doi:10.1111/j.0361-3666.2010.01175.

Individual preparedness and mitigation

actions for a predicted earthquake in
Istanbul
Sıdıka Tekeli-Yeşil, Necati Dedeoğlu, Marcel Tanner, Charlotte Braun-
Fahrlaender and Birgit Obrist1

This study investigated the process of taking action to mitigate damage and prepare for an earth-
quake at the individual level. Its specific aim was to identify the factors that promote or inhibit
individuals in this process. The study was conducted in Istanbul, Turkey—where an earthquake
is expected soon—in May and June 2006 using qualitative methods. Within our conceptual
framework, three different patterns emerged among the study subjects. Outcome expectancy,
helplessness, a low socioeconomic level, a culture of negligence, a lack of trust, onset time/poor
predictability, and normalisation bias inhibit individuals in this process, while location, direct
personal experience, a higher education level, and social interaction promote them. Drawing on
these findings, the paper details key points for better disaster communication, including whom
to mobilise to reach target populations, such as individuals with direct earthquake experience
and women.

Keywords: awareness, earthquakes, individual preparedness, Istanbul, mitigation,

Turkey

Introduction
Rapid population growth, urbanisation, environmental degradation, poverty and
social inequalities pave the way for increased impact of natural hazards in the mod­
ern world. Global trends in natural hazards show that, although there has been a
decrease in fatalities, the number of people affected by and the estimated damage
due to disasters have risen since the middle of the twentieth century.2 This situation
has led to some changes in disaster management practice. In recent years, risk reduc­
tion—that is, prevention of the impact of a hazard, mitigation of the damage, and
preparedness for the event—has attracted the most attention. In addition, it has been
acknowledged that risk reduction comprises a continuous series of tasks carried out
by social, economic, governmental and professional sectors and should not only
involve specialised experts in various fields. This is why disaster risk reduction is
described in the context of the International Strategy for Disaster Reduction (ISDR)
as a shared responsibility of governments, communities and individuals (ISDR, 2004).
Corresponding to ISDR definitions, studies and experiences indicate that the ac­
tive participation of individuals and the community is essential for success of any kind
of disaster management activity (Burningham, Fielding and Trush, 2008). However,
there are few examples of good practices. On the one hand, a ‘top-down’ approach

Disasters, 2010, 34(4): 910−930. © 2010 The Author(s). Journal compilation © Overseas Development Institute, 2010
Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA
Table 1 Selected literature on individual hazard preparedness and factors affecting hazard-related actions

Type of hazard Country (city/state/area) Factors Reference Comments

Flood England and Wales • Social class Burningham, Fielding and The paper focused mainly on
• Personal flood experience Trush (2008) awareness
• Length of time in residence
• Denial/rejection of ‘at-risk’ status
• Invisibility of the risk
Earthquake Turkey (Istanbul) • Location Kalaça, Bay and Çalı (2007) Economic level (known socio-
• Demographic characteristics (age, tenure) economic level of the neighbour-
• Education level of individuals hood where respondents live)
• Risk perception
• Economic level
• Earthquake experience
• Direct experience
Earthquake Turkey (Antalya) • Age (only for awareness) Dedeoğlu (2006) Fatalism was not a major
• Earthquake experience factor affecting action
• Education level of individuals
• Economic level of individuals
Earthquake Turkey (Istanbul) • An absence of capacities Inelmen, Işeri-Say and The study examined participa-
• Unfamiliarity with the community-based organisation (CBO) Kabasakal (2004) tion in a relevant local CBO
• Lack of time
• Fatalism
• Lack of trust
Cyclone Australia (Cairns) • Risk perception Anderson-Berry (2003) –
• Direct personal cyclone experience
• Hazard awareness education
Earthquake Turkey (Istanbul) • Education level of individuals Fişek et al. (2002) Risk perception and attitudes
• Economic level of individuals (income) do not demonstrate a straight­
• Gender (only for risk perception) forward relationship with action
Earthquake United States (southern California • Location Lindell and Prater (2000) All stated factors cause hazard
and western Washington) • Demographic characteristics adjustments within a causal
• Earthquake experience chain
• Hazard intrusiveness
Individual preparedness and mitigation actions for a predicted earthquake in Istanbul
911

Type of hazard Country (city/state/area)
Earthquake United States (Los Angeles)
Volcanic eruption New Zealand (North Island)
Earthquake Japan (Tokyo-Yokohama) and
United States (Los Angeles)
Earthquake Iran (Tehran and Rasht)
Erosion Bangladesh
Earthquake United States (California)
Earthquake United States (Los Angeles)
912 
Factors Reference Comments
• Demographic characteristics Lindell and Whitney (2000) Risk perception does correlate
• See one’s self as responsible with intentions and action
• Outcome expectancy (efficacy of adjustments)
• Direct personal experience (important for awareness; Johnston et al. (1999) Risk perception and knowledge
Sıdıka Tekeli-Yeşil et al.
stimulates information seeking) do not show a straightforward
• Normalisation bias relationship with action
• Optimistic bias
• Economic level of individuals (income) Palm (1998) The paper mainly investigated
• Perceived vulnerability of the home the impact of culture on risk
• Age (only in Japan—not in US) perception and action and the
• Sense of control over one’s destiny (only in US—not in Japan) differences between residents
of Tokyo and Los Angeles
• Fatalistic attitudes Asgary and Willis (1997) Economic status was signifi-
• Hazard perception cant only in one of the
• Economic status research areas
• Economic level of individuals Mamun (1996) The sydy investigated resettle-
• Education level of individuals ment in safer areas
• Erosion experience
• Helplessness and fatalism
• Kinship
• Dependence on cultivation
• For risk perception: Mileti and Fitzpatrick (1992) Risk perception was important
• Risk communication factors (i.e. salience, message style and for information seeking
frequency etc.)
• For action:
• Information seeking
• Social support and interaction (i.e. knowing that other people
are taking steps)
• Denial Lehman and Taylor (1987)
• Poor predictability/onset time
 Individual preparedness and mitigation actions for a predicted earthquake in Istanbul 913

is still most common in disaster/risk management, while on the other, people appear
uninterested in and unwilling to take preparedness action and to reduce risks (Dedeoğlu,
2006; Lehman and Taylor, 1987).
  This paper focuses on the latter with the overall aim of investigating the process
of taking action to mitigate earthquake damage and prepare for such an event at the
individual level in Istanbul, Turkey, where scientists predict that a major earthquake
will occur in the near future (Parsons et al., 2000). The specific aim of the study is
to identify the factors that promote or inhibit individuals in this process. This paper
draws on the first part of an ongoing study in Istanbul.

Background
Many studies in various countries with different economic and social profiles have
investigated the factors that motivate individuals to take hazard precautions or hinder
them in doing so. Despite the different backgrounds of the countries and the nature
of the hazards probed, they seem to share certain characteristics in terms of human
hazard responses. Socioeconomic and education level, hazard experience, and demo­
graphic features seem to be common factors in various studies in this field, as sum­
marised in Table 1.
  The studies presented in Table 1 demonstrate that individual preparedness for
earthquakes or other natural hazards is a complex process determined by many inter­
acting social, economic, environmental and personal factors. In this context, we have
developed a conceptual framework that we employed while collecting and analysing
the data (see Figure 1). To summarise the framework: in the case of risk of hazard
occurrence, awareness of risk—covering knowledge of the danger and its conse­
quences, how to cope with it, and risk perception—is a prerequisite for undertaking

Figure 1 Conceptual framework for disaster mitigation and preparedness action

Source: Tekeli-Ye ş il (2010).

914  Sıdıka Tekeli-Yeşil et al.

protective measures. The next step is to evaluate the costs to and the benefits for
individuals in terms of the economic, physical and psychological costs of taking
action. According to the results of this evaluation, the sequence might continue with
a person’s attitudes towards and intentions regarding taking action to mitigate earthquake
damage and to prepare for such an event, followed by the action taken. Intervening
personal, social or environmental factors can influence, positively or negatively, each
step and the transition phases. Sometimes, because of a disaster or some other devel­
opment, the action concerned may be followed by an evaluation of all or some of the
steps. The evaluation phase is not considered in this study.

Methodology
The study presented in this paper is the first part of a larger investigation. Qualitative
research methods, namely focus-group discussions (FGDs) and in-depth interviews,3
were used to test the research questions.

Research site
Active faults traverse Turkey. The Marmara region, in which Istanbul is located, expe­
rienced two severe earthquakes on 17 August and 12 November 1999, with mag­
nitudes of 7.4 and 7.2 on the Richter scale, respectively. According to the Turkish
Court of Accounts (2002–03), these events claimed nearly 18,000 lives and caused
severe damage. Istanbul was only slightly affected by them compared to other cities
in the Marmara region, but some authors have estimated that there is a 62 per cent
(± 15 per cent) probability of an earthquake of magnitude ~7 on the Richter scale in
the Marmara region in any 30-year period (Parsons et al., 2000). The province of
Istanbul has one metropolitan municipality and 32 district municipalities. We selected
two of these districts for the study: Bakırköy, located in the first-degree earthquake
risk zone and expected to be one of the districts that would suffer the most in the
expected earthquake; and Beykoz, located in the second-degree risk zone and expected
to be one of the districts that would suffer the least.4 Within the districts, there are
a variety of neighbourhoods (subdistricts) of different socioeconomic status.

Sample and data collection

We conducted 12 FGDs in total. The participants came from existing social groups—
such as people working in the same place, members of a neighbourhood association
or painting-course students—in the two districts. All participants were adults who
were able to make decisions about their homes. We considered the economic status
and education level of the participants in order to assign the participant groups to
a high, moderate or low socioeconomic level (SEL), based on information from
key contacts on participant profiles (see Table 2). We used a semi-structured question
guide during all of the discussions. The question guide was checked via a pre-test
 Individual preparedness and mitigation actions for a predicted earthquake in Istanbul 915

Table 2 Distribution of the focus-group discussions

Socioeconomic level (SEL) Bakırköy Beykoz Total

Number of FGDs (number of people Number of FGDs

who participated in each FGD)

Low SEL 2 (8 + 7) 2 (10 + 6) 4

Moderate SEL 3 (7 + 8 + 6) 2 (10 + 8) 5

High SEL 2 (10 + 7) 1 (6 ) 3

Total 7 (53) 5 (40) 12 (93)

discussion and covered topics in the conceptual framework. The groups consisted
of between six and 10 people. Gender was not considered while building up the
groups; however, two groups consisted of only men and one group consisted of only
women. The other nine groups involved both females and males. Most of the discus­
sions took place where participants were likely to meet (such as in the workplace or
in the office of the association of which the participants were members).
  In addition, we carried out 11 in-depth interviews with authorities, experts in the
social and natural sciences, administrators and those responsible for implementing
various mitigation and preparedness programmes, and other key informants.
  All discussions and interviews were audiotaped with the permission of the partici­
pants and then they were transcribed verbatim. In all FGDs, beside the moderator,
an observer was present to take notes and observe responses. The Maxqda® software
application was employed for data management. Codes and sub-codes were organised
according to the framework of the study and the question guide to extract relevant
sections for the analysis.

Results
Analysis of the collected data revealed that participants did not follow a common path
towards action. Within our conceptual framework, three different patterns emerged:

• The first and most common pattern was interruption of the process towards action
after the ‘awareness’ step.
• The second and less common pattern was that, after the ‘awareness’ step, respond­
ents progressed to a subsequent step or steps, but again intervening factors inter­
rupted the sequence before the respondents successfully introduced measures.
• The third and least common pattern was the completion of the sequence through
to action.

  The results presented take into consideration these three patterns. First, we discuss
patterns 1 and 2 and the factors that inhibit further steps towards action. Next, we
916  Sıdıka Tekeli-Yeşil et al.

analyse pattern 3 and explore those factors that aided participants and made it possible
for them to complete the sequence. Citations from the discussions and in-depth inter­
views are provided throughout the text to allow respondents to use their own words.

Pattern 1: interruption of the sequence after the ‘awareness’ step

Participants demonstrated relatively good knowledge of the earthquake risk and its
consequences, but knew less about how they might adequately respond. Their
knowledge was usually derived from statements made by scientists in the media.
Participants typically knew what kind of a risk existed in particular parts of Istanbul:
those from Bakırköy were aware that their district is characterised by relatively
more risk due to its proximity to the fault and less stable soil conditions, while those
from Beykoz knew that their area is characterised by relatively less risk. Participants
also knew quite well what kind of consequences Istanbul would endure in the event
of an earthquake. They expected the damage to be somewhat localised and for it to
be especially severe in locations where gecekondus5 are common. According to many
participants, accessibility would be a big problem due to damage to or the closure of
roads and highways.
  Participants also discussed the eventual economic and psychological ramifications.
Some pointed out that, during the 1999 earthquakes, Istanbul was able to demonstrate
solidarity towards and provide support for the affected cities; however, if Istanbul
itself were severely affected, none of the neighbouring cities would be able to act in
the same manner, since they lack the resources to do so and will likely be affected
themselves. Participants, though, had less and in some cases only superficial knowl­
edge of what to do to prepare for an earthquake or to mitigate damage. When
people were asked about possible mitigation and preparedness measures, the most
common ones mentioned were an earthquake bag/kit and having the building
tested for construction quality, followed by fixing tall furniture to the wall. However,
those who knew that an earthquake bag is part of earthquake preparedness did not
necessarily know about its contents.
  Although the general perception of risk with regard to Istanbul city was high among
all groups, participants did not think that they themselves were at risk. Some of the
participants, who were at risk, tended to rationalise their situation as follows:

IB: ‘a ground investigation was done here in Osmaniye [a subdistrict of Bakırköy] by ex-
perts. They said, “Here is better than other parts of Bakırköy”. Here is limestone, not stable
and soft, but still it is better compared with other places, in Bakırköy’.
FB: ‘Here is the most solid ground within Bakırköy’.
EH: ‘There were limestone-kilns, in the past, stones were obtained from here’.6

  In-depth interviews confirmed the above findings. Experts also thought that people
knew the risk, but that this did not lead to critical awareness and finally to action.
 Individual preparedness and mitigation actions for a predicted earthquake in Istanbul 917

Pattern 2: interruption of the sequence after an evaluation of costs and

benefits and/or attitudes/intentions steps
Some of the respondents who were aware of the risk tried to take further steps. Most
of the participants who could move forward to informal cost–benefit analysis accorded
priority to short-term needs. They preferred to invest their limited time and money
in meeting daily needs and not in strategies related to the ‘probability’ that an earth­
quake will happen in 20–30 years:

‘Much research about Turkey has shown that our culture has low levels of future orientation
. . . Especially the absence of the habit of future orientation, seeing everything in the
context of today or interrelated with the past, is a very important factor. Perhaps it is not
inappropriate in a country such as Turkey. As people have accepted this, they have authen-
ticity. It is not possible to make plans, as done in countries like Sweden or Switzerland for
three to four to 10 years, in a country that has to deal with continuous economic crises,
disasters and social problems’.7

  Furthermore, some of the participants thought that some measures, such as an

earthquake kit, remind people daily of the threat and cause psychological problems.
Instead of confronting such problems, they prefer not to take steps:

PH: ‘. . . another reason is the psychological issue. There is death at the end of such a
disaster. To prepare for death is not something that everybody can tolerate’.8

  While many of the participants demonstrated positive attitudes towards precautions,

a few were inclined or intended to take steps towards earthquake mitigation and
disaster preparedness. Some participants mentioned that, if the state provided cheap
retrofitting credit, they would consider adapting their houses.
  Small events, such as our discussions, may motivate people and serve as a reminder
of existing intentions.

AH: ‘Personally I am thinking to take further steps. Due to this discussion, an earthquake
bell is ringing in my ears. I will have a look at my earthquake bag’.9

Patterns 1 and 2: factors creating obstacles

With regard to personal, social and environmental intervening factors, the most
evident ones generating obstacles to further progress in patterns 1 and 2 are pre­
sented below.

Personal intervening factors

Outcome expectancy (absence of belief in micro-scale, individual and non-
structural measures)
Almost all participants expressed belief in preparedness and mitigation activities on
a macro scale, yet many did not express much belief in the effectiveness of measures
918  Sıdıka Tekeli-Yeşil et al.

taken on the micro scale or at the individual level. In addition, they believed more
in structural (such as retrofitting a building) than non-structural mitigation or indi­
vidual preparedness measures (such as fixing tall furniture to the wall or possession
of an earthquake bag).

SH: ‘I do not understand, could you please explain it to me? I nail the shelf to the wall
in order to prevent it falling on my head. What would happen if this shelf falls on my
head when the building is collapsed? In Japan they say: “that thing should not fall down,
this thing should not injure me”. Building is strong in Japan, it does not collapse’.10
OB: ‘Overall measures should be taken. Our houses are not the problem. When measures
are not taken in total in your living environment, then the measures taken by yourself
have no meaning. Unless an improvement [restoration] begins in the whole of Istanbul,
the measures that we would take personally have no meaning’.11

  In some cases, though, explanations by participants of the absence of belief in the

effectiveness of measures and individual preparedness seem to relate to transferring
responsibility to others (such as to the state, government or local authorities).

Helplessness
Fatalism or religious attitudes towards disasters did not play an important role in
disaster preparation. However, this comment needs some refinement, because help­
lessness was very common and in some cases was openly portayed as fatalism, with
helplessness as the underlying factor and fatalism the coping strategy. Several times
during the discussions we had to explore more deeply and clarify contradictions
between fatalistic expressions and measures that individuals intended to introduce
or had done so. One type of response was: ‘first you have to take all possible meas­
ures and then trust God’. A second type is illustrated below:

MB2: ‘Huh, helplessness we give up, it means helplessness you throw in the towel . . .’.
AB: ‘It means you acquiesce to everything from the beginning, for example you are a big/
large man. Me? This man could beat me, it is so simple, when this earthquake comes,
I can’t do anything’.12
ZB: ‘The Turkish people are under constant danger . . . Earthquakes are just one of
these [dangers]. Therefore, people living in the community should be a little bit fatalist.
It should not be seen as related with the religious attitude; otherwise [if people are not a
little bit fatalist], we would all be mentally ill.13

  According to some of the respondents to the in-depth interviews, traumatisation

and false perceptions of the damage also led to a sense of helplessness.

‘Except the people who went to Kocaeli [the most affected area in the 17 August 1999 earth-
quake] to join solidarity activities, most of the Istanbul residents witnessed the earthquake
on television. This witnessing experience was a little bit traumatic and battering, because
 Individual preparedness and mitigation actions for a predicted earthquake in Istanbul 919

those who did not personally observe the consequences of the earthquake and only saw it on
television, with screens of collapsed buildings and battered people images on televisions they
perceived it as if Kocaeli had totally collapsed . . .’.14

Social intervening factors

Low SEL
The taking of mitigation and preparedness actions was not very common among
group participants with a lower socioeconomic level. Regardless of the district in
which they lived, almost all of them had implemented either no or very few measures:

SB: ‘I haven’t seen any of my neighbours taking such measures for the earthquake, I also
haven’t heard about it from anybody. As I said a few minutes ago, it depends on resources
. . . you know people earn their living with difficulty, nobody thinks about precautions.
Okay, they know that an earthquake will hit, but, for example, my neighbour, now he
is looking for a basement to rent, really nobody cares whether the walls are cracked or
whatever it is’.15

  The experts interviewed were divided on this issue: those with engineering or
administrative backgrounds highlighted the importance of economic status, while
those implementing public preparedness programmes or investigating such pro­
grammes said that there is no direct relationship between taking precautions and
economic status.

‘We have seen that the most important problem is financial . . . We asked how important
is it for you to arrange financial sources or credit support for the retrofitting: the response
was 70 per cent very important, 20 per cent important . . . the only solution is financial
support’.16
‘I don’t have the impression that people without economic or material problems undertake
such measures more comfortably or easier . . . The objection “we don’t have economic means
to do this” arises when we work with low-income groups. But in reality this is just an
L-profile [an appliance to fix tall furniture to the wall), simple to use, can be found in
any market selling building materials, in any hardware store. Even when we explain this,
the criticism “we don’t have economic means to use this” still comes up, but it is not really
true. Again safety culture, I relate such objections to the fact that the cornerstones of safe
attitudes in the community are not yet in place’.17

Culture of negligence in the context of multiple and/or constant risks in

the community
A culture of negligence, which can be described as ignorance of security rules or
safety regulations in general in many aspects of daily life (such as while driving or
while working), was mentioned by respondents during in-depth interviews and group
discussions in different terms or using different examples (such as a lack of a culture
of safety). Due to the existence of multiple risks and/or constant threats in daily
920  Sıdıka Tekeli-Yeşil et al.

life, people and even institutions act in a negligent manner with respect to such
risks. Unfortunately, in a vicious cycle, this culture of negligence is also increasing
or reproducing existing risks in daily life. Statements by group participants also under­
lined multiple risks in daily life as a cause of the negligence:

OB: ‘In our community the problem is: this signboard can also fall down on my head [as
security measures may not necessarily be applied or controlled], I may also fall into a hole
dug by the municipality [and not filled later], I may also die because of a natural disaster.
Here is a country full of surprises. Therefore, earthquakes or such events are extreme things
for us’.18

Lack of trust
Trust (in the institutions that plan, apply and control retrofitting projects or other
preparedness programmes) was flagged as an important determinant in making
decisions in favour of mitigation of damage. Neither commercial nor public insti­
tutions have gained individuals’ trust. In the FGDs, participants pointed out also
that scientists and universities enjoy a high degree of trust among group members,
although recent disputes among scientists over the magnitude and the location of
the epicentre of the forecasted earthquake have damaged this somewhat.

Environmental intervening factors

Onset time/poor predictability and normalisation bias
Earthquakes are sudden-onset hazards and can strike at any time without warning.
Therefore, measures should be constant; they cannot be seasonal or periodic. Partici­
pants frequently mentioned such features as a hindrance to pre-emptive action:

ÖB: ‘Of course then [if people know when an earthquake will hit] a measure would be
taken, now what should the people do for an earthquake that will happen in 25 years? It
will happen in 25 years, also what will I be till that time? . . . [W]hen scientists would
say “it will happen in two years” people adjust themselves according to two years not to
25 years. Now if they would say “in two years the earthquake will hit” I would move
to a new place, I would go to my village [ from which he migrated] [as] the ground of my
village is more solid.19

  Both in the discussions and in the interviews it was noted that risk was soon for­
gotten after life returned to normal in 1999. Small-scale earthquakes in the country
or an earthquake somewhere else in the world may serve to refresh minds with regard
to intentions and actions, but these phases are usually very short.

‘This is a troublesome task, if I would give an example from myself; I carried a whistle,
spare money with me all the time in the beginning. But even I leave the whistle behind
when I change my bag. Even I, as a conscious educator, do this. My earthquake bag was
ready and it was ready standing somewhere in the hallway, now it has been removed’.20
 Individual preparedness and mitigation actions for a predicted earthquake in Istanbul 921

‘We asked about the earthquake bag: in the first years [after the earthquakes] 50 per cent
of the participants raised their hands [meaning they have an earthquake bag]; this fell to
zero more recently’.21

Pattern 3: completion of the sequence with action

As mentioned above, most of the participants had good knowledge of earthquake
risk and the consequences. Knowledge of coping with earthquakes, though, varied
among group members. The participants of groups with moderate and high SELs
and with direct experience of an earthquake were best informed about how to deal
with earthquakes at the individual level. In addition, these people were keener to
acquire additional information. One of the respondents in the in-depth interviews,
who was preparing and presenting a weekly radio programme on mitigation of
damage and earthquake preparedness, described his audience as people who are in­
terested in the subject, are responsive, and who are already preparing themselves.
  Participants of groups with moderate and high SELs and participants with direct
experience of an earthquake demonstrated positive attitudes and more intention to
engage in further actions.
  Many participants had taken some kind of action, but it is difficult to assess the
quality and the completeness of these precautions. Although there was a difference
in action taken between groups in Bakırköy and Beykoz (that is, between low- and
high-risk zones), there was a relatively greater difference between SELs within and
between these districts.
  The most frequently mentioned precaution taken, as noted above, is possession of
an earthquake kit/bag: this was also one of the most well-known preparatory steps.
Fixing tall furniture securely to walls was the second most cited action for mitigat­
ing damage and preparing for an earthquake. Placing large or heavy objects on
lower shelves and storing breakable items in low or closed cupboards with latches
were also highlighted, although less often, even though such measures cost next to
nothing to implement.
  A geotechnical inspection of the building site, an assessment of building quality,
and retrofitting the building were activities that had been undertaken primarily by
participants of groups with moderate and high SELs. However, some participants
in the groups with low or moderate SELs had carried out ‘building investigations’
or ‘retrofitting’ themselves or they had been done by a friend with a relevant pro­
fessional background (such as a civil engineer, an architect or a foreman), but who
might not necessarily be working in this specialised field.
  Having a plan for what an entire family should do and on reuniting during and
after an earthquake was rarely mentioned; when it was, it was mainly by participants
with moderate or high SELs. Participants with a family plan tended to be those with
children. Very few families, though, had rehearsed their plan.
  Other precautions cited include learning what to do during and after an earth­
quake. While earthquake insurance is compulsory, it was not referred to very often
in discussions, although some people said that they did have it. Changing residence
922  Sıdıka Tekeli-Yeşil et al.

was rarely mentioned. Those who had done so were either better off or such a move
was already on their family’s agenda. In these cases, a safe neighbourhood or a
building known to be safe was chosen. Some female participants stated that they
had learned how to switch off the gas and the electricity. Having a fire extinguisher
and knowing how to use it and earthquake training or participation in a volunteer
group were the least mentioned preparedness activities.
  There were references to ‘extreme’ precautions, especially among high-SEL par­
ticipants, such as having a steel/iron shelter or their own rescue apparatus.
  The experts interviewed thought that, even though there are good examples of
individual preparedness, a high percentage of people in Istanbul are totally unpre­
pared for an earthquake.

Motivating factors for pattern 3

With regard to personal, social and environmental intervening factors, the most
evident motivating factors leading to completion of the sequence through action are
presented below.

Personal intervening factors

Direct personal experience
In most groups, participants who demonstrated the highest level of preparedness and
were most motivated to take action were those who had some direct experience of
the 1999 earthquakes. People who had experienced these events directly (such as
through losing close relatives or friends and/or participating actively in solidarity
action) and those who had participated in subsequent rescue efforts were more likely
to take precautions and show a greater intention to engage in further action:

SB: ‘I was one of the volunteers who ran to act in solidarity the next day after the ’99
earthquake. I was sent by the company [ for which he was working at that time] . . . we
didn’t have any knowledge about organising rescue or solidarity activities, we couldn’t do
anything, we saw the disaster, we were in an incapable position, in terms of helping people,
rescuing them, a terrible disaster, terrible damage. After this earthquake, I thought about
what I can do at home, I did some little things. I saw the disasters there after two earthquakes,
also at Kaynaşlı [a subdistrict close to the epicentre of the second earthquake in 1999], there
my father-in-law died. After that I was more worried . . .’.22
EB: ‘You asked us why we have been involved in the group [a local, non-professional
volunteer rescue team]: to be conscious. I lost many relatives during the earthquake [on
17 August 1999]. The son of my uncle [his corpse] was brought out from the rubble after
27 hours and I was like this [standing] there, we couldn’t do anything; we were waiting
behind the red band and waiting for his body to come out’.23

Higher education level

Mitigation and preparedness actions were more common among group participants
with higher education levels. Even within the groups with a low SEL, those with
 Individual preparedness and mitigation actions for a predicted earthquake in Istanbul 923

more education were more likely to take precautions. Awareness was clearly better
among participants with better education.
  In-depth interviews also indicated that education level plays a significant role in
disaster mitigation and preparedness, especially in relation to knowledge of the risk,
the potential consequences, and how to cope with an event.

Social intervening factors

Social interaction
Participants who had introduced some measures had often tried to persuade those
in their circle to take similar steps. Respondents found it easier to implement meas­
ures when they saw examples that persuaded them about the efficacy of the initiatives.
For example, many participants did not believe that insurance would be beneficial
or would function properly in the event of a major earthquake. However, a mem­
ber of a high SEL group in Beykoz had obtained not only compulsory earthquake
insurance but also additional private insurance for both his home and his workplace
after seeing one of his friends rebuild his factory after the 17 August 1999 earthquake
using funds from an insurance claim.
  Many people in the FGDs and the in-depth interviews had not come across ex­
amples of individual preparedness, which would have motivated them.

‘In Turkey the majority don’t believe, accept anything without having seen it. This is how
we spread our impact: we have a strategy like the following: we select three neighbour-
hoods [subdistricts] in each district. Because when these neighbourhoods are involved, the
neighbouring neighbourhoods take the attitude “we should get involved too”’.24

  Furthermore, many of the interviews indicated that social interaction and social
networks play a key role in implementing mitigation and preparedness programmes
among women and in districts where traditional relationships still exist. In almost
all of the in-depth interviews, muhtars25 were also flagged as key to tapping into pub­
lic and different subdistrict social networks.

Environmental intervening factors

Location of the house
Although the difference in the taking of precautions between higher and lower risk
areas was not too great, more participants in Bakırköy (higher-risk area) have intro­
duced measures compared to participants in Beykoz (lower-risk area).

Discussion
The study supplied very rich data, from which we have presented only the most prom­
inent findings. Because of the nature of the qualitative methods, we cannot generalise
based on these findings.
924  Sıdıka Tekeli-Yeşil et al.

Table 3 Comparison of the study findings and the findings in cited literature*

Literature Our study

Obstacle factors

Low socioeconomic level  

Outcome expectancy (absence of belief in the efficacy of  

measurements)

Helplessness  

Culture of negligence No information 

Lack of trust  

Onset time/suddenness of onset  

Normalisation bias  

Fatalism / 

Motivating factors

Location (living in higher-risk areas)  

High risk perception / 

Direct personal experience  

Hazard experience  

Higher education level  

Social interaction  

Note:  indicates association with taking action;  indicates no association.

  The study was designed to understand better action taken by individuals to

mitigate damage and to prepare for a future earthquake. We considered the whole
process, therefore, and could not evaluate any one theme in great detail—each
point warrants a separate investigation. Table 3 compares the findings of the study and
existing literature.
  The risk of an earthquake in the Marmara region was highlighted after the earth­
quakes in 1999 and this information has been broadly disseminated among the
citizens of Istanbul, including the FGD participants. The amount of knowledge of
the risk and its consequences was considerable, and participants presented a realistic
appraisal of the risk. However, in general, participants were less informed about
how to cope with an earthquake at the individual level, although those with high­
er education levels were better informed than those with lower education levels.
The source of this knowledge may have been myriad information in the media and
interpersonal discussions after the 1999 earthquakes. Yet, much of this informa­
tion—which people did not seek so much as were exposed to—focused on the
ramifications of the recent earthquakes or the characteristics of the predicted one,
and not necessarily on adequately preparing people for the next event. The work
of Mileti and Fitzpatrick (1992) on information-seeking behaviour might explain
 Individual preparedness and mitigation actions for a predicted earthquake in Istanbul 925

this gap between knowledge of risk and potential consequences and how to cope
with it. Usually, knowledge played a minor role in the taking of precautions when
it was obtained passively, but it facilitated precautions when it was acquired actively.
  Risk perception among the participants with respect to Istanbul itself was high,
but in both districts, risk perception for self was lower than risk perception for the
city as a whole. High-risk perception for Istanbul did not seem to have an influence
on action taking. Risk perception led to action if participants personalised the risk.
Further research is needed to study the relationship between risk perception for self
and undertaking protective measures, and how people personalise risk.
  Our findings indicate that immediate or short-term benefits are quite important
for making decisions on mitigation and preparedness activities. In addition, prepared­
ness actions might entail psychological costs for some individuals.
  Even though economic status plays an important and direct role in structural meas­
ures, it should not influence many non-structural measures. However, economic
status was frequently cited in the FGDs as a barrier even to non-structural measures
that would cost little or nothing. Here the role of economic status may be indirect,
and education level or outcome expectancy may be confounding this finding. The
findings showed that some of the building investigations were carried out by resi­
dents themselves or by friends with an appropriate professional background, but
who are not directly responsible or appointed to do such work. This might be seen as
a sign of intention to engage in mitigation, but in the context of a lack of resources
or the existence of economic problems. The difference among experts on the role of
economic status seems to relate to the type of measure. Structural measures, accorded
priority by administrators and engineers, cost more and hence these individuals iden­
tified economic status as important, whereas their colleagues in less investment-
intense areas did not do so. These differences highlight the importance of adopting
a multidisciplinary approach to disaster management.
  Some studies have identified fatalism as playing a role in disaster preparedness in
Turkey (Inelmen, Işeri-Say and Kabasakal, 2004) while according to others, fatalism
does not seem to be a major factor in earthquake preparedness (Dedeoğlu, 2006). Our
findings show that fatalism and religious attitude do not have a significant impact
on mitigation and preparedness activities. Almost all participants, including those
with a strong religious belief, were aware of the reasons for earthquakes and the causes
of damage. However, a sense of helplessness was very common and was sometimes
expressed as if it were fatalism. Risa Palm (1998) in her study in Japan suggests that,
in some cultures, acceptance of destiny does not imply passive acceptance of fate,
but instead a realistic assessment of elements that one cannot control. This perspec­
tive may also apply to the scenario we investigated. Qualitative studies such as this
one seem best placed to explore fully such attitudes with all their contradictions and
inconsistencies. However, witnessing the disaster through the media led to help­
lessness and disbelief about protective behaviour among some group participants.
The images broadcast on television produced an inaccurate perception of damage and
precautions: the impression of the 17 August 1999 earthquake in Izmit (Kocaeli)
926  Sıdıka Tekeli-Yeşil et al.

was that absolutely everything had been destroyed; although the level of damage was
very high, in fact, only five per cent of buildings were totally destroyed (Boğaziçi
University, 2003). This false picture also generated disbelief in the effectiveness of
mitigation and preparedness measures.
  Group participants showed low levels of trust in actors in the construction industry
and its related services and in sources of information. This problem is complex and
requires more intensive investigation. Given the trust shown in general in universi­
ties and scientists, these actors might play an important role in disaster communication.
  Our findings on the role of onset time and poor predictability in taking action
support those of Lehman and Taylor (1987), who suggested that individuals at risk
of a highly probable catastrophic event of unknown timing may cope with the threat
through denial and obliviousness. In addition, though, normalisation of life between
the previous and the next damaging event caused group participants to forget about
the risk or to perceive there to be no urgency to act. Low levels of future orientation
in society (Inelmen, Işeri-Say and Kabasakal, 2004) might be another reason why onset
time and normalisation bias act as moderators in the process of taking action.
  In addition to the existing findings in the literature, our study revealed that a
culture of negligence and the existence of constant multiple risks in daily life led
group participants to take earthquake preparedness less seriously or not to prioritise
earthquakes among other risks. Group participants tended to be unconcerned about
various risks in their daily lives and did not take security measures in other aspects
of life (such as while driving). A high percentage of smokers (PIAR, 1988) and a
low percentage of users of safety belts while driving a car (Traffic Research Center,
1999) in Turkey verify these statements by group participants and experts on the
culture of negligence.
  In the groups, education level, direct experience of an earthquake and socioeco­
nomic level were all found to play key roles in almost all action steps.
  After the two earthquakes in 1999, many people went to the disaster area either
to search for their relatives/friends or to offer support. They witnessed the impact
of the earthquake directly. Among the group participants, this experience was a
strong facilitator of earthquake preparedness. The experiences of such individuals
could be mobilised to reach different groups in subdistricts or neighbourhoods.
  Our data indicate that social interaction and social networks were important fac­
tors motivating group participants to undertake protective measures. Women and
muhtars were highlighted as key persons to play a part in social interactions. Further
research is needed to examine how such interactions occur among different social
groups in order to enhance the effectiveness of public programmes.
  In conclusion, our findings mainly correspond with earlier studies on the roles of
socioeconomic and education levels, outcome expectancy, helplessness, lack of trust,
suddenness of onset, normalisation bias, location, direct personal experience and
social interaction. Furthermore, a cultural phenomenon—negligence—was also
found to have an association with failure to implement earthquake precautions
among group participants.
 Individual preparedness and mitigation actions for a predicted earthquake in Istanbul 927

Recommendations
It is widely acknowledged in the disaster community that hazards trigger disasters
when they encounter vulnerabilities—hence the basis of disaster management is vul­
nerability reduction. Our study shows that a lower socioeconomic and education
level, which are commonly listed as components of people’s vulnerability, are the main
factors that hinder participants in taking earthquake precautions. However, the causes
of such vulnerabilities are rooted in international and national political, economic
and social factors26 and there is little in the way of recommendations on these issues
that falls within the scope of this paper. Below, therefore, we offer some recommen­
dations for practical approaches to enhancing existing mitigation and preparedness
programmes and for further research.
  Our study pointed up two main problems with current public activities that aim
to increase risk awareness and corresponding mitigation and preparedness activities.
The first issue is the content of such activities. Any kind of activity that seeks to
increase awareness among the public should focus more on what individuals can do
to prepare themselves or to reduce their own losses than on (without excluding)
information on actual risk and its features, about which people already have some
knowledge.
  The second issue concerns how this information is communicated. Awareness is
a prerequisite for taking action; however, conventional awareness programmes that
merely disseminate information are not very useful in helping people take further
steps. Such programmes should take into consideration the socioeconomic and
education level of target populations and use interactive methods so that people can
personalise the risk and acquire the ability and self-confidence to cope with it.
Women, muhtars and people with direct experience of an earthquake might be among
the key persons to mobilise to reach target populations. Scientists, who enjoy a high
level of trust, might also play a role in disaster communication; their message, though,
should concentrate on measures that could be taken rather than confusing commu­
nications about faultlines and the magnitude of the forecasted earthquake. Mitigation
and preparedness programmes can involve some methods that allow social interac­
tions to occur, such as persuasive examples of successful actions taken by both
‘ordinary’ people and public figures that clearly show the effectiveness of the meas­
ures. Social interaction might also be used to disseminate information among tradi­
tional neighbourhoods and women. Furthermore, in light of statements made in the
groups about belief in the effectiveness of macro- rather than micro-scale measures,
state institutions should set an example to the public by introducing both structural
and non-structural measures and fulfil their responsibility to secure citizens’ lives
and well-being.
  People tend to take precautions that do not cost much in the way of time and
money. If more extensive precautions—such as retrofitting of buildings—were
sponsored, more people might engage in these activities. There should be invest­
ment in developing easier means of mitigation and preparedness. Efforts to overcome
the culture of negligence and create instead a culture of safety in the community
928  Sıdıka Tekeli-Yeşil et al.

would help not only in implementing earthquake precautions, but also many other
measures related to various daily risks. Schools might be a starting place for such
actions. Safety in daily life (with regard, for instance, to traffic, the home and the
neighbourhood) could be integrated into the curriculum. Considering the statements
about normalisation bias, it seems that keeping the earthquake hazard on the agenda
by providing frequent messages to the public, aids preparedness.
  All of the issues raised in this paper could be examined in more depth, but we
particularly recommend further research on the following:

• the relationship between one’s own perception of risk and undertaking measures;
• how people personalise the risk; and
• the lack of trust and how to overcome this in disaster communication.

  Finally, we suggest further research on quantification and the generalisation of our

findings in this paper.

Correspondence
Sıdıka Tekeli-Yeşil, Department of Epidemiology and Public Health, Swiss Tropical
and Public Health Institute, Socin Str. 57, P.O. Box 4002 Basel, Switzerland.
Telephone: +41 61 2848284; fax: +41 61 2848105
E-mail: sidika.tekeli-yesil@unibas.ch or sidikatekeli@hotmail.com

Endnotes
1
Sıdıka Tekeli-Yeşil is Researcher at the Swiss Tropical and Public Health Institute, Switzerland;
Necati Dedeoğ lu is Director of the Department of Public Health, Medical Faculty, Akdeniz
University, Turkey; Marcel Tanner is Director of the Swiss Tropical and Public Health Institute,
Switzerland; Charlotte Braun-Fahrlaender is a Project Leader at the Swiss Tropical and Public
Health Institute, Switzerland; and Birgit Obrist is a Project Leader at the Swiss Tropical and Public
Health Institute, Switzerland.
2
See http://www.emdat.be/natural-disasters-trends (accessed 12 April 2010).
3
For detailed information on these methods, see (among other works): Denzin and Lincoln (2000);
Tashakkori and Teddlie (2003); Morgan (1988); Bernhard (2000); and Kitzinger (1995).
4
For detailed information on earthquake zones and expected site-dependent intensity distribution,
see Afet Işleri Genel Müdürlüğü Deprem Araştırma Dairesi/General Directorate of Disaster Affairs,
Earthquake Research Department, http://www.deprem.gov.tr; and Boğaziçi University (2003).
5
Gecekondus (unauthorised squats) are usually makeshift one-storey houses built illegally on govern­
ment land on the outskirts of the city.
6
Conversation between a female and two male participants, FGD, Bakırköy—moderate SEL.
7
In-depth interviews, male, scientist.
8
Female, FGD, Bakırköy—moderate SEL.
9
Female, FGD, Beykoz—moderate SEL.
10
Female, FGD, Beykoz—moderate SEL.
11
Female, FGD, Bakırköy—moderate SEL.
 Individual preparedness and mitigation actions for a predicted earthquake in Istanbul 929

12
Males, FGD, Bakırköy—low SEL.
13
Male, FGD, Bakırköy—high SEL.
14
In-depth interviews, female, social worker, who has coordinated public mitigation/preparedness
programmes.
15
Male, FGD, Bakırköy—low SEL.
16
In-depth interviews, male, engineer, who is coordinating a structural mitigation programme.
17
In-depth interviews, female, social worker, who has coordinated public mitigation preparedness
programmes.
18
Male, FGD, Bakırköy—high SEL.
19
Male, FGD, Bakırköy—low SEL.
20
In-depth interviews, female, disaster preparedness educator of a relevant association.
21
In-depth interviews, female, coordinator,disaster preparedness education unit of a university institute.
22
Male (who has undertaken many mitigation and preparedness measures at home), FGD, Beykoz—
moderate SEL.
23
Male (who had undertaken many mitigation and preparedness measures at home and a member of
a local volunteer rescue team), FGD, Bakırköy—moderate SEL.
24
In-depth interviews, male, programme manager of a preparedness project.
25
A muhtar is the elected head of a subdistrict/neighbourhood.
26
For detailed information on the components of vulnerability and their links to root causes, see
Wisner et al. (2003).

References
Anderson-Berry, L.J. (2003) ‘Community vulnerability to tropical cyclones: Cairns, 1996–2000’.
Natural Hazards. 30(2). pp. 209–232.
Asgary, A. and K.G. Willis (1997) ‘Household behaviour in response to earthquake risk: an assessment
of alternative theories’. Disasters. 21(4). pp. 354–365.
Bernhard, H.R. (2000) Social Research Methods: Qualitative and Quantitative Approaches. Sage Publications,
Thousand Oaks, CA.
Boğaziçi University (2003) Earthquake Risk Assessment for the Istanbul Metropolitan Area. Final Report.
Boğaziçi University Press, Istanbul.
Burningham, K., J. Fielding and D. Thrush (2008) ‘It’ll never happen to me’: understanding public
awareness of local flood risk’. Disasters. 32(2). pp. 216–238.
Dedeoğ lu, N. (2006) ‘Knowledge, attitude and practice of residents of Antalya, Turkey about earth­
quake preparedness’. Paper presented to the International Disaster Reduction Conference, Davos,
Switzerland, 27 August–1 September.
Denzin, N. and Y. Lincoln (eds.) (2000). The Handbook of Qualitative Research. Sage Publications,
Thousand Oaks, CA.
Fi ş ek, G.O., N. Yeniçeri, S. Müderrisoğ lu and G. Özkarar (2002) ‘Risk perception and attitudes
towards mitigation’. Paper presented to the second annual meeting of the International Institute for
Applied Systems Analysis (IIASA) and the Disaster Prevention Research Institute (DPRI) of Kyoto
University on ‚Integrated Disaster Risk Management: Megacity Vulnerability and Resilience’,
Laxenburg, Austria, 29–31 July.
Inelmen, K., A. Işeri-Say and H. Kabasakal (2004) ‘Participation lethargy in disaster preparedness
organizations within the framework of a Turkish CBO’. International Journal of Sociology and Social
Policy. 24(10/11). pp. 130–158.
ISDR (International Strategy for Disaster Reduction) (2004) Living with Risk: A Global Review of
Disaster Reduction Initiatives. 2004 version. ISDR Secretariat, Geneva.
930  Sıdıka Tekeli-Yeşil et al.

Johnston, D.M., M.S. Bebbington, C. Lai, B.F. Houghton and D. Paton (1999) ‘Volcanic hazard
perceptions: comparative shifts in knowledge and risk’. Disaster Prevention and Management. 8(2).
pp. 118–126.
Kalaça, S., A. Aytekin Bay and Ş. Çalı (2007) ‘Istanbul’da Depreme Kar şı Bireysel Düzeyde Önlem
Almayı Belirleyen Faktörler’ (Factors affecting earthquake preparedness at individual level in
Istanbul). Paper presented to the eleventh National Public Health Congress, Denizli, Turkey,
23–26 October.
Kitzinger, J. (1995) ‘Qualitative research: introducing focus groups’. British Medical Journal. 311. pp. 299–302.
Lehman, D.R. and S.E. Taylor (1987) ‘Date with an earthquake: coping with a probable, unpredict­
able disaster’ Personality and Social Psychology Bulletin. 13(4). pp. 546–555.
Lindell, M.K. and C.S. Prater (2000) ‘Household adoption of seismic hazard adjustments: a comparison
of residents in two states’. International Journal of Mass Emergencies and Disasters. 18(2). pp. 317–338.
Lindell, M.K and D. Whitney (2000) ‘Correlates of household seismic hazard adjustment adoption’.
Risk Analysis. 20(1). pp.13–25.
Mamun, M.Z. (1996) ‘Awareness, preparedness and adjustment measures of riverbank erosion-prone
people: a case study’. Disasters. 20(1). pp. 68–74.
Mileti, D.S. and C. Fitzpatrick (1992) ‘The causal sequence of risk communication in the Parkfield
earthquake prediction experiment’. Risk Analysis. 12(3). pp. 393–400.
Morgen, D. (1998) ‘Practical strategies for combining qualitative and quantitative methods: applica­
tions to health research’. Qualitative Health Reseacrh. 8(3). pp. 362–376.
Palm, R. (1998) ‘Urban earthquake hazards’. Applied Geography. 18(1). pp. 35–46.
Parsons, T., S. Toda, S. Stein, A. Barka and J.H. Dieterich (2000) ‘Heightened odds of large earth­
quakes near Istanbul: an interaction-based probability calculation’. Science. 288. pp. 661–665.
PIAR (Piar Gallup Ara ş tırma Şirketi) (1988) Sigara Alı şkanlıkları ve Sigarayla Mücadele Kampanyası
Kamuoyu Araştırması Raporu (Cigarette Smoking Habits and Campaign of Fighting with the Cigarette
Gallup Poll Report). PIAR, Istanbul.
Tashakkori, A. and C. Teddlie (eds.) (2003) Handbook of Mixed Methods in Social and Behavioral Research.
Sage Publications, Thousand Oaks, CA.
Tekeli-Yeşil, S. (2010) ‘Being prepared. Vulnerability in hazard-prone megacities: the case of Istanbul’.
In H. Brauch et al. (eds.) Coping with Global Environmental Change, Disasters and Security Threats,
Challenges, Vulnerabilities and Risks. Springer Verlag, Berlin. Forthcoming.
Traffic Research Center (1999) ‘Ülkemizde emniyet kemeri kullanımı’ (Usage of safety belt in our
country). Traffic Research Center Publications, Ankara.
Wisner, B., P.M. Blaikie and T. Cannon and I. Davis (2003) At Risk: Natural Hazards, People’s Vulner-
ability, and Disasters. Second edition. Routledge, London.