Annals of Valahia University of Targoviste.

Geographical Series (2017), 17(2): 145-151

DOI: 10.1515/avutgs-2017-0013
ISSN (Print): 2393-1485, ISSN (Online): 2393-1493
© Copyright by Department of Geography. Valahia University of Targoviste

ASSESSING PERCEPTION OF FLOODS WITHIN THE FRAMEWORK OF

VULMIN PROJECT: METHODOLOGICAL REMARKS

Andra Costache1, Laura Comănescu2, Alexandru Nedelea2

1
Valahia University of Târgoviște, Department of Geography St. Lt. Stancu Ion, no. 34-36,
Târgoviște, Romania, 0245206105 Email: andra_cost@yahoo.com
2
University of Bucharest, Faculty of Geography, Department of Geomorphology – Pedology -
Geomatics St. Nicolae Bălcescu, no. 1, Bucharest, Romania, 0213053775 Email:
lauracomanescu@yahoo.com, alexnedelea10@yahoo.com

Abstract
The paper emphasizes the methodological steps undertaken in the assessment of floods perception, which
was conducted in tree study areas from Romania (Banat Plain, Bend Subcarpathians and Siret river basin).
The reasons behind the design of the questionnaire and interview guide are explained, in relation with the
research objectives. The research endeavor allows not only to better understand the local context of
vulnerability to floods and flash floods, but also to incorporate the results in a broader framework of
vulnerability assessment, enhancing the quality of scientific services provided to stakeholders.
Keywords: questionnaires, interviews, floods perception, Banat Plain, Siret river basin, Bend Subcarpathians

1 INTRODUCTION

The current tendency in the research of different types of hazards (including floods)
incorporates several directions, such as human perception analysis (Bradford et al., 2012),
probabilistic risk assessment (Heijmans, 2001), integration of several indicators sets so as to
determine vulnerability indices, involving population, entrepreneurs, civil society and decision-
making factors (categories which may generically be grouped together under the name of
stakeholders), as well as scientists, in the attempt to assess vulnerability to extreme events and to
build resilience.
The papers focusing on human perception to different types of hazards have seen a boom
over the past decades (Bland, O’Leary, Farinaro, Jossa, Trevisan, 1996; Loewenstein, Weber, Hsee,
Welch, 2001; Chesney, Cannibal, Baines, 2002; Sjöberg, 2002; Weber, Blais, Betz, 2002; Cheval,
2003; Bickerstaff, 2004; Cannon, Davis, 2004; Sjöberg, Moen, Rundmo, 2004; Armaş, 2008; APA,
2009; O’Sullivan et al., 2012); however, there is a long series of preoccupations with perception of
population on the environmental changes and relating extreme events (Fischhoff and Furby, 1983).
Within the framework of the VULMIN project, an important component of the endeavours
meant to assess human vulnerability to environment changes and floods was to understand the local
communities’ perception of these pressure factors. Integration of the knowledge and expertise of
both population and local authorities into the vulnerability assessment enabled not only defining the
local context of vulnerability to floods and environmental changes but also taking some important
steps towards reaching a major general objective of the project, i.e. provision of scientific services
to local population and decision-making factors (stakeholders).
The perception studies may be conducted in any research stage: in the preliminary stage,
with a view to defining objectives and working hypotheses, in the stage of ongoing researches (with
the aim of validating hypotheses) and in the final stage, in order to validate conclusions.

145 Unauthenticated
Download Date | 4/14/18 9:56 AM
 Within the framework of the VULMIN project, in order to assess perception of floods and
environmental changes, use was made of two sociological research techniques, i.e. the sociological
questionnaire (to assess perception of population) and sociological interview (in connection to
perception studies involving local authorities). For both techniques used, classic stages of a
sociological research were taken, namely: a. projection of research; b. research itself; c. analysing
information and formulating conclusions.

2 RESEARCH PROJECTION

Given the interdisciplinary nature of the endeavour (environmental psychology,

environmental sociology, perception of hazards), analysis was made of current methodological
approaches in this field and preliminary documentation was made on local vulnerability context in
the areas selected for the case studies, located in the Banat Plain, the Bend Subcarpathians and the
lower basin of the Siret river, which were previously affected by floods.
The objectives of the perception studies were established following a thorough
documentation conducted at local level. For instance, the general and the specific objectives of the
perception studies conducted in the Banat Plain were:
1. Understanding the local context of the human community’s vulnerability to environmental
changes and extreme events
1.1. Understanding population’s perception of the main types of environmental changes
occurred in the Banat Plain and their intensity.
1.2. Understanding population’s perception of changes occurred in the frequency and
intensity of extreme events.
1.3. Identification of extreme events perceived as threats by the local population.
1.4. Identification of potential consequences of environmental changes and extreme events
on local community welfare.
2. Identification of the flood vulnerability components
2.1. Highlighting the relation between components of hazard perception in the area under
study i.e.:
- previous experience (as a witness to flood)
- awareness of exposure to flood
- concern/fear of floods
- actions taken to prevent and mitigate flood effects, adopted at individual and community
level
2.2. Identification of information required, at individual and community level, in order to
diminish flood vulnerability.
2.3. Identification of efficient means of communication between local community and
decision-making factors, of the one hand and, local community and scientists, on the other
hand.

For the mentioned case study, we consider of paramount importance the research hypotheses
laid down in relation to objective 2.1, as the relation between hazard perception components is
essential for the consolidation of the adaptive. Therefore, we started from the premise that previous
direct experimentation of the analysed extreme events favours raising awareness of flood exposure,
and both lead to higher concern level/degree of fear of future occurrence of similar phenomena,
which entails initiation of some measures to prevent and reduce impact.
Considering the research objectives and hypotheses, the population subjected to research
was established. Irrespective of how a survey is conducted, the representativeness of the sample is
extremely important, both for the purpose of the study and in relation to the population involved. It
is considered that when the number of inhabitants is lower and the space is limited, it is easier to
obtain representativeness, in comparison to large spaces and dense population. The selection criteria

146 Unauthenticated
Download Date | 4/14/18 9:56 AM
used for the questioned people mainly took into consideration their residence in the areas affected
by floods and their previous experience of floods (direct or indirect witnesses to floods).
At the end of this stage, option was made for procedures to be used in the research. In
addition, instruments necessary to collect information were prepared (e.g. list of questions to be
included in the questionnaire and the interview guide). Both the questionnaire and the interview
may be applied over the telephone, via mail or face to face. As far as accuracy and significance of
results are concerned, there are advantages and disadvantages for each of them. In case of a survey
conducted on the basis of a letter mailed (self-administered questionnaires), the costs are lower, but
the number of responses gathered drops considerably and the number of questionnaires impossible
to validate therefore rises; in addition, accuracy of the responses is affected as the respondent is not
provided with further information, and the interview operator cannot offer clarifications when
required.
A telephone survey involves additional costs and a certain pressure on the subject;
moreover, the person conducting the survey may influence the subject. However, the operator can
offer a series of explanations which may render the responses more accurate. The most efficient
method is the face to face method, namely the one of the questionnaires administered by interview
operators. In this case the refusal rate is lower, the respondent may receive real-time clarifications,
and the interview operator may also register further information such as non-verbal language
elements. In terms of the disadvantages relating to questionnaires administered by interview
operators one may determine higher costs and the fact that some responses may be influenced (e.g.:
by question rephrasing or due to causes relating to attitude and personality of the operator). In this
research, the questionnaires were applied by members of the research team who have been
instructed with regards to suitable conduct during performance of the survey in the field.
Research instruments were used to collect quantitative and qualitative data. In order to
obtain optimal results, the stage when questionnaire questions are formulated is very important.
Research instruments, i.e. questionnaire and interview guide, were elaborated for each and every
case study, starting from the objectives and the research hypotheses previously laid down.
The applied questionnaires include open-ended items, structured items (closed-ended
questions) and mixed questions. Special attention was attached to introduction of some control
questions to check how honest the responses were as well as the respondents’ attention and how
well they understood the questions. By content of the questions used, they are: questions requesting
factual data (they have a concise, direct, straightforward content); opinion questions (which aim
for thinking, opinion, attitude and reaction of the interlocutor); motivation questions (which
highlight arguments regarding opinions or facts); testimonies (which request opinions coming from
people who represent a community). Questions formulated were clear, short, not repetitive,
impossible to suggest the response and applicable to large categories of respondents.

2.1.Development of the Questionnaire and Interview Guide for the Case Study
Conducted in Cudalbi, Galaţi County

One of the case studies conducted in the Siret river basin was in Cudalbi, a rural settlement
located in Galați county. In order to understand and compare the opinion of the population in
relation to authorities we used the questionnaire (applied to inhabitants) and the interview (applied
to authorities). The data collected were statistically processed and later interpreted.
The questionnaire on perception of flood-related risk included 26 questions (most multiple
choice or open-ended questions), structured in 3 sections: data on subjects, data on community and
households, data on previous experience and flood-related information level (Comănescu and
Nedelea, 2016):
Section 1- Data on subjects comprise domicile identification elements (important data to
locate respondents in relation to the area affected by floods), age (provide information on
previous experiences of people interviewed) and gender (element which confers a specific

147 Unauthenticated
Download Date | 4/14/18 9:56 AM
 psychology and implicitly a type of reaction), training level (very important for the
characteristics of the responses to questions in the following sections).
Section 2- Data on community, households and dwellings aim to set the homestead location
compared to the area affected by floods, type and material used in building the house, age,
height and number of floors, existing rainwater drainage system in the region and other
structural measures.
Section 3 – Experience, level of knowledge and information with regards to natural hazards
is the most extensive. This chapter focuses on how floods were managed (pre-, during and
post-event), namely how the population became conscious and perceived the involvement of
the authorities. The focus in on: population’s experience in relation to floods, determination
of causes leading to floods, volume of information received by population concerning the
probability of occurrence of such phenomena, reaction of authorities and actions taken by
authorities, raising awareness on volunteerism.
The interview applied to local authorities in Cudalbi comprised several questions (all open-
ended questions), of which main purpose was to correlate their opinion with the citizens’ opinions
on pre- and post-event management (Comănescu, Nedelea, 2016). The questions targeted:
understanding the causes leading to floods, information on warning systems and existing action
plans for the parish, existence of signalling drills and their frequency, how the authorities reacted
and managed to mobilize resources, actions taken by authorities, citizens’ reaction during the crisis
and in relation to assessing needs and distribution of relief support; existence of civic spirit and
mutual aid within the community.

2.2.Development of Questionnaires for Case Studies Conducted in the Banat

Plain and in the Bend Subcarpathians

The questionnaire “Perception of the Environmental Changes and Natural Hazards in the
Banat Plain” included 57 items, grouped together in 3 main sections.
a. Local Context of Human Communities’ Vulnerabilities to Environmental Changes and
Extreme Events. The section focuses on the landscape changes, climate changes, quality of water
and exposure to extreme events in general (Costache et al., 2014). This section has been developed
to achieve the specific goals of objectives 1.1. - 1.4. In order to avoid excessive limitation of the
respondents’ selection choices (and therefore include young adults in the questionnaire sample), the
questions concerning landscape and climate changes referred to the past two decades.
We considered that the terminology used (“landscape”, “climate changes”) may have
different meanings to respondents. This is the reason why the questions in this section were
predominantly structured and facilitated collection of responses and understanding of their
connotations. An example of an item designed to this end is: Mention the changes in the local
climate that you have seen over the past years; this item also stands for a control question in
connection to other two items (Do you consider that the dangerous climate phenomena have
become more frequent in the past 20 years?, respectively What was the manifestation of the
dangerous climate phenomena over the past 5 years compared to the previous timeframe?), as the
phenomenon evolution tendency was stated for these items (e.g.: wind intensifications were more
frequent and stronger).
A succession of items (How afraid are you of the dangerous natural phenomena which are
to come?, Which were the consequences of the dangerous natural phenomena which have affected
your area in the past 10 years? and Do you consider that the dangerous natural phenomena
occurring in the past 10 years have affected your family welfare and to what extent?) was designed
to check one of the working hypotheses formulated for objective 1.3. Identification of extreme
events perceived as threats by local population, respectively: The greater the social and economic
impact of an extreme event, the greater the fear of such an event occurring in the future.

148 Unauthenticated
Download Date | 4/14/18 9:56 AM
 b. Water quality. A separate section was assigned to questions referring to water quality
formulated so as to highlight the drinkable water supply sources, potential changes in quality and
the impact on the inhabitants’ health. Data were later on correlated with the results obtained after
the analysis of the water samples from these areas.
c. Vulnerability to floods. The section included questions regarding: i. personal experience
of respondents and individual post-event rehabilitation; ii. raising awareness to floods and fears of
floods; iii. perception of actions necessary to prevent and mitigate effects of floods. In this section
there were more open-ended questions and the respondents could offer details on how they
experienced floods occurred in their area. The questions in this section allowed highlight of
connections between the components of the perception of hazards, which were investigated in the
following stage, namely the stage of data analysis through some correlation tests.
To understand the respondents’ perception on the capacity to react and adapt to flood
occurrence, questions were designed so as to find out: level of confidence in different institutions
involved in such endeavours, perception of authorities’ preparedness level in case of occurrence of
similar events as well as the role of the social support networks in the mitigation of the flood
impact. Considering the difficulties relating to evacuation of population in case of previous events
occurred in the region (2005), an item was introduced to strictly understand how prepared the
population is to follow instructions given out by authorities in relation to crisis situations.
The identification questions were detailed with a view to capturing the influence of the
social and demographic, economic and cultural variables (ideological options) on the perception of
the environment. Additionally, consideration was taken of the elements enhancing the vulnerability
in relation to households, as well as family structure, characteristics of households and geographic
position compared to floodable areas.
In the case study conducted in the Bend Subcarpathians, one applied a 64-item questionnaire
organized in four sections: local context of vulnerability of human communities exposed to floods
and flash floods; vulnerability to floods/high floods; quality of drinkable water in the areas affected
by flash floods; identification questions.
Some of the items were similar to the items in the questionnaire applied in the Banat Plain.
Nevertheless, the peculiarities of the region as well as the specific objectives in this case study,
oriented on perception of flash floods, imposed a different structure, with specific items. Therefore,
some new questions were introduced in the section addressing flood vulnerability. They refer to:
 personal experience and raising awareness to exposure (e.g.: existence of warnings
regarding imminent occurrence of floods/flash floods in the locality, detailing on the impact
of previous events on households, individual behaviour during flash floods which affected
the locality, perception of annual and seasonal frequency of floods, connection between
configuration of water catchment area and occurrence of flash floods);
 post-event rehabilitation (individual involvement in the locality rehabilitation actions);
 measures to prevent and mitigate effects of floods (the extent to which the population has
been informed on how to behave in case of floods and flash floods, communication channels
favoured by respondents).

3 CONCLUSIONS

The stage of actual collection of information was preceded by the pilot survey which was
used to test the validity of the research instruments. For the case study conducted in the Banat Plain,
one considered that the field research in 2005 (conducted after occurrence of floods) provided the
information required to validate the questionnaire. The questionnaire in the Bend Subcarpathians
(2015) was projected on the experience acquired in the two case studies (Banat 2005 and 2013).
The field campaigns were carried out in a single stage in Banat Plain (May 2013) and in two stages
in the Bend Subcarpathians.

149 Unauthenticated
Download Date | 4/14/18 9:56 AM
 Following validation of questionnaires and encoding of information, use was made of
statistic analysis instruments available in Excel and SPSS such as descriptive statistics of frequency
analysis type, parametric (chi-square) and non-parametric correlation tests. These instruments were
necessary to process the databases.
The described research stages allowed making correlations between the environmental
context and perceptions, attitudes and population’s behaviours. In addition, a useful psychological
dimension was integrated into the vulnerability assessment. The results may be analysed in terms of
endeavours necessary to increase the capacity to respond and adapt to floods and environmental
changes. Furthermore, some similar items used in case studies applied in different regions provide a
comparative perspective.
Population’s perception of environment is shaped by the social and cultural context and the
collective history, which overlap individuals’ aspirations, needs and history. The psychological
dimension of the man-environment relation has been investigated throughout time and various
approaches were used, such as:
- deterministic approaches, with explicative models focused, for instance, on adaptation or
environment constraints, which imply the need for control and predictability;
- interactionist and transactional approaches, which focus on the individual’s capacity to resist
stress and on cognitive aspects of the man-environment relation (e.g.: manner in which
individuals assess threats and manner in which they perceive adjustment and adaptation to
such threats);
- systemic approaches, which bring on, among others, the need to understand social
representations (informative, cognitive, normative elements, beliefs and opinions – Moser,
2009) and the manner in which the environment defines social, cultural and economic
identity of the individual. Within the same approaches, the concept of “behaviour setting”
emerged (Barker, 1968), respectively a behavioural scheme associated to a particular place
(Bechtel, Marans and Michelson, 1987).
Irrespective of the type of paradigm by which we choose to explain the relation between the
individual and the environment, it is essential to accept the influence of the psychological
dimension (with all unquantifiable elements in its structure) on the human vulnerability to extreme
events.

ACKNOWLEDGEMENTS

The current study was conducted in the framework of the project “Vulnerability of the environment
and human settlements to floods in the context of Global Environmental Change (VULMIN)”
(PN‐II‐PT‐PCCA‐2011‐3.1‐1587) funded by the Ministry of National Education for the period
2012-2016 (http://www.igar-vulmin.ro). The questionnaire applied in the lower basin of the Siret
river (Cudalbi) was developed by professor PhD Liliana Zaharia.

REFERENCES
American Psychological Association (APA) (2009), Psychology & Global Climate Change.
Addressing a multifaceted phenomenon and set of challenges. A report of the American
Psychological Association Task Force on the Interface Between Psychology and Global Climate
Change, http://www.apa.org/science/about/publications/climate-change.aspx, accessed on
25.03.2016
Armaş, I. (2008), Percepţia riscului natural: cutremure, inundaţii, alunecări de teren, [Natural risk
perception: earthquakes, floods, landslides], Editura Universităţii din Bucureşti, Bucureşti, 217
p. [in Romanian]
Barker, R.G.(1968), Ecological psychology: Concepts and methods for studying the environment of
human behavior, Stanford University Press, Stanford, California

150 Unauthenticated
Download Date | 4/14/18 9:56 AM
Bechtel, R.B., Marans, R. W., Michelson, W., (editors) (1987), Methods in environmental and
behavioral research, Van Norstrand, New York
Bradford R.A., O'Sullivan J.J., van der Craats I.M., Krywkow J., Rotko P., Aaltonen J., Bonaiuto
M., De Dominicis S., Waylen K., Schelfaut K. (2012), Risk perception - issues for flood
management, Nat. Hazards and Earth System Sciences, 12, 2299-2309
Bickerstaff, K. (2004), Risk perception research: socio-cultural perspectives on the public
experience of air pollution, Environment International, 30, 6, 827–840.
Bland, S. H., O’Leary, E. S., Farinaro, E., Jossa, F., Trevisan, M. (1996), Long-term Psychological
Effects of Natural Disasters, Psychosomatic Medicine, 58,18–24.
Cannon, T., Davis, I. (2004), At Risk: Natural Hazards, People’s Vulnerability and Disasters, 2nd
edition, Routledge, London, N.Y., 471 p.
Chesney, P. E. S., Cannibal, G. L., Baines, R. (2002), The Development of a New Holistic
Assessment Model for the Inclusion of Public Perceptual Values in the Decision-Making
Process, Risk Analysis, III, WIT Press, Southampton, Boston, 171-180.
Cheval, S. (2003), Percepţia hazardelor naturale. Rezultatele unui sondaj de opinie desfăşurat în
România (octombrie 2001-decembrie 2002), Riscuri şi catastrofe, II, 49 - 59.
Comănescu L., Nedelea A. 2016, Floods and public perception on their effect. Case Study: Tecuci
Plain (Romania), year (2013), Procedia Environmental Sciences - International Conference –
Environment at a Crossroads: SMART approaches for a sustainable future, 32, 190 – 199, doi:
10.1016/j.proenv.2016.03.024
Costache, A., Bălteanu, D., Sima, M., Dumitraşcu, M., Grigorescu, I., Dragotă, C. (2014), A
Community-Based Assessment of the Exposure to Environmental Change and Extreme Events in
the Banat Plain (Romania), 14th SGEM GeoConference on Ecology, Economics, Education And
Legislation, SGEM2014 Conference Proceedings, 1, 3-10.
Fischhoff, B., Furby, L. (1983), Psychological dimensions of climatic change, in Chen R. S.,
Boulding, E., Schneider S. H. (editors), Social science research and climate change: An
interdisciplinary appraisal, The Netherlands: Reidel, 180–203
Heijmans, Annelies (2001), Vulnerability – A Matter of Perception, Disaster Management Working
Paper, IV, Benfield Grey Hazard Research Center, London, 17 p.
Loewenstein G. F., Weber E.U., Hsee C. K., Welch N. (2001), Risk as feelings, Psychological
bulletin, 127, issue 2, p. 267.
Moser, G. (2009), Introducere în psihologia mediului [Introduction to environmental psychology],
Polirom, Iaşi, 285 p. [in Romanian]
O'Sullivan J. J., Bradford R. A., Bonaiuto M., De Dominicis S. , Rotko P., Aaltonen J., Waylen K. ,
Langan S. J. (2012), Enhancing flood resilience through improved risk communications, Nat.
Hazards Earth Syst. Sci., 12, 2271-2282
Sjöberg L. (2002), Factors in risk perception, Risk Analysis, 20, issue 1, 1-12
Sjöberg, L., Moen, B. E., Rundmo, T. (2004), Explaining Risk Perception. An Evaluation of the
Psychometric Paradigm in Risk Perception Research, Rotunde Publikasjoner, Trondheim,
Norway, 33 p.
Weber E U, Blais A‐R, Betz N. E. (2002), A domain‐specific risk‐attitude scale: Measuring risk
perceptions and risk behaviors, Journal of behavioral decision making, 15, 4, 263-290.

151 Unauthenticated
Download Date | 4/14/18 9:56 AM