approaches such as IT infrastructure library (ITIL), COBIT, ISO17799 or BS7799 limits the ability of the ICT professionals to perceive subjectivity. These standards generate a shared use of language. In practice ICT professionals of the outsourcer tend to separate fundamentally the outsourced infrastructure to behavioural issues. Neither have they seen the need to involve the individuals nor to look for participation of stakeholders. On the other hand only issues are solved which are defined in a contract and service level agreement (SLA). Consequently behaviour and process are not seen as intertwined aspects. Again, ICT-security faces a general management challenge. Other authors pay more attention to how security is handled by management in general. Especially Hancock (2002) deals with the management of a security event as part of corporate crisis management. We identified in practice that the importance of security gets management attention if a real crisis happened. Regularly the crying wolf syndrome appears this means that every single security problem is treated as a corporate crisis. Consequently not only management attention is rising but general sensitivity to threats and vulnerabilities is exploding. The (re-) action of top management is often not suitable because they set security temporally on highest importance (but if the crisis is gone they are losing focus). Again, these management decisions do not offer any solution to the generic security problem, because not all individuals are involved continuously. Ongoing involvement is fundamentally important to get them participating in the security activities and taking over ownership for it. Besides we see in practice that top management support is important but middle management has to leverage compliance by triggering all users. Our research is inspired by the idea to secure the weakest link, so there is no way to avoid focusing on the individuals behaviour. Knowledge about the entire vulnerability landscape is important but the key is to expose a security ideology to build up the language and involve people to own the weakest link challenge, change the existing environment, and secure it (e.g. minimise incidents). Only the individuals can minimise security risks.

4. Choosing the suitable research methodology

The sophistication of hacker tools is steadily intensifying whilst at the same time resulting in less technical knowledge being needed to break into systems. The necessary resources are given, but does it mean that they are also used? Schneier (2000) put it into the following words: Cyberspace crime includes everything youd expect from the physical world: theft, racketeering, vandalism, voyeurism, exploitation, extortion, con

www.ejbrm.com

119

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (117-124)

games, fraud (p.15). And if we are speaking about vulnerabilities and threats we do not focus only on external attacks or those which are driven by intention. Malicious or benevolent behaviour is possible, which has different effects if the hacker is an expert or a novice (to get more information about the types of threats and vulnerabilities see Stanton et al., 2005). We will address this challenge later on, but in the first step we will explain why we use critical research. Organisations have to bear in mind that every user could be the weakest link. Traditional views assume a way of deducting future occurrences from past ones. This implies objectivity is given. We reject this view point, because of the subjectivity of human theory. Behaviour is socially constructed depending on language and ideologies. Our research emphasises relationships and gives privilege to investigating the organisational context. For us language is central, and we propose that ICT security research should put more focus on meaning and perception. No academic group as far as we can tell has considered a holistic ICT security approach. We do not want to give the impression that this is fully uncharted territory but rather that it is not explored much by management research nor does it take into account the highly subjective nature of ICT security.

5. Using critical research and postmodernism

In this paper we propose a research methodology that harnesses aspects of critical research and postmodernism. We do this because it enables us to access different voices and consider the balance power involved. Organisational behaviour (e.g. leadership, motivation) is strongly qualitative. Broader and richer descriptions help to build up more sensitivity to the ideas and meanings. The main tasks for critical research are to break up established ways of using language and to problematise language. For Alvesson and Deetz (2000) critical theory is participative and dialogic addressing insights, critique, and transformative re-definition. Critical research calls for at least the first 2 elements (Alvesson and Deetz 2000, p. 164). Interpretation produces insights by addressing the non-obvious, by making sense, and by enriching the situation whilst broadening the angle of perspectives to highlight problems and also possibilities. Our critique (through deconstruction of language and balances of power) seeks to undo frozen meanings; an example being that ICT has nothing to do with the soft stuff of research inquiry. It then becomes possible to attend to detail (e.g. behaviour of the individual), to institutional macro-oriented themes (e.g. ICT attacks and cyberspace crime), and to counteract taken-for-granted goals and ideas (e.g. ICT has been bounded as hard stuff while leaving out soft stuff). Our critique explores domination and the repression of ICT labelling. Just as transformative re-definition as a counterpart to insight and critique helps us to produce new ways of seeing and thinking, so ICT-related hard stuff is bridged to the human-related soft stuff. The context for individuals action has to be re-thought, re-felt, and re-experienced. In a way of transformative re-definition we will try to indicate an alternative way, and not wasting valuable management time. Listening to and understanding language and power issues are crucial themes. In Alvesson and Deetz (2000) point of view postmodernism could mean one of three things: (1) using Freud more unconventionally, (2) heeding structuralist language theory from Saussure, and (3) focusing on language. We will focus on the third one in this paper. Language is intrinsically related to meaning and perception, because it becomes central through discursive practises. The combination between postmodernism and critical theory may well provide the best remaining option (Alvesson and Deetz 2000, p. 103). For Alvesson and Deetz (2000) postmodernism is more a pull theory and critical theory more a push theory. The goal is to develop critical sensitivity. Therefore de-familiarization becomes a key (Alvesson and Deetz 2000, p.178). We negate current ICT approaches to get the familiar foreign (e.g. wasting time: The mission impossible with respect to technology-oriented security approaches). Our research will set direction for ICT security, commitment to the individuals and sharpness for the importance of the soft stuff.

6. The subjective nature of ICT security

The subjective nature of ICT security is formatively driven by the individual. Cultural, social, personal, and psychological factors influence behaviour. The latter one includes motivation, perception, learning, beliefs and attitudes influencing behaviour. This paper does not offer the space to discuss all these facets nor

www.ejbrm.com

120

Academic Conferences Ltd

Andreas Wagner and Carole Brooke

present the silver bulletin solution but raises attention to promote subjective approaches within ICT security decision making. Therefore we will pick some of the most influential ones for security. To describe the inability of technology and the importance of behaviour we will describe the case that documents in trash are often more valuable than the same data in a computer. It is a human problem. First of all the right perception of threats and vulnerabilities is crucial. Often people overestimate or underestimate the probabilities of attacks because they do not have the experience, training, education on one hand or the right drivers on the other hand (our term drivers we will explain later on). In the case of there being too many alarms the individuals will learn to ignore them (e.g. in an extreme instance the crying wolf syndrome causes total ignorance). In this case the individual is not driven by anything such as motivation, attitude or intention because people stop asking why things happen. They only do their work without thinking about the consequences. (An example might be where an employee is called by another employee or external partner whom they do not personally know and with whom they exchange confidential data over the telephone because they think the one who is calling will be trustworthy). These cases offer a hint that is important to take a deeper look into why people do not understand risks or are not acting accordingly to the known. Practice shows that there is a big difference how people are behaving at work if you compare it with the behaviour at home. In our point of view it is all about the right combination of ability and drivers. We put the term ability in the context that the individual knows what compliance is. Each individual has to have the understanding why security is crucial for the organisation and how to behave if a certain incident appears (or even having a hint what to do or whom to involve). For us the right perception (what is going on), experience (comparing past incidents with the current problem), fundamental education (why security is so important) and training (how to behave in a certain situation) are intertwined. There are several authors (Hazari 2005, Krueger and Dickson 1994, Sitkin and Weingart 1995) researching the area of perception to get more insights into the notion of perceived behavioural control and self-efficacy. But this research regarding perception alone does not help. For us perception has to be bridged the other aspects. In a way experience has a certain impact on education and training, and vice versa. For an experienced person it is easier to match practical insights with explicit knowledge (which is educated) or tacit knowledge (which is trained). The other way round education and training helps to reach a better learning curve regarding the ability to behave. Experienced behaviour is the outcome. Vom Brocke and Buddendicks research (2005) focuses on skills and competence which is generated by education and training. Again, their research allows us to question education and training in general but a contextualisation is required. We will move forward, albeit incrementally, to generate a holistic view while raising attention to promote suitable security drivers. Drivers are extrinsic and intrinsic parameters to take action according to compliance. In our point of view security drivers can be subsumed by motivation, attitude and intention. For some authors (Cullen 1997, Osterloh and Frey 2000) extrinsic motivation, e.g. using punishment and rewards, is the key. If we put it in the context of security it is not sufficient to use only punishment and rewards because they are reactive. Intrinsic motivation is more important than extrinsic motivation, because it is the only way to transfer tacit knowledge. Tacit knowledge plays an important role for security behaviour. Technology- and process-oriented approaches focus on explicit knowledge and our research shows that there is still a gap regarding secure behaviour. Attitude and intention are more long-term drivers, which we are not addressing further in this paper. To bridge attitude and intention to ICT security we identified a supportive nature of the organisations mission to safeguard existing resources (e.g. security as a long-term objective to stay competitive), but it is a long way to go to get such high-level organisational stuff transferred to the individual. In a way this has an effect on the relationship between ICT and ideology.

7. Filling the gap

Filling the gap means stopping the waste of time by following only technology- and process-oriented security approaches. We have just touched these approaches the organisations implement and maintain in practice. The key to fill the existing gap is that organisations have to pay more continuous attention to the users behaviour. From a critical point of view the individual is unique.

www.ejbrm.com

121

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (117-124)

Ability and drivers have to be understood and organisations have to follow a holistic approach with respect to an embedded organisational development and change. Organisations must trigger a sense of urgency, because business becomes e-business so the survival of organisations is dependent on the security of their core resources. There is no way round to be connected with partners, suppliers, and lots of remote users. Business gets enabled through the use of technology (ICT revolutionised business and there is just a new wave coming up: Leveraging business through web services and service-oriented architectures (SOA)). That is one side of the coin. The other one is gaining vulnerabilities and threats with which the organisation is confronted. Management has to reflect change, to support knowledge creation, and to create a security culture. The environment changes, competition moves and the organisation itself goes through an evolutionary process. Attention has to be paid to human and organisational challenges. Understanding has to be generated, language has to be modified and the suitable behaviour has to be put in place. We see a continuous transition towards ICT security culture change as the right mission to follow. Transition bridges the traditional and the new approach by developing the organisation from ending the traditional views to a new beginning. Our view is supported by Clegg (2004), who calls for a replacement of push-based with pullbased approaches to involve the individuals, to get them participating, and to influence their behaviour by generating ownership. Changing the users behaviour means driving the individual through a psychological transition process. We see knowledge creation in a key role to create a security culture, which minimises the waste of time. In this context we cite Balogun and Jenkins (2003), who claim to re-conceive change management and put it into a knowledge-base perspective. Knowledge generation supports the change process with the mission to generate a security culture. Nonaka and Takeuchis (1995) knowledge creation spiral puts attention to the tacit and explicit dimensions of knowledge while middle management plays an important role to drive this process. In practice wasting time begins and ends with middle management. We underline the importance of ability, especially tacit knowledge, and intrinsic drivers to solve the dilemma of the weakest link. Knowledge creation is dependent on, but also influences, culture. Relationships within the organisation provide a basis for a form of socialisation which is important in exchanging tacit knowledge. Sharing knowledge can be enabled by adequate communication to create useful relationships between the individuals to share their knowledge and learn from each other. For Balogun and Jenkins (2003) an enabling context has to legitimise this behaviour. We come back to the important role of middle management as enabler for secure behaviour. The culture has to trigger thinking and understanding to be efficient and effective regarding the security challenges the organisation is confronted with. The twin of each change is communication. We see that security culture has to be brought to life by communication and can only survive by continuous communication. To generate a common understanding, suitable language and a security view regarding conversations and dialogues with the individuals are necessary. The enabling context (influenced by and supporting ideology) has to be owned by the users and driven by middle management. A lot of socially driven activities have to be orchestrated by individuals (and legitimised by top management) to meet both formal and informal requirements. Language is the key to enable perception and anticipation on one hand and to leverage action and responsiveness on the other hand. From a critical research standpoint a 100 percent security culture is not achievable. It is unrealistic or utopia. Security culture can reduce risks and helps to handle vulnerabilities and threats. An organisation can only move forward into a secure direction if the culture supports the suitable perception of current problems and anticipation of future incidents. Real ownership by the individual regarding security acknowledges the effectiveness of a security culture development. Communication, response, convenience, language, expectations, understanding, participation, and a lot of similar aspects improve the senses to anticipate security vulnerabilities and threats. To see what is going on is not good enough. Each individual has to take action and to leverage his/her responsiveness. In this context we refer to Locke (1991), who focuses on the motivation sequence and developed a model which combines a lot of well-know theories. Overall Lockes model describes that a certain performance outcome can be affected if adequate ability and knowledge is given, but a precondition is that the individual chooses commitment to a certain outcome. Again, we see that the aspect of ownership is stressed, so management has to put security at the heart of the organisation.

www.ejbrm.com

122

Academic Conferences Ltd

Andreas Wagner and Carole Brooke

Barlow and colleagues (2005) explored how to transform business to a sensitive and responsive organisation. For them the organisation has to develop itself upside down, which we would like to see as accompanied by middle management (and legitimised by top management). Again, we can conclude that a knowledge-driven culture with respect to security has to be realised to minimise risks. Barlow and colleagues (2005) underline this no waste mentality (p.191).

8. Conclusions
The traditional view of organisations controlling and knowing all activities has to disappear and to be replaced by a mission generating ownership of all internal and external users to secure the core resources. Yet even if people are committed 100 percent security is impossible. Technology- and process-oriented best practices and standards can not support the mission to have an immune network. Because of its subjective nature, especially the weakest link phenomenon in ICT security, these approaches can not fulfil the mission they are driving for. Organisation development, knowledge creation and security culture are fragments to get the security puzzle solved. Throughout we are arguing in this paper to put the individual into the focus of interest. Understanding language and power are crucial to leverage ICT security (which is supported by technology and processes but enabled by the individuals themselves). In order to protect and deliver organisational success in practice more and more secure organisations get rid of these traditional assumptions and approaches, because they are misguided. The key to secure ICT infrastructure is employees (anticipative) perception and the (responsive) action they take in accordance with the learned and perceived need for an understanding of compliance. There is a need for further research, which we are opening up with our paper.

9. Recommendations
A more critical attitude can support practitioners and researchers, particularly in ICT security. The cumulative effect would be a move to the subjectivity of the individual. In practice more and more organisations stop believing that if vulnerabilities and threats can be understood, risk is manageable by hard stuff-oriented best practices and standards. Holistic research can balance the above mentioned limitations. The mission could be the minimisation of ICT incidents, but it is not realistic to assume this will achieve a secure environment. Overall we tend to connect ICT security to performance goals of the organisation. Only then, can it break free of the view that security is owned by the ICT department. We propose that it is all about people management, especially language and power and its subjective nature.

10. Further research

This paper as a preliminary research of a PhD thesis is backed up by literature of connected areas and the experience we gained in practice and research. We highlight the need for further research in the field of ICT security. Qualitative research in all its facets would generate more insights into this area. Research of a more quantitative nature (e.g. large surveys) could be done to explain further the market views of vendors and analysts, which underline that the understanding of people is one of the key issues securing organisations business. It would be interesting to explore technology- and process-orientated approaches and the way they could be leveraged by a holistic people-centred practice. Perhaps then we can create a mission that is less impossible?

References
Alvesson, M., and Deetz, St. (2000) Doing critical management research, Sage, London. Backhouse, J., and Dhillon, G. (1996) Structures of responsibility and security in information systems, European journal of information systems, 5, 2-9. Backhouse, J. (2004). Risk management in cyberspace, London School of Economics and Politics, London. Balogun, J., and Jenkins, M. (2003) Re-conceiving change management; a knowledge-base perspective, European management journal, 21 (2), 247-257. Barlow, S., Parry, St., and Faulkner, M. (2005) Sense and respond: The journey to customer purpose, Palgrave, London. Brooke, C. (1994) Information technology and the quality gap, Journal Employee Relations, 16 (4), 22-34. Clegg, C.W., and Walsh, S. (2004) Change management: Time for a change, European journal of work and organisational psychology, 13 (2), 217-239. Cullen, D. (1997) Maslow, monkeys and motivation theory, Organization, 4 (3), 355-373. Ghosh, A. K. (1998) E-commerce security, Wiley & Sons, New York. Gosh, A. K. (2001) Security and privacy for e-business, Wiley & Sons, Weinheim. Gonzalez, J. J. (2002) A Framework for human factors in information security, Rio de Janeiro: WSEAS Int. Conf. on Information Security.

www.ejbrm.com

123

ISSN 1477-7029

Electronic Journal of Business Research Methods Volume 5 Issue 2 2007 (117-124)

Hancock, B. (2002) Security crisis management - The basics, Computers and security, 21 (5), 397-401. Hazari, S. (2005) Perceptions of end-users on the requirements in personal firewall software: An exploratory study, Journal of organizational and end user computing, 17 (3), 47-65. Krueger, N. Jr, and Dickson, P. R. (1994) How believing in ourselves increases risk taking: Perceived, Decision Sciences, 25 (3), 385-400. Locke, E. A. (1991) The motivation sequence, the motivation hub, and the motivation core, Organizational behavior and human decision processes, 50, 288-299. Nonaka, I., and Takeuchi, H. (1995) The knowledge-creating company, Oxford University Press, Oxford. Osterloh, M., and Frey, B. S. (2000) Motivation, knowledge transfer, and organizational forms, Organization science, 11 (5), 538-550. Pye, G., and Warren, M. (2005) Benchmarking e-business security: A model and framework, Proceedings of 3rd Australian information security management conference, 80-87, Edith Cowan University, Perth. Sauer, Ch. (1993) Why information systems fail: A case study approach, Waller, Hatfield. Schneier, B. (2000) Secrets and lies, Wiley & Sons, New York. Sitkin, S. B., and Weingart, L. R. (1995) Determinats of risky decision-making behaviour: a test of the mediating role of risk perceptions and propensity, Academy of management, 38 (6), 1573-1592. Stahl, B. C. (2005) Reflective responsibility for risk, unpublished paper, Leicester. Stanton, J. M., Stam, K. R., Mastrangelo, P., and Jolton, J. (2005) Analysis of end user security behaviors, Computers and security, 24 (2), 124-133. Thomson, M. E., and von Solms, R. (1998) Information security awareness: Educating our users effectively, Information management and computer security, 6 (4), 167-173. vom Brocke, J., and Buddendick, Ch. (2005) Security by learning the contribution of e-learning to security awareness management, unpublished paper, Muenster. Zegers, N. (2000) Battling insider attacks: Deterring improper security behaviour, Erasmus University Rotterdam, Rotterdam.

www.ejbrm.com

124

Academic Conferences Ltd