Copyright © 2013 by the author(s). Published here under license by the Resilience Alliance.

Frankenberg, E., B. Sikoki, C. Sumantri, W. Suriastini, and D. Thomas. 2013. Education, vulnerability, and
resilience after a natural disaster. Ecology and Society 18(2): 16. http://dx.doi.org/10.5751/
ES-05377-180216

Research, part of a Special Feature on Education and Differential Vulnerability to Natural Disasters
Education, Vulnerability, and Resilience after a Natural Disaster
Elizabeth Frankenberg 1, Bondan Sikoki 2, Cecep Sumantri 2, Wayan Suriastini 2 and Duncan Thomas 1

ABSTRACT. The extent to which education provides protection in the face of a large-scale natural disaster is investigated.
Using longitudinal population-representative survey data collected in two provinces on the island of Sumatra, Indonesia, before
and after the 2004 Indian Ocean tsunami, we examine changes in a broad array of indicators of well-being of adults. Focusing
on adults who were living, before the tsunami, in areas that were subsequently severely damaged by the tsunami, better educated
males were more likely to survive the tsunami, but education is not predictive of survival among females. Education is not
associated with levels of post-traumatic stress among survivors 1 year after the tsunami, or with the likelihood of being displaced.
Where education does appear to play a role is with respect to coping with the disaster over the longer term. The better educated
were far less likely than others to live in a camp or other temporary housing, moving, instead, to private homes, staying with
family or friends, or renting a new home. The better educated were more able to minimize dips in spending levels following the
tsunami, relative to the cuts made by those with little education. Five years after the tsunami, the better educated were in better
psycho-social health than those with less education. In sum, education is associated with higher levels of resilience over the
longer term.
Key Words: development; disaster; education; resilience; vulnerability

INTRODUCTION policy (Linnerooth-Bayer et al. 2005, Telford and Cosgrave

Disasters are threats to population well-being that derail 2007, Buttenheim 2010, Horton 2011, Padgett and Warnecke
socioeconomic progress, strain social safety nets, and require 2011).
complex assistance and recovery interventions. Over the last
This study uses population-representative longitudinal survey
decade alone, Indonesia, Sri Lanka, Pakistan, China, Haiti,
data collected before and after the 2004 Indian Ocean tsunami
and Japan have all experienced natural disasters with death
to consider the specific question of whether education
tolls in the tens of thousands. The high mortality disasters in
moderates the immediate and longer-term impacts of a
recent years, combined with predictions that these events will
catastrophic natural disaster. Addressing this question with
increase in frequency as a result of global warming and rising
data from a catastrophic disaster collected over a 5-year time
population densities in coastal areas, have increased interest
frame speaks to the broader theoretical line of inquiry
in more fully understanding the factors that underlie
regarding the influence of socioeconomic status on
trajectories of disaster recovery over the longer term.
“vulnerability” and longer-term “resilience” in the aftermath
Several challenges impede building this deeper understanding of a major disaster.
of disaster recovery from the existing research literature. These
It is important to note that, in general, the better educated tend
include the difficulty of studying events of catastrophic
to earn more, have greater wealth, and live longer and healthier
magnitude, the limited size and representativeness of the
lives (Lutz and Samir 2011). They also tend to live in areas
samples and follow-up periods of available data, and
that are less prone to natural disasters and to be better protected
consequently the relative lack of empirical studies focusing
against shocks either through some form of formal or informal
on longer-term outcomes for large representative populations
insurance or through greater diversification of their
(National Research Council (NRC) 2006, Galea and Maxwell
livelihoods as well as their financial assets and social support
2009, Sastry and Vanlandingham 2009). One emerging
(Strauss and Thomas 2008). As a result of these choices, it is
theoretical insight is the conceptualization of disaster impact
difficult to disentangle whether the better educated are better
and recovery in terms of vulnerability and resilience, with the
able to recover over the longer term from a disaster that has
attendant recognition that each is embedded in a context of
equal immediate impacts regardless of education level, or
social processes that may, themselves, contribute to pre-
whether the better educated suffer fewer immediate impacts
existing variation in inequality in multiple dimensions in a
at the outset because of prior investments in risk mitigation
society (NRC 2006, Tierney 2007). This theoretical
and insurance.
perspective complements needs on the empirical side to
identify the population sub-groups who suffer the most The Indian Ocean tsunami, in combination with the data we
devastating and longest-lasting impacts of disaster. These analyze, provides a window into this issue for several key
challenges are recognized as critical for both science and for reasons. First, it is reasonable to treat the tsunami as

1
Duke University, 2SurveyMeter
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

completely unanticipated, at least along the coast of the island of two-storey buildings. Along parts of the west coast of Aceh,
of Sumatra. Second, in the communities that bore the most the water removed bark from trees as high as 13 m (Borrero
intense force of the waves, the tsunami had a devastating 2005). Where rivers emptied into the ocean, the water moved
impact on livelihoods. Third, we draw on an unusual data set inland as much as 6–9 km, flooding plains and arable land. In
that provides information on multiple dimensions of impact other areas, the water reached about 3–4 km inland (Kohl et
and recovery for a population-representative sample first al. 2005, Umitsu et al. 2007).
interviewed 10 months before the tsunami and then for 5 years
From the perspective of contrasting the impact of this
after the event.
catastrophic disaster on the better educated relative to those
The destruction wreaked by the tsunami was massive and far with less education, two important features of the Indian Ocean
reaching. Importantly for this research, we show that in tsunami distinguish it from other natural and manmade
severely damaged areas, the immediate impacts were broadly disasters. First, the tsunami was completely unexpected. The
the same for people across the entire distribution of education. last major tsunami on the coast of mainland Aceh took place
In contrast with many natural disasters, education conferred over 600 years ago (Monecke et al. 2008). Waves reached
little protection from the tsunami’s short-term effects. some parts of coastal Aceh within minutes of the earthquake,
However, over the longer term, our analysis indicates that the and retreating water, a signature of an impending tsunami, was
better educated are substantially more resilient with respect to not interpreted as a sign of danger by the vast majority of the
psychosocial health and economic status. population. Only residents of Simeulue island, where a smaller
tsunami occurred in 1907, systematically relocated to higher
We conclude that the better educated are better placed to
ground, and, correspondingly, the survival rate in Simeulue
mitigate the deleterious consequences and to embrace new
was very high (Gaillard et al.,2008). The unexpectedness of
opportunities in the aftermath of even a major large-scale and
the tsunami contrasts sharply with disasters for which there is
unanticipated disaster. It is not possible to ascribe a causal
some advance warning, such as hurricanes, tornadoes, and
interpretation to this evidence. Those who have invested more
floods. Warnings before Hurricane Katrina, for example,
in education may be more entrepreneurial, nimble, and better
enabled people with more resources—which the better
equipped to take on new opportunities and challenges after a
educated often have—to protect themselves and their
major disaster. The greater resilience of the better educated
livelihoods at least partially from the disaster.
may also arise because of better access to financial resources
or greater availability of social resources after the tsunami. The second important feature of the tsunami for this research
is that the force with which the tsunami waves hit the shore
THE DISASTER varied locally as a function of geophysical factors. The height
At 7:58 a.m. on 26 December 2004, an earthquake measuring and inland reach of water from the tsunami were a complicated
an estimated 9.3 on the Richter scale occurred off the coast of function of both the vertical displacement of the seafloor
Sumatra, Indonesia. Faulting from the earthquake lasted 8 (which varied along the rupture) and features of coastal
minutes, temporarily disrupting the earth’s rotation and topography, such as the slope of the coastal zone, elevation of
generating a 1,200-km rupture along the floor of the Indian the beachfront, and the direction of the wave relative to the
Ocean (Bunting et al. 2007). The vertical displacement from land (Ramakrishnan et al. 2005). Accordingly, a component
the quake was 5–15 m, which generated huge tsunami surges of the intensity of the tsunami’s impact is random and is
that ultimately reached the shores of all countries that rim the unrelated to the education of those living in the area. It is,
Indian Ocean (Kerr 2005, Lay et al. 2005, Marris 2005, therefore, reasonable to treat the tsunami as a large and
Sinadinovski 2006). unanticipated natural disaster that is unlikely a priori to have
The first of the waves slammed into the island of Sumatra spared the better educated.
within 15 minutes of the earthquake. In Aceh, the Indonesian
DATA AND MEASUREMENT
province closest to the rupture, the tsunami engulfed
Research on the impact of disasters has been limited by a dearth
communities along 800 km of coastline. Studies estimate that
of population-representative data that follow samples of
the tsunami killed 130,000 individuals, with another 30,000
sufficient size before and after the disaster. Constructing
classified as missing (Rofi et al. 2006, Doocy et al. 2007).
population-representative samples after an event that displaces
Some 700,000 individuals were displaced, and damage to
a large fraction of the population is extremely difficult, and
property and infrastructure was valued at 4.5 billion (The
few studies have access to information on populations before
Consultative Group on Indonesia 2005).
a major disaster strikes. The studies that do are rarely well
In areas severely damaged by the tsunami, the water swept positioned to locate and interview the individuals who move
away everything in its path including roads, bridges, and from place to place in the disaster’s aftermath (Buttenheim
buildings. At the beachfront in Banda Aceh, water depths were 2010, Gray et al. 2011, Horton 2011).
approximately 9 m and even further inland reached the height
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

Data imagery and prepared by the United States Geological Survey,

We draw on longitudinal data that we designed and collected the United States Agency for International Development, the
as part of the Study of the Tsunami Aftermath and Recovery Dartmouth Flood Observatory, and the German Aerospace
(STAR). The study covers individuals who were living in Center (Gillespie et al. 2007). Additionally, in each
coastal areas of the Indonesian provinces of Aceh and North community, we conducted interviews with local leaders, who
Sumatra before the tsunami. Building on the foundation provided their own assessments of the extent of destruction to
provided by a pre-earthquake baseline survey that interviewed the built and natural environment due to the tsunami and
nearly 27,000 individuals, we conducted follow-up surveys earthquake, and our survey supervisors completed a
annually for 5 years after the disaster. questionnaire that detailed damage due to the tsunami and
earthquake based on their own direct observation.
The baseline survey participants were part of a socioeconomic
survey, SUSENAS, conducted by Statistics Indonesia in We used these sources of information to construct a categorical
February 2004, 10 months before the Sumatran-Andaman indicator of damage to the enumeration area. This indicator is
earthquake. Statistics Indonesia has conducted the SUSENAS a strong and significant predictor of many tsunami-related
annually throughout Indonesia since 1963. The survey, which outcomes derived from the household data including
is widely used in the international scientific and policy mortality, injuries, post-traumatic stress disorders, and extent
communities, is regarded as being of very high quality. It is of damage to houses and land (Frankenberg et al. 2008). By
designed to represent the population at the “kabupaten” this indicator, 95 of the 410 STAR enumeration areas are
(district) level. The baseline for STAR consists of households classified as severely damaged. We conducted the analyses
located in 13 districts along the coast of Aceh and North for this paper based on data from respondents who were living
Sumatra when they were interviewed in the 2004 (pre- in the severely damaged areas at the time of the pre-earthquake
earthquake) wave of SUSENAS. We selected these 13 districts baseline.
because they were geographically positioned so that their
Before the earthquake, when interviewed as part of the 2004
coastlines were at risk of inundation from the tsunami waves
baseline survey, one respondent in each household reported
although not all parts of the coast were, in fact, inundated. This
socioeconomic and demographic characteristics for
provides communities that were directly impacted by the
themselves and all other household members. The first follow-
tsunami and coastal communities that were not directly
up survey, STAR1, took place between May 2005 and July
affected. Within these 13 districts, all SUSENAS enumeration
2006, in collaboration with Statistics Indonesia and with the
areas were included in STAR, for a total of 410 enumeration
assistance of their field supervisors. In STAR1, we collected
areas (EAs) in 369 villages. Although all of the districts
both individual and household-level data, drawing on and
included in STAR had a potentially vulnerable coastline, the
augmenting the baseline questionnaire. Every adult member
extent to which the tsunami inundated the 410 enumeration
of every household was eligible to be interviewed, and
areas varied considerably as a function of position relative to
information about every child was collected from a parent or
the earthquake’s epicenter, shape of the coastline, distance
primary caretaker. Every member of the baseline household
from the ocean, elevation of the land, and the presence of rivers
survey who survived the tsunami was eligible to be tracked
or canals flowing into the ocean.
and interviewed in their new location. In addition, village
To characterize the tsunami’s destructive effect on each leaders and informants at local schools and health facilities
enumeration area, we developed a classification method that provided information as part of a large community-level
combines information from remote-sensed satellite imagery, survey.
reports from community informants, and observations of
STAR1 was the first of five annual post-tsunami surveys.
survey team supervisors. We use several biophysical measures
These data are combined with the subsequentr waves, STAR2
derived from satellite images, which were linked to the exact
—STAR5, the last of which took place between September
location of each EA using global positioning system (GPS)
2009 and December 2010.
measurements made during the follow-up survey. We
constructed one of these measures by comparing satellite In this paper, we focus on 3,812 individuals who were between
imagery from the National Aeronautics and Space 20 and 59 years old at the time of the baseline survey and were
Administration’s Moderate Resolution Imaging Spectroradiometer living in the 95 enumeration areas classified as heavily
(MODIS) for 17 December 2004, a week before the tsunami, damaged. We put enormous effort into identifying the baseline
to imagery for 29 December 2004, 3 days after the tsunami. respondents who had died in the tsunami, which involved
The proportion of land cover that the tsunami changed to bare finding surviving members of baseline households, following
earth (through scouring or sediment deposition) was manually up with neighbors and community leaders in the devastated
assessed for a 0.6-km2 area centered over each GPS point. areas, visiting camps and barracks, and consulting registers
These estimates of damage were cross-validated with kept in each village of those who died or were missing (see
estimates of damaged areas derived from remotely sensed Frankenberg et al. 2011 for more details). Of all the deaths
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

recorded in STAR and attributed to the tsunami, 92% occurred sequelae of exposure to disasters and because higher levels of
in the severely damaged enumeration areas on which we focus socioeconomic status, as measured by education and other
here. indicators, have been shown to be protective in previous
studies (Armenian et al. 2000; see also Norris et al. 2002 for
High re-interview rates are critical to the success of
a review). The stress measures are constructed from
longitudinal surveys. We developed and implemented
information on post-traumatic stress reactions, which were
extensive tracking protocols to find not only those who
assessed using seven items from the 17-item Post-Traumatic
continued to live in their baseline locations but also those who
Stress Disorder (PTSD) Checklist Civilian Version (PCL)
moved. This is particularly important in the context of a
(Weathers et al. 1993). The instrument has been validated
disaster that causes massive disruption and displacement—as
among veterans, those exposed to disasters, violence,
is the case with the tsunami.
accidents, and sexual assault, and survivors of bone-marrow
Of the age-eligible respondents who were, at baseline, living transplants and has been used in both advanced and developing
in the severely damaged areas, 28% were dead as of the first countries (Blanchard et al. 1996, Smith et al. 1999). Adult
follow-up survey. Of the remaining respondents, 87% were respondents in the post-tsunami surveys were asked about
members of interviewed households in the first follow-up. specific symptom items that, in combination, covered three
Persistent attempts to track all survivors in subsequent waves distinctive psychological domains of post-traumatic stress.
paid off: over 95% have been interviewed at least once and These data were used to construct PTSR scales developed in
90% were assessed in the final interview. Frankenberg et al. (2008), which range from 0 to 21. Higher
values reflect higher PTSR, and thus poorer psychosocial
Measures health. Psychosocial counseling has been shown to mitigate
In this section, we describe the measures that we use to indicate the effects of disasters on psychosocial well-being, and efforts
vulnerability to the immediate and shorter term impacts of the were made to strengthen mental health services in Aceh in the
tsunami, as well measures that capture longer term outcomes years following the disaster (Prasetiyawan et al. 2006). We
and can be interpreted as indicators of resilience. assess who obtained counseling.
As the water came ashore, it swept up many people, killing Our final set of indicators of well-being provides a summary
some, exposing others to traumatic experiences, and damaging of the economic status of each respondent and their families.
or destroying most of the homes in its path. Measures of these Income is often used as a measure of economic well-being in
outcomes serve as indicators of the disaster’s immediate socioeconomic studies, but it is complicated to interpret,
impact. We begin by examining mortality. Among those who particularly after the tsunami destroyed farmland and
survived, we also analyze experiences during the tsunami, businesses, resulting in a substantial decline in employment
including hearing or seeing the water come ashore, being and earnings. Moreover, a large fraction of the population—
caught in the water or injured by it, seeing others struggling especially women—did not earn income before the tsunami.
in the water, or having one’s house damaged or destroyed as
a result of the disaster. For these reasons, we examine economic resources at the
household rather than individual level. Because a large aid
In the months that followed the tsunami, residents of the effort was mounted after the tsunami, it is important that the
communities that were heavily damaged struggled to cope measure of economic well-being include assistance from the
with the magnitude of the event, and assistance began to arrive. public sector as well as from friends and family. A measure
To capture these dynamics, we focus on two dimensions: the that meets these criteria is household consumption, which
built environment and psychosocial resources. includes the imputed value of goods produced at home and
With respect to the built environment, damage to housing and goods and services provided in kind, during the month before
infrastructure from the tsunami resulted in massive population the survey. The value (market and imputed) of consumption
displacement. We analyze whether respondents were is more likely to reflect economic well-being than income as
displaced from their original residence and, if they were, it incorporates not only goods and services provided by family,
whether they lived in temporary housing (defined as living in friends, and the public sector but also consumption from
a camp or temporary settlement, a tent, or barracks) at some drawing on savings or selling assets.
point during the 2 years after the tsunami. We also investigate Another advantage of household consumption is that it is
whether they received assistance from the government or measured in every wave of the survey, including the baseline.
international agencies to build or repair housing. Accordingly, we can trace the evolution of expenditure, and
With respect to psychosocial resources, we develop indicators its relationship to education, before and after the tsunami.
of post-traumatic stress reactivity (PTSR) and receipt of Generally, we expect households to attempt to mitigate the
mental health counseling. We believe it appropriate to focus impact of a large negative shock on their well-being by keeping
on PTSR because it is one of the most common psychological reductions in consumption to a minimum—that is, smoothing
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

consumption over time. If the better educated have more activity, and help ensure that individual-level measures of
resources that they can call on—from savings, family education are not simply proxying for community-level
members, or from other sources—then they are likely to reduce variations in resources before the disaster or degree of
consumption less than those with fewer resources during times destruction during the disaster.
when resources are relatively scarce. We will provide direct
For each individual, i, at time t, the model
evidence on this question.
θit = α + β Eit + γ Xit + µc + εict [1]
In an effort to take into account variation in consumption with
household composition, we adjust expenditure for the number is estimated by ordinary least squares. Unobserved
of household members and examine household per capita heterogeneity is captured by εict. Estimates of variance–
expenditure (PCE). To evaluate the importance of this covariance matrices and all test statistics take into account
approximation, we explore changes in household clustering at the enumeration area level and are robust to
composition. Per capita expenditure is specified in a arbitrary forms of heteroskedasticity (Huber 1981).
logarithmic form because its distribution has a very long right
Education is measured by the highest grade attained. We
tail, and it is well approximated by the log normal distribution.
interpret grade attained as a crude summary of the respondent’s
We also examine the share of the budget allocated to food. school-related level of human capital, but fully recognize that
Higher food shares have been interpreted as indicative of lower human capital is a far broader concept that reflects a wide array
levels of well-being since at least Engel (1895), and the food of skills, personality traits, health, and cognition. Because this
share is the foundation of many measures of poverty used research focuses on the relationship between education and
across the globe. Food shares provide an alternative indicator each of the outcomes described above, we have carefully
of economic well-being that complements the logarithm of explored the shape of this relationship. Two model
PCE and, to some extent, takes into account changes in prices. specifications are reported in the tables. The first model is
linear in education and the second is piecewise linear with a
EMPIRICAL METHODS knot at completion of grade 6. (Experiments with knots at other
The goal of this research is to provide new insights into the points and models that include indicator variables for
moderating effect of education in the face of a major shock to education do not yield additional insights, and so we report
population health and well-being as a result of a large and these specifications.)
unanticipated disaster. For each outcome, θ, described in the
previous section, we examine its association with education, Table 1 reports the distribution of education of respondents in
E, after adjusting for age, X, in a multivariate regression the baseline survey, conducted before the tsunami, and
framework. We control for age because, in the study sample, includes both those who survived the tsunami and those who
levels of education are significantly higher among younger did not. As shown in Table 1, among all respondents who were
cohorts. In all analyses based on individual-level data, models living, at baseline, in communities that were severely damaged
are stratified by sex. All models also control location of by the tsunami (the study sample used in the regression
residence at baseline in a flexible way. analyses reported below), the average respondent completed
9.4 grades of school. This is equivalent to finishing junior
The impacts of some disasters accrue disproportionately in secondary school. In contrast, the average respondent in the
communities whose locations on marginal land make them entire STAR sample of baseline respondents—which includes
relatively vulnerable and whose residents tend to be poor and areas that were not severely damaged—has completed 8.3
perhaps poorly educated. This was not generally the case with grades of school. This difference underscores the fact that,
respect to the tsunami, which affected wealthy communities unlike many natural disasters, education levels were on
of business owners, and public servants located in cites along average higher in the areas that were more likely to be severely
the coast as well as relatively poorer communities of fishermen damaged by the tsunami.
and farmers, but left the more remote inland communities
untouched. However, rather than rely on heterogeneity in The distribution of education of male and female adults who
education levels across communities that were all badly were living in severely damaged areas is displayed in panel B
damaged, we draw contrasts among people who were living of Table 1. Very few adults in the sample had no schooling.
in the same community. Formally, all of the regression models Among the one-third who had some primary education, the
include community (enumeration area) fixed effects, µc, which vast majority completed primary school (six grades). About
absorb the influence of all community-specific variation that one in five attended junior high school (grades 7–9), about
does not change over time and affects the outcome, θ, in a one-quarter attended senior high (grades 10–12), and slightly
linear and additive way. These fixed effects capture the extent over one in six attended some college. Males are significantly
of damage in the community because of the earthquake and better educated than females, with 49% of males in the sample
tsunami, as well as prior levels of infrastructure and economic attending senior high school or more, vs. only 42% of females.
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

Table 1. Distribution of age and education

All sample Females Males

A. Highest completed grade
1. All areas 8.3 8.0 8.7
[0.03] [0.05] [0.05]
2. Severely damaged areas 9.4 9.2 9.7
[0.07] [0.10] [0.09]
B. Grade completed (% sample)
(among respondents in severely damaged areas)
No school 1.8 2.6 1.0
Some primary school (Grade 1–5) 8.6 9.0 8.1
Completed primary school (Grade 6) 23.6 26.2 21.0
Junior secondary school (Grades 7–9) 20.5 19.9 21.1
Senior secondary school (Grades 10–12) 27.8 24.2 31.5
College (>Grade 12) 17.6 18.1 17.1
Age (% sample)
20–29 35.9 38.0 33.7
30–39 30.2 29.4 31.1
40–49 20.8 20.4 21.3
50–59 13.0 12.1 13.9
Sample size 3415 1732 1683
Notes: [Standard errors].

The table also displays the age distribution of baseline fixed effects. Panel B1 reports estimates from a model that is
respondents, with about one-third in their twenties, one-third linear in education. Panel B2 reports estimates from a spline
in their thirties, and the rest in their forties and fifties. function that allows the shape to be piece-wise linear in
education with a knot at completed grade 6 (or completed
RESULTS primary school). Sample sizes are in panel C.
This section presents results from estimating the regression
model described above, which provides a summary of how For females, education does not confer a survival advantage.
educational attainment is related to outcomes that represent Better-educated females are no more likely to survive the
both vulnerability to the tsunami’s immediate and short-term tsunami than females with little education (column 1). For
impacts and longer-term resilience in the disaster’s aftermath. males, some evidence suggests that education is protective, at
Attention is restricted to respondents who were, at the time of least among those who advanced beyond primary school
the pre-tsunami baseline survey, living in enumeration areas (column 2). Males who completed senior high school are about
that were subsequently severely damaged by the tsunami. The 6.5 percentage points more likely to survive the tsunami than
samples include those who stayed in the areas and those who those who left school after completing primary education.
moved away so that they are representative of the population However, those who completed primary school are themselves
exposed to the full brunt of the tsunami. about 8 percentage points more likely to die in the tsunami
than those who never attended school.
Table 2 presents the results of estimating the model for our
measures of the tsunami’s immediate impact: mortality, One interpretation of these results is that, in part, the mortality
exposure to traumatic experiences, and damage or destruction differences by education of males reflect differences in height,
to housing. In the table (and in subsequent tables for other strength, and possibly other dimensions of human capital. This
outcomes), panel A displays the average for the outcome. interpretation is consistent with the evidence that mortality
Mortality in the heavily damaged areas was extremely high: rates are higher among females than among males. As shown
30.2% of females and 19.1% of males died in the tsunami. in Frankenberg et al. (2011), older males and females were
Close to the coast, mortality was even higher. In communities also more likely to die in the tsunami relative to prime age
within a kilometer of the shoreline, 54.5% of females and males and females, respectively.
33.3% of males died. To explore the idea that education may be an important proxy
Panel B displays estimates of the shape of the relationship for strength, we re-estimate the models for respondents who
between education and each outcome, controlling age and EA were living within 1 km of the coast, where mortality was
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

Table 2. Mortality and exposure at time of the tsunami

I. Mortality at time of tsunami II. Exposure at time of tsunami III. Damaged/

Destroyed Housing
All respondents ≤1 km from coast Females Males Females Males
Females Males Females Males Injured or See others Hear or see Injured or See others Hear or see
caught in the struggle the water caught in the struggle the water
water water
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
A. Mean
30.2 19.1 54.5 33.3 25.9 35.5 76.8 34.6 44.6 80.8 63.47 76.39
[3.62] [2.41] [8.33] [7.15] [2.75] [3.30] [3.52] [2.77] [3.03] [3.03] [3.87] [4.00]
B. Education (highest grade)
B.1 Linear
-0.05 -0.59 -0.38 -1.58 -0.79 -0.71 -0.02 -0.10 -0.45 -0.24 0.02 -0.62
[0.17] [0.22] [0.44] [0.63] [0.39] [0.45] [0.32] [0.44] [0.44] [0.31] [0.28] [0.37]
B.2 Spline (0–6)
-0.42 1.16 -1.14 0.17 -0.23 -0.73 -0.22 0.95 -0.40 0.29 0.45 -1.04
[0.44] [0.65] [1.29] [2.45] [0.98] [0.87] [0.61] [0.97] [1.16] [0.81] [0.70] [0.63]
Spline (>7)
0.09 -1.07 -0.09 -2.07 -0.96 -0.70 0.04 -0.35 -0.46 -0.37 -0.14 -0.46
[0.28] [0.29] [0.81] [0.84] [0.48] [0.56] [0.37] [0.54] [0.54] [0.40] [0.36] [0.52]
C. Sample size
1920 1889 314 318 1036 1036 1036 1166 1166 1166 1036 1166
Notes [Standard errors] take into account clustering at the enumeration area level and are robust to heteroskedasticity. Includes controls for respondent age
and enumeration area fixed effects.

highest. Results are reported in columns 3 and 4. Close to the All subsequent regression analyses focus on those who
coast, education is unrelated to survival for females but is even survived the tsunami. Many of the survivors experienced
more strongly predictive of survival among males, suggesting harrowing events as the water came ashore. We next turn to
that strength is not the full explanation for the link between the question of how education is related to these experiences,
education and survival. as reported by survivors who were living at baseline in areas
that were subsequently severely damaged by the tsunami
We cannot test directly whether education is a proxy for height
(these results are presented in columns 5 through 10 of Table
and strength because neither height nor strength was measured
2). We begin with females. Overall, a quarter of women were
at baseline and, therefore, is not known for those who died in
caught up in the water or injured, just over one-third witnessed
the tsunami. However, using data from areas that were not
others struggling in the water, and about three-quarters heard
damaged by the tsunami, where mortality is very low, we can
or saw the water come ashore. Better-educated women were
estimate the association between height and education using
less likely to report being caught in the water, injured, or
data from the post-tsunami resurveys. On average, a male who
watching others struggle. For example, a female who
completed senior high is 163 cm tall; a male who only
completed senior high school is estimated to be between 4 and
completed primary school is 160 cm tall and a male who did
5 percentage points less likely to have been caught up in or
not complete primary school is 142 cm tall. Controlling age,
injured by the water relative to a female who completed
in a piece-wise linear specification, each year of education is
primary school, and this difference is statistically significant.
associated with a 4.7 cm (standard error = 0.3 cm) increase in
Education is unrelated to hearing or seeing the tsunami come
height until completing primary school and then an increase
ashore among females.
of 0.3 cm (standard error = 0.08 cm) for each additional grade
thereafter. Because the association between education and Among male survivors about one-third report being caught up
mortality is positive for males up to completion of primary in the water or injured, and nearly 45% saw others struggling
school and is then negative and significant only for males with in the water. These rates are higher than for females because
more than primary schooling (whereas the association females were less likely to have lived to report these
between education and height is positive throughout the experiences than were males. For males, conditional on
education distribution), it seems unlikely that education serves surviving, exposure to the tsunami is unrelated to education.
only as a proxy for height and strength in these regression
The results presented thus far relate to the vulnerability of
models. The education–height association is not strong enough
individuals along the health-related dimensions of mortality
nor the appropriate shape to fully explain the relationship
and exposure to traumatic experiences. The physical
between survival and education of males.
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

Table 3. Displacement and receipt of housing assistance

Females Males
Displaced by Lived in temporary Received housing Displaced by Lived in temporary Received housing
tsunami housing assistance tsunami housing assistance
[1] [2] [3] [4] [5] [6]
A. Mean 66.25 46.33 51.25 66.47 52.06 55.75
[5.45] [4.37] [4.43] [4.13] [3.95] [4.10]
B. Education (highest grade)
B.1 Linear 0.17 -1.31 -0.78 -0.39 -1.39 -0.84
[0.36] [0.44] [0.37] [0.41] [0.37] [0.39]
B.2 Spline (0–6) 1.22 -1.64 -1.73 -0.73 -0.47 -0.57
[0.92] [0.86] [1.03] [0.73] [0.88] [0.83]
Spline (>7) -0.11 -1.21 -0.50 -0.30 -1.61 -0.90
[0.42] [0.46] [0.48] [0.52] [0.41] [0.49]
C. Sample size 1036 1036 1036 1166 1166 1166
Notes [Standard errors] take into account clustering at the enumeration area level and are robust to heteroskedasticity. Controls included for respondent age
and enumeration area fixed effects..

vulnerability of respondents’ homes is another dimension with A key dimension of the recovery and reconstruction effort was
strong implications for the well-being of survivors. The last the provision of assistance with building or repairing houses
two columns present results for experiencing damage or that were damaged or destroyed by the tsunami. About half of
destruction to one’s home. Overall, two-thirds of women and the tsunami survivors received housing assistance from the
three-quarters of men report that their home was damaged or government or a nongovernmental organization (NGO)—a
destroyed by the tsunami—again reflecting differential fraction that is substantially less than the fraction whose houses
survival rates of men and women. But for neither men nor were damaged. The better educated were just as likely to have
women are damage or destruction of housing related to level their home damaged in the tsunami, but they were less likely
of education. to receive housing assistance. Although the decline in the
probability of receiving aid as education increases is
We turn next to outcomes that represent respondents’
statistically significant, the rate of decline is small, and
experiences as the aftermath of the disaster unfolded. We
differences in the probability of receiving assistance for the
explore these along the dimensions discussed above: housing
best educated relative to those with little education are modest.
and psychosocial health. Table 3 focuses on housing in the
disaster’s aftermath. Whereas the models are reported Another form of assistance, but one that relates to psychosocial
separately for males and females, results do not differ by sex, dimensions of the disaster, is receipt of mental health
and so we do not distinguish between males and females in counseling. As shown in the first two columns of Table 4,
the discussion. regardless of sex, around one in six survivors received some
form of counseling after the tsunami. Although the
We first consider displacement, which we define as moving
relationship between education and receipt of counseling is
within the first 4 months following the tsunami. About two-
positive, the association is only statistically significant for
thirds of survivors were displaced by the tsunami, and being
males who completed more than primary schooling.
displaced is not associated with education.
Columns 3 through 6 of Table 4 report levels of PTSR. At the
About half the survivors who were living in areas that were
time of the first interview after the tsunami, levels of PTSR
severely damaged lived in temporary housing—typically a
are higher among females (6.60 on a scale of 21) than among
camp—at some point during the 2 years after the tsunami
males (5.89). For neither sex are there differences across the
(columns 2 and 5). The rest of the displaced moved to private
education distribution. At the time of the most recent
homes—either moving in with family or friends or renting a
interview, approximately 5 years later, PTSR levels have
private home. The better educated, particularly those who
declined substantially, although they remain higher for
completed more than primary school, were significantly less
females relative to males (3.63 vs. 2.73 on the same 21-point
likely to move to temporary housing, but this is not because
scale). In addition, PTSR is significantly lower among the
the better educated were less likely to be displaced.
better educated, indicating that those with more education are
Displacement risks are not related to education. Rather, the
more resilient in terms of psychosocial well-being. For
education advantage with respect to avoiding temporary
example, the difference in PTSR of a male who completed
shelter likely reflects the greater availability of financial and/
senior high school relative to a male who did not attend school
or social resources of those who are better educated.
is about half the average level of PTSR for all males at the
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

Table 4. Mental health counseling and post-traumatic stress reactivity

I. Received mental health counselling II. Post-Traumatic Stress Reactivity

Females Males Females Males
At any time during At any time during First interview afer Last First interview after Last
study period study period tsunami interview tsunami interview
[1] [2] [3] [4] [5] [6]
A. Mean 14.58 13.38 6.60 3.63 5.89 2.73
[1.63] [1.42] [0.16] [0.21] [0.17] [0.16]
B. Education (highest grade)
B.1 Linear 0.34 0.50 0.06 -0.08 -0.03 -0.11
[0.30] [0.30] [0.05] [0.03] [0.04] [0.03]
B.2 Spline (0–6) 0.73 -0.50 -0.02 0.05 0.02 -0.19
[0.74] [0.95] [0.12] [0.10] [0.10] [0.09]
Spline (>7) 0.23 0.74 0.08 -0.11 -0.04 -0.09
[0.38] [0.33] [0.06] [0.04] [0.05] [0.04]
C. Sample size 1036 1166 860 860 829 829
Notes [Standard errors] take into account clustering at the enumeration area level and are robust to heteroskedasticity. Controls included for respondent
age and enumeration area fixed effects.

final interview; for females, a comparable comparison educated were substantially better protected from the declines
accounts for about one-quarter of the average PTSR level. in PCE—and economic resource availability—that occurred
in the aftermath of the tsunami. Thus, the better educated were
In Table 5, we turn to indicators of the characteristics of
better able to smooth consumption after this large shock, and
households that respondents are in at each wave of the study.
inequality across the education distribution rose. As years
These characteristics include the logarithm of PCE, lnPCE,
since the tsunami passed, the gap in spending between the
the share of spending that goes to food, household size, and
better and less educated remained larger than it was before the
the share of household members who are under 15. Because
tsunami, but the difference is significant only during the first
these indicators are measured at the household level, models
2 years after the tsunami.
are not estimated separately for males and females. The
relationship between lnPCE and education of the respondent The piece-wise linear model, in panel 1C, establishes that
is estimated separately for every wave of the study, which consumption smoothing is more effective only among those
provides evidence on evolution in the trajectory of lnPCE who have completed primary education. In fact, for those who
across the education distribution as time passes. did not complete primary school, education and PCE are not
related.
Panel 1A of the table reports the association between lnPCE
and education for each wave. Specifying expenditure in The interpretation of variation in the relationship between PCE
logarithms means that the coefficient estimates can be and education is not entirely straightforward. First,
interpreted as representing percentage changes. By estimating immediately after the tsunami, prices rose substantially for
the models separately for each year after the tsunami, and by many goods and, as a result, relative prices also changed.
including enumeration area fixed effects, the models absorb Changes in overall prices are captured by the intercept. The
the impact of changes in prices over time and at the local area data to reliably compute such indices do not exist.
level. This is important in the context of the tsunami—and
If prices of the consumption bundles consumed by the better
most natural disasters—when supply chains are severely
educated rose more than those consumed by the less educated,
disrupted and shortages of food, housing, and transport are
then real resources of the better educated will be lower than
accompanied by high rates of price inflation that vary across
those of the less educated, and this could explain the apparently
time and space.
greater consumption smoothing by the better educated. This
In the pre-tsunami wave, the better educated spend more: per is unlikely to be the case, as prices of food, housing, and
capita expenditure is 2.16% higher for each year of completed transport rose the most, and these goods tend to account for a
education. This is a reflection of the fact that education and larger share of the budget of poorer households. Rather, our
economic success are positively correlated. In the year after estimates of the differences in consumption smoothing are
the tsunami, each year of education was associated with a likely to be lower bounds because prices for the goods the least
4.18% increase in PCE. The difference between the post- and educated spend most of their money on are the prices that
pre-tsunami association, 2.02%, is displayed in panel 1B. That likely rose the most.
difference is statistically significant, indicating that the better
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

Table 5. Economic resources and household composition

Survey wave Pre- Post-tsunami survey wave (years after tsunami)

tsunami 1 2 3 4 5
[1] [2] [3] [4] [5] [6]
1. lnPCE
1A. Education 2.16 4.18 3.51 3.07 2.90 2.74
[0.30] [0.67] [0.47] [0.50] [0.48] [0.44]
1B. Difference (relative to pre-tsunami relationship) 2.02 1.34 0.91 0.73 0.57
[0.73] [0.55] [0.58] [0.56] [0.53]
1C. Spline (0–6) -0.01 1.10 0.44 1.44 0.47 1.46
[0.66] [1.59] [1.54] [1.01] [0.95] [1.06]
Spline (>7) 2.74 4.96 4.30 3.50 3.52 3.07
[0.34] [0.87] [0.54] [0.64] [0.57] [0.56]
2. Food share
Spline (0–6) 0.06 -0.13 -0.17 -0.38 -0.34 0.01
[0.18] [0.33] [0.36] [0.27] [0.28] [0.27]
Spline (>7) -0.45 -1.34 -1.03 -0.87 -0.77 -0.73
[0.08] [0.20] [0.17] [0.13] [0.16] [0.14]
3. Household size
Spline (0–6) 0.07 0.09 0.05 0.06 0.03 0.05
[0.04] [0.04] [0.04] [0.03] [0.03] [0.04]
Spline (>6) 0.010 -0.010 -0.020 -0.02 -0.03 -0.02
[0.02] [0.02] [0.02] [0.02] [0.02] [0.02]
4. Share of HH members age <15
Spline (0–6) 0.46 0.88 0.71 0.60 0.51 0.74
[0.34] [0.41] [0.34] [0.35] [0.37] [0.37]
Spline (>6) 0.07 -0.01 -0.18 -0.16 0.06 0.05
[0.16] [0.17] [0.16] [0.17] [0.16] [0.17]
5. Sample size 3413 2692 2641 2627 2636 2732
Notes [Standard errors] take into account clustering at the enumeration area level and are robust to heteroskedasticity. All models include enumeration area
fixed effects.

Additional evidence suggests this is the case. As shown in by household size. Rather than rely on some other ad hoc
Panel 2 of Table 5, and paralleling results for lnPCE, food adjustment, we investigate whether there were changes in
shares are unrelated to education for those with primary household size and composition following the tsunami that
schooling or less and then decline with education for higher are systematically related to education.
levels. The rate of decline is substantially and significantly
Panel 3 of Table 5 reports results for household size and Panel
greater in the year immediately after the tsunami, rising from
4 reports results for the share of household members who are
0.5% per grade of schooling before the tsunami to 1.3% after
age under 15. For both cases, there is evidence that household
the tsunami. Food shares rose the most for the least educated,
size and the share age under 15 rose with education
whereas the better educated were more able to protect their
immediately after the tsunami—but only among those who
budget allocations to other goods. The estimated rate of decline
had no more than primary schooling—the group of people for
of foods shares as education increases becomes less steep as
whom lnPCE and food shares are not related to education.
time since the tsunami rises, but it remains significantly
Among the better educated, there is no evidence that household
different from the pre-tsunami rate for 3 years after the tsunami
size or the share age under 15 is related to education or that
—longer than is the case for lnPCE.
these associations changed over time. We conclude that
The second reason that interpretation of changes in PCE is variation in household size and composition is not likely to
complicated is that household size and composition may have substantially contaminate our interpretation of changes in
changed over time. For the purposes of interpreting the economic well-being based on variation in lnPCE and food
evidence on the lnPCE-education profile, these changes need shares. That evidence clearly indicates that, relative to those
to differ across the distribution of education. A long literature with little education, the better educated were more successful
focuses on the development of equivalence scales that seek to in smoothing consumption—or mitigating the deleterious
enable direct comparisons of well-being across different impact of the tsunami on spending—immediately after the
household structures. No consensus has been reached on a tsunami whereas those with less education took longer to
theoretically and empirically valid approach to this complex adjust.
problem and so, faute de mieux, we have adjusted expenditure
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

DISCUSSION spending after the disaster. But, over time, the difference by
This research has traced out the tsunami’s immediate impacts education level diminishes. A similar pattern is documented
on health and well-being by comparing the markers before and for the share of budget spent on food, an indicator used as a
after the tsunami for the same people. We have also traced out proxy for economic well-being.
the evolution of the same markers during the 5 years after the
Thus, although the tsunami took a huge immediate toll on
tsunami. Throughout, we have given special prominence to
individuals at all levels of education, the evidence suggests
variation in these trajectories across the distribution of
that the better educated were more effective at adjusting to the
education and, thereby, have provided new insights into the
changed reality of their lives relative to those with less
extent to which heterogeneity in resilience and recovery in the
education. In part, this likely reflects the resources they had
aftermath of a major disaster is related to education.
before the tsunami as well as their experiences in the months
With respect to its immediate impact, we consider the tsunami after the tsunami, when they were able to afford to move to
as a threat to life and property. Education, particularly above private homes rather than live in camps. Moreover, although
the primary level, provides some protection against death for the destruction of livelihoods resulted in reductions in the
men, but not for women. Conditional on survival, education economic resources of all households, the better educated were
provides some protection from being caught in the water or more able to mitigate declines in consumption levels relative
injured for women, but not for men. If one assumes that those to the cuts in spending made by those with less education.
who perished in the tsunami were caught in the water, then Finally, 5 years after the tsunami, the better educated were in
rates of exposure to the tsunami’s greatest threats to safety better psychosocial health than those with less education,
were similar for males and females, and decline with education indicating a more rapid recovery.
for both. However, on the dimension of physical destruction
The faster recovery of the better educated in the face of a major
of perhaps greatest salience to our respondents—the loss of a
natural disaster does not appear to be because they thought an
home—the better educated are no less likely to suffer than
earthquake or tsunami was more likely than those with less
anyone else.
education. In the baseline survey conducted before the
Thus, in terms of the tsunami’s immediate impacts, the tsunami, a randomly selected subsample of 15% of the
protective effect of education was limited. But does education respondents were asked whether they thought they were living
distinguish decisions and outcomes in the aftermath of the in a location that was at high risk of a natural disaster. In all
tsunami? of Aceh and North Sumatra, 9% of the respondents answered
this question in the affirmative, and 9% of those respondents
The role of education varies by outcome. For those with more
said that the greatest risk was from an earthquake or tsunami.
than a primary school education, increases in education were
In the areas that were severely damaged by the tsunami, 12%
associated with a reduced likelihood of living in temporary
of the respondents said they lived in a place that was at risk
housing in the form of camps, tents, or barracks. This is not
of a natural disaster, and 13% of them indicated the greatest
because of greater access to housing assistance from official
risk was from an earthquake or tsunami. Importantly, for this
sources for the better educated. In fact, education is associated
research, responses to these questions are not associated with
with a significant reduction in the likelihood of receiving such
the education levels of the respondents.
housing assistance, although the size of the effect is small.
Instead, the protective effects of education are likely a
Turning to psychosocial dimensions of well-being, in the
reflection of greater accumulated financial resources and
period shortly after the tsunami, levels of PTSR did not differ
possibly social resources available to the better educated in
across the education distribution. Over time, however,
times of need. It is also possible that those who have invested
declines in PTSR proceed more rapidly for the better educated.
more in education make better choices in times of adversity,
As with housing, this does not reflect better access to mental
are more entrepreneurial, and are more effective at taking on
health services among the better educated. Receipt of any
new challenges. Although the results presented here are
counseling was rare.
important for the design of policies that seek to mitigate the
Because data on levels and patterns of household spending are impact of large-scale disasters, understanding the pathways
available before the tsunami, as well as annually for 5 years through which the better educated were more able to weather
afterward, we investigate the evolution of spending before and the storm of the Indian Ocean tsunami remains an important
after the tsunami, and how those trajectories differ by and pressing question for scientific inquiry.
education level. In the year after the tsunami, absolute levels
of spending decline for everyone, but the size of the decline
shrinks as education rises. In other words, after the tsunami, Responses to this article can be read online at:
the difference in spending levels by education increases, http://www.ecologyandsociety.org/issues/responses.
suggesting that the better educated were better able to protect php/5377
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

2008. Ethnic groups’ response to the 26 December 2004

Acknowledgments: earthquake and tsunami in Aceh, Indonesia. Natural Hazards
47(1):17–38.
This work has been supported by grants from the World Bank,
the MacArthur Foundation (05-85158-000), the National Galea, S., and A. Maxwell. 2009. Methodological challenges
Institute for Child Health and Human Development in studying the mental health consequences of disasters. Pages
(HD052762, HD051970), the National Institutes on Aging 579–593 in Y. Neria, S. Galea, and F.H. Norris, editors. Mental
(AG031266), and the National Science Foundation Health and Disasters. Cambridge University Press, New
(CMS-0527763). All opinions and errors are those of the York, New York, USA.
authors.
Gillespie, T., J. Chu, E. Frankenberg, and D. Thomas. 2007.
Assessment and prediction of natural hazards from satellite
LITERATURE CITED imagery. Progress in Physical Geography 31(5):459–470.
Armenian, H., M. Masahiro, A. Melkonian, A. Hovanesian, Gray, C., E. Frankenberg, T. Gillespie, C. Sumantri, and D.
N. Haroutunian, P. Saigh, K. Akiskal, and H. Akiskal. 2000. Thomas. 2011. Population movements in Sumatra after the
Loss as a determinant of PTSD in a cohort of adult survivors tsunami. Paper presented at the 2011 Annual Meetings of the
of the 1998 earthquake in Armenia: implications for policy. Population Association of America, Washington, D.C., USA.
Acta Psychiatrica Scandinavia 102:58–64.
Horton, J. 2011. Natural disasters—taking a longer term view.
Blanchard, E.B., J. Jones-Alexander, T. C. Buckley, and C. Lancet 377: 439.
A. Forneris. 1996. Psychometric properties of the PTSD
checklist (PCL). Behavior Research and Therapy 34:669– Huber, P. 1981. Robust statistics. Wiley, New York, New
673. York, USA.

Borrero, O. S. 2005. The great Sumatra earthquake and Indian Kerr, R. 2005. Model shows island muted tsunami after latest
Ocean tsunami of December 26, 2004. EERI Special Indonesian quake. Science 308:341.
Earthquake Report March 2005:1–8. Kohl, P., A. O’Rourke, D. Schmidman, W. Dopkin, and M.
Bunting, T., C. Chapman, P. Christie, S. Singh, and J. Sledzik. Birnbaum. 2005. The Sumatra-Andaman earthquake and
2007. The science of tsunamis. Oilfield Review Autumn 2007. tsunami of 2004: the hazards, events, and damage. Prehospital
and Disaster Medicine 20(6):356–363.
Buttenheim A. 2010. Impact evaluation in the post-disaster
setting: a case study of the 2005 Pakistan earthquake. Journal Lay, T., H. Kanamori, C. Ammon, M. Nettles, S. Ward, R.
of Development Effectiveness 2(2): 197–227. Aster, S. Beck, S. Bilek, M. Brudzinski, R. Butler, H. DeShon,
G. Ekstrom, K. Satake, and S. Sipkin. 2005. The Great
Doocy, S., Y. Gorokhovich, G. Burnham, D. Balk, and C. Sumatra-Andaman earthquake of 26 December 2004. Science
Robinson. 2007. Tsunami mortality estimates and 308:1127–1133.
vulnerability mapping in Aceh, Indonesia. American Journal
of Public Health 97:S146–151. http://dx.doi.org/10.2105/ Linnerooth-Bayer, J., R. Mechler, and G. Pflug. 2005.
AJPH.2006.095240 Refocusing disaster aid. Science 309(5737): 1044–1046.

Engel, E. 1895. Die Lebenskosten Belgischer Arbeiter- Lutz, W., and K. C. Samir. 2011. Global human capital:
Familien Fruher and Jetzt. International Statistical Institute integrating education and population. Science 333:587–592.
Bulletin 9:1–74. Marris, E. 2005. Inadequate warning system left Asia at the
Frankenberg, E., J. Friedman, T. Gillespie, N. Ingwersen, R. mercy of tsunami. Nature 433:3–5.
Pynoos, I. Rifai, B. Sikoki, A. Steinberg, C. Sumantri, W. Monecke, K., W. Finger, D. Klarer, W. Kongko, B. McAdoo,
Suriastini, and D. Thomas. 2008. Mental health in Sumatra A. Moore, and S. Sudrajat. 2008. A 1,000-year sediment record
after the tsunami. American Journal of Public Health 98 of tsunami recurrence in northern Sumatra. Nature 455:1232–
(9):1671–1677. http://dx.doi.org/10.2105/AJPH.2007.120915 1234.
Frankenberg E., T. Gillespie, S. Preston, B. Sikoki, and D. National Research Council. 2006. Facing hazards and
Thomas. 2011. Mortality, the family and the Indian Ocean disasters: understanding human dimensions. The National
tsunami. Economic Journal 121(554):F162–182. Academies Press, Washington, D.C., USA.
Gaillard, J., E. Clavé, O.Vibert, Azhari, Dedi, J. Denain, Y.
Efendi, D. Grancher, C. Liamzon, D. Sari, and R. Setiawan.
 Ecology and Society 18(2): 16
http://www.ecologyandsociety.org/vol18/iss2/art16/

Norris, F., M. Friedman, P. Watson, C. Byrne, E. Diaz, and Umitsu, M., C. Tanavud, and B. Patanakanog. 2007. Effects
K. Kaniasty. 2002. 60,000 disaster victims speak: Part I. An of landforms on tsunami flow in the plains of Banda Aceh,
empirical review of the empirical literature. Psychiatry 65 Indonesia, and Nam Khem, Thailand. Marine Geology
(3):207–239. 242:141–153.
Padgett A., and T. Warnecke. 2011. Diamonds in the rubble: Weathers, F. W., B. T. Litz, D. S. Herman, J. A. Huska, and
the women of Haiti. Institutions, gender equity and human T. M. Keane. 1993. The PTSD checklist: reliability, validity,
development in Haiti. Journal of Economic Issues XLV(3): and diagnostic utility. Paper presented at the Annual Meeting
527–558. of the International Society for Traumatic Stress Studies, San
Antonio, Texas, USA.
Prasetiyawan, E. Viora, A. Maramis, and B. Keliat. 2006.
Mental health model of care programmes after the tsunami in
Aceh, Indonesia. International Review of Psychiatry 18)
6):559–562.
Ramakrishnan, D., S. Ghosh, V. Raja, R. Vinuchandran, and
A. Jeyram. 2005. Trails of the killer tsunami: a preliminary
assessment using satellite remote sensing and GIS techniques.
Current Science 88(5):709–712.
Rofi, A., S. Doocy, and C. Robinson. 2006. Tsunami mortality
and displacement in Aceh province, Indonesia. Disasters 30
(3):340–350.
Sastry, N., and M. Vanlandingham. 2009. Prevalence of and
disparities in mental illness among pre-Katrina residents of
New Orleans one year after Hurricane Katrina. American
Journal of Public Health 99(S3):725–731.
Sinadinovski, C. 2006. The event of 26th December 2004—
the biggest earthquake in the world in the last 40 years. Bulletin
of Earthquake Engineering 4:131–139.
Smith, M.Y., W. Redd, K. DuHamel, S. Vickberg, and P.
Ricketts. 1999. Validation of the PTSd checklist-civilian
version in survivors of bone marrow transplantation. Journal
of Trauma Stress 12(3):485–493.
Strauss, J., and D. Thomas. 2008. Health over the life course.
Pages 3375–3474 in T. P. Schultz and J. Strauss, editors.
Handbook of development economics, Volume 4. North-
Holland Biomedical Press, Amsterdam, The Netherlands.
Telford, J., and J. Cosgrave. 2007.The international
humanitarian system and the 2004 Indian Ocean earthquake
and tsunami. Disasters 31(1):1–28.
The Consultative Group on Indonesia. 2005. Indonesia:
preliminary damage and loss assessment. The December 26,
2004 Natural Disaster. Technical Report, Bappenas and the
International Donor Community, Jakarta, Indonesia. [online]
URL: http://www.unep.org/tsunami/reports/damage_assessment.
pdf
Tierney, K. 2007. From the margins to the mainstream?
Disaster research at the crossroads. Annual Review of
Sociology 33:503–535.