Anda di halaman 1dari 28

IJEC (2013) 45:279–305

DOI 10.1007/s13158-012-0077-7

ORIGINAL ARTICLE

Assessment of the Psychosocial Development


of Children Attending Nursery Schools in Karen
Refugee Camps in Thailand

Akiko Tanaka

Published online: 5 February 2013


Ó Springer Science+Business Media Dordrecht 2013

Abstract The Karen, an ethnic minority group in Burma, have experienced a


prolonged state of exile in refugee camps in neighboring Thailand because of ethnic
conflict in their home country. Nursery schools in the three largest Karen refugee
camps aim to promote the psychosocial development of young children by pro-
viding a child-centered, creative, learning-friendly environment. Psychosocial
development and potentially concerning behaviors of two to five-year-old children
in nursery schools were examined by use of a psychosocial checklist. The results
showed that psychosocial development of the children increased with age, with most
five year olds being proficient in playing cooperatively with other children. Sadness
or emotional outbursts were observed for a third of the children. Difficulty sepa-
rating from parents was also observed. The results also showed that children who
attended the nursery schools for more than a year were better at playing coopera-
tively with other children and were more aware of their own and others’ feelings.
On the other hand, children who were newer to the nursery schools were more polite
and better at following rules and controlling their feelings when frustrated. The
results indicate that nursery schools can be a promising practice to promote healthy
psychosocial development of children in protracted refugee situations.

Keywords Refugee  Preschool  Psychosocial development  Nursery school 


Mental health  Children  Violence  Conflict  Burma  Karen  Refugee camp

Résumé Les Karens, un groupe ethnique minoritaire de Birmanie, ont vécu en état
d’exil prolongé dans des camps de réfugiés dans le pays voisin de la Thaı̈lande en
raison de conflits ethniques dans leur pays d’origine. Les écoles maternelles des
trois plus vastes camps de réfugiés karens visent à promouvoir le développement
psychosocial des jeunes enfants en mettant à leur disposition un environnement

A. Tanaka (&)
WellShare International, 122 West Franklin Avenue, Suite 510, Minneapolis, MN 55404, USA
e-mail: tana0011@umn.edu; atanaka@wellshareinternational.org

123
280 A. Tanaka

centré sur l’enfant, créatif et propice à l’apprentissage. Le développement psy-


chosocial ainsi que les comportements potentiellement à risque d’enfants de ma-
ternelles de 2 à 5 ans ont été analysés à l’aide d’une liste de contrôle psychosociale.
Les résultats révèlent que le développement psychosocial de ces enfants augmente
en fonction de leur âge, la plupart des enfants de 5 ans démontrant une capacité à
jouer en coopération avec d’autres enfants. Un tiers des enfants manifestent de la
tristesse ou des accès d’émotions. On observe également qu’ils ont de la difficulté à
se séparer de leurs parents. Les résultats montrent aussi que les enfants qui fréqu-
entaient les écoles maternelles depuis plus d’un an étaient mieux disposés à jouer en
coopération avec d’autres enfants et plus conscients de leurs propres sentiments et
de ceux des autres. D’autre part, les enfants qui étaient plus nouveaux en maternelle
étaient plus polis, suivaient mieux les règlements et contrôlaient mieux leurs
émotions quand ils étaient frustrés. Les résultats indiquent que les écoles matern-
elles peuvent être une pratique prometteuse pour promouvoir le sain développement
psychosocial d’enfants en situation prolongée de réfugiés.

Resumen Los Karen, un grupo étnico minoritario de Birmania, han experimen-


tado un estado de exilio prolongado en campos de refugiados en la vecina Tailandia
debido al conflicto étnico en su tierra natal. Las guarderı́as de los tres campos de
refugiados Karen más grandes, tienen el objetivo de promover el desarrollo psi-
cosocial de los niños brindándoles de un ambiente enfocado en ellos, creativo y que
invita al aprendizaje. Se examinó el desarrollo psicosocial y comportamientos po-
tencialmente preocupantes de niños de dos a cinco años usando pautas de chequeo
psicosocial. Los resultados mostraron que el desarrollo psicosocial de los niños
aumenta con la edad, y que la mayorı́a de los niños de 5 años son capaces de jugar
de manera cooperativa con otros niños. Un tercio de los niños presentó tristeza o
arrebatos emocionales. También se observó dificultad al separarse de los padres.
Los resultados también muestran que los niños que asistieron a guarderı́as por más
de un año eran mejores jugando de forma cooperativa con otros niños y estaban más
conscientes de los sentimientos de sı́ mismos y de los demás. Por otro lado, los niños
nuevos en las guarderı́as eran más corteses y eran mejores siguiendo reglas y
controlando sus sentimientos cuando se frustraban. Los resultados indican además
que las guarderı́as pueden ser una práctica prometedora para promover un desarrollo
psicosocial saludable de los niños que viven en campos de refugiados por perı́odos
de tiempo prolongado.

Background

Burma (also called Myanmar), a country in Southeast Asia surrounded by Thailand


and Laos on the east, the People’s Republic of China on the northeast, India on the
northwest, and Bangladesh on the west, has been ruled by a military regime since
1962. Apart from the majority Burman ethnicity, there are more than 100 different
ethnic minority groups within the country, making up approximately one-third of
the population and occupying 55 % of the land, mainly in the border areas of the
country (Human Rights Documentation Unit 2009). The Burman-led central

123
Psychosocial Development of Karen Refugee Children 281

government and ethnic minority groups have a long history of military conflicts,
with the minority groups seeking increased autonomy or independence. The Karen
are an ethnic minority group who reside mostly in the southern and eastern parts of
the country with a claimed population of seven million, although official census
figures from 1983 show the number to be much smaller (South 2011). Military
conflict between the ruling government and the Karen started in 1949, soon after
Burma’s independence from Britain, making it the oldest continuing civil war in the
world (Human Rights Documentation Unit 2009).
The military regime of Burma has been known for its suppression of
prodemocracy movements and exploitation and human rights violations against
its citizens (Bureau of Democracy, Human Rights, and Labor 2011).1 Its
exploitation and attacks against the Karen, mostly in villages nestled in jungles
along the Thai–Burmese border, have been well documented (Bureau of Democ-
racy, Human Rights, and Labor 2007, 2011; Human Rights Documentation Unit
2009; The Karen Women’s Organization 2007). Karen villages have been burnt or
forcibly relocated, and villagers’ property and food supplies have been confiscated
or destroyed. The villagers have been cut off from economic sustenance, raped,
tortured, shot and killed on sight, and forced into labor, for example serving as
porters for the Burmese military. Numerous landmines have been placed by both the
Burmese military and armed Karen groups around villages and jungles nearby.
These difficulties have displaced many Karen in jungles along the Thai–Burmese
border, and have pushed a significant number across the border to Thailand since
1984 (Burmese Border Consortium 2004). As of September 2006, there were seven
official Karen refugee camps in Mae Hong Son, Tak, Kanchanaburi, and
Ratchanaburi provinces of Thailand bordering Burma, housing a total of over
128,000 people (Thailand Burma Border Consortium 2006). As of the end of 2005,
of all people of concern in Thailand (including asylum seekers and refugees from
Laos, Cambodia, China, and all ethnicities from Burma) 48 % were children (under
18 years old) and 14 % were under five years old (UNHCR 2006).
Because of the long-term refugee situation for the Karen, many children have
been born in the camps and have grown up there. Although no longer experiencing
the chaos, insecurity, and extreme deprivation often existing in the newer camps,
long-term encampment of refugees has its own challenges. The residents of the
camp suffer from a lack of freedom of movement, crowding, lack of sufficient
employment opportunities and dependency on minimal international aid, frustration,
despair, or helplessness that comes from uncertainty about their future, and a lack of
constructive activities. Karen children who were a part of more recent waves of
migration to the camps have experienced violence and an arduous journey to
Thailand, and may have been separated from or lost parental figures. Even if the
children were born in the camps, their parents were likely to have been exposed to
traumatic experiences that prompted their migration and may be suffering mental
health consequences from these and other stresses of life in refugee camps.

1
The new military-backed government after the election of November 2010 implemented a series of
reforms and signed a ceasefire with the Karen in January 2012; however, the situation on the ground in
Karen areas has not substantially changed (Hindstrom, 2012; Karen Human Rights Group, 2012).

123
282 A. Tanaka

Mental health issues experienced by refugee children are well-documented.


However, most studies of mental health among children affected by conflict focus
on school-age children and adolescents or cover a wide age range in their original
community or after resettlement (Bronstein and Montgomery 2011; Goldstein et al.
1997; Kinzie et al. 1986; Lustig et al. 2004; Mollica et al. 1997), and the mental
health status of young children in refugee camps has hardly been described. In
addition, most of the literature focuses on the mental health of children who have
been exposed to significant trauma themselves, and the mental health effects of a
long-term refugee situation on children who were born in camp or may not
remember the events that led to the camp are uncertain. However, even if the
children did not experience traumatic events themselves, their parents probably did,
and they may be traumatized and their mental health affected. Parental mental
health and experience of trauma have been found to affect their children’s mental
and behavioral health (Bombay et al. 2009; Dekel and Goldblatt 2008; Kiernan and
Huerta 2008; Riley et al. 2009; Vaage et al. 2011; Yehuda et al. 2001).
Data on positive psychosocial development, for example prosocial behavior,
adaptation, and peer relationships, of refugee children have been scarce and mixed,
with some studies showing higher or comparable levels of positive behavior
exhibited by children with refugee or war experience (Fazel et al. 2009 (prosocial
behavior); Macksoud and Aber 1996; Raboteg-Saric et al. 1994), and some showing
lower levels (Fazel et al. 2009 (peer problems); Habir et al. 1994; Kerestes 2006).
These data are mostly for older children, with the exception of Raboteg-Saric et al.
(1994) who focused on preschoolers during the war in Croatia. Studies of non-
refugee children who have experienced similar adversity, for example poverty,
homelessness, maternal mental health issues, and exposure to violence, have shown
potential delay in development, including psychosocial development (Alaimo et al.
2001; Bassuk and Rosenberg 1990; Flisher et al. 1997; Kiernan and Huerta 2008;
Lamphear 1985; Osofsky 1995; Riley et al. 2009).
Thus, despite the multitude of factors that are likely to negatively affect refugee
children psychologically and socially, there has been a scarcity of data in this area
for young children, especially those in refugee camps. In addition, there has been a
lack of data on primary prevention programs that aim to support positive
psychosocial development of young refugee children. The need for such programs
is especially pronounced given that protracted refugee situations have become the
norm in the world (Milner and Loescher 2011), which means that children may
spend a large part of their childhood in refugee camps.
One program designed to promote psychosocial development of young refugee
children is the nursery schools run in the three largest Karen camps by a
community-based women’s organization and supported by an international non-
governmental organization (NGO). The schools provide protection, nutritious
meals, and supportive, child-centered curricula. This paper presents data on the
psychosocial development of young Karen children in nursery schools in these three
refugee camps in Thailand with the purpose of:

123
Psychosocial Development of Karen Refugee Children 283

1 describing the psychosocial development of children and potential behaviors of


concern; and
2 examining whether or not nursery schools with child-centered curricula
promoted psychosocial development of young children.

Experiences of Karen Children

Refugee Camps

This paper focuses on nursery schools in three Karen refugee camps in Tak province
of Thailand: Mae La, Umpiem Mai, and Nu Po. These camps, established in the mid
to late 1990s by consolidation of smaller camps that were set up across the border
from Burma since 1984, are located within 10 km of the border. As of September of
2006, the three camps had a combined population of 81,805 (48,956 in Mae La,
19,561 in Umpiem Mai, and 13,288 in Nu Po) (Thailand Burma Border Consortium
2006). The year 2006 saw an increase in the influx of refugees to the Karen camps
because of a military offensive by the government forces against villages in the
Karen State (Human Rights Documentation Unit 2007), and the start of a large scale
effort to resettle the refugees in a third country (UNHCR 2007).
With houses built from bamboo on stilts and roofs made of leaves, the camps
resemble Karen villages along the border, just many times larger. There are schools,
clinics, markets, tea shops, churches, temples, and mosques in the camps. The
camps are very organized, divided into zones and sections, and managed by Karen
refugee committees (Burmese Border Consortium 2004). The definitive difference
between the villages from which the refugees fled and the camps is the barbed wire
fence surrounding them. The residents of the camps are prohibited from leaving
without permission and risk arrest, detention, and deportation if caught without
documentation (Human Rights Documentation Unit 2007). Crowding and lack of
space is also a problem (Burmese Border Consortium 2004).
Camp residents rely on food rations provided by aid organizations for most of
their staple foods, with other food commodities expected to be supplemented by
refugees themselves (Thailand Burma Border Consortium 2007). Other supplies, for
example house-building materials, stoves, cooking fuel, and some clothing, are
provided by aid agencies. Whereas most of the Karen refugees made their living
farming in their home villages, opportunity to farm in the camps is limited by space.
Households in the camps generate income by casual labor, employment with
agencies inside the camps, sale of own products or handicrafts, or engagement in
other trades (Cardno Agrisystems Limited 2009). Although leaving the camps for
work is prohibited by Thai authorities, some refugees do sneak out to work on farms
or providing skilled labor, for example carpentry (Cardno Agrisystems Limited
2009; M. Thitiphanawan, personal communication, May 23, 2012). As of 2009, in
four camps along the Thai–Burmese border surveyed, including Mae La and Nu Po,
the average monthly household earned income was 960 Thai baht (approximately 29
dollars using 2009 currency exchange rate) (Cardno Agrisystems Limited 2009).

123
284 A. Tanaka

Karen Culture and Socialization of Children

The Karen consist of several subgroups, most in the camps along the Thai–Burmese
border being Pwo or Sgaw. They are traditionally animists or Buddhists, with a
significant number having been converted to Christianity by British missionaries in
the nineteenth century. Karen society is described as hierarchical, with status
differentiated on the basis of a variety of factors, for example age, gender, education
level, and home region (urban/rural, mountain/lowlands) (Karen Buddhist Dhamma
Dhutta Foundation 2010). Deference is paid to those who are older, parents, or other
authority figures, for example teachers, monks, or priests. This is seen in how they
address each other, not talking back, and customs such as who is served a meal or
handed a tobacco purse first, seating arrangements, and how younger individuals ask
permission to pass behind a seated elder (Anders 1980; M. Thitiphanawan, personal
communication, June 5, 2012). The elders or teachers, on the other hand, are
believed to be responsible for leading, guiding, and directing those who are younger
(M. Thitiphanawan, personal communication, June 5, 2012).
In the camps, the average family size is 4.7 and the household size approximately
six (Cardno Agrisystems Limited 2009; Thailand Burma Border Consortium 2007).
A third of the families consist of nuclear families whereas over half consist of
extended or multi-family households (Cardno Agrisystems Limited 2009). The
number of households headed by a single female is small, approximately 6 %
(Thailand Burma Border Consortium 2007).
Description of Karen parenting or socialization of children, either in their villages
or refugee camps is scarce, except in anecdotal stories. What limited information
there is on Karen child rearing indicates that when there is an infant in the
household, the mother often stays at home. If the mother works outside the home,
young children are often supervised by relatives, for example a grandmother, other
adults, or older siblings (Anders 1980; M. Thitiphanawan, personal communication,
May 23, 2012). Children are expected to work on household chores at an early age
(Phan and Lewis 2010). When children misbehave, they are typically disciplined by
their parents by beating, shouting, or threats, although a wide range of discipline
practices is used across families (M. Thitiphanawan, personal communication, May
23, 2012). Conformity and good behavior are also enforced by teachings by elders,
monks, or priests (Ya, personal communication, May 13, 2012).
Life in refugee camps brings about new challenges in parenting. During focus
groups conducted by the author in 2006 with parents, teachers, and partner
organizations of the nursery schools, participants mentioned concerns for children’s
safety (playing on the streets, being hit by a car, playing in the river) and lack of
monitoring of children while parents worked, some even going away for an
extended period outside the camp for work. Some bad habits that children had
developed (roaming around, stealing) were also mentioned. Children are also
exposed to increased problem behavior by adults in the camps, for example
domestic violence, alcohol and drug use, and depression, which may have increased
in the Karen community because of exposure to violence and militarism, poverty,
and the stress and frustration of living in the camps (Ya, personal communication,
May 13, 2012; Burmese Border Consortium 2004: Human Rights Documentation

123
Psychosocial Development of Karen Refugee Children 285

Unit 2007). Children are more easily exposed to these behaviors in the camps
because of crowding and the proximity to neighbors (Ya, personal communication,
May 13, 2012). Health issues (e.g., diarrhea) and insufficient nutrition were also
mentioned in the focus groups as concerns for children (also see Thailand Burma
Border Consortium 2007).

Description of the Program

The nursery schools in Karen refugee camps along the Thai–Burmese border started
informally. NGO support in Mae La camp began in 1998; in 2000 it began in
Umpiem Mai and Nu Po camps. At the time of this evaluation, a Karen women’s
organization operated the nursery schools day to day whereas the NGO provided
technical and material support and supervision. In 2006, 35 nursery schools in the
three camps employed 200 teachers, who were themselves refugees and residents of
the camps, and served 3,901 pre-kindergarten-aged children (2,355 in Mae La, 936
in Umpiem Mai, and 610 in Nu Po). The minimum qualification for the teachers was
having completed grade 6 and being at least 18 years old.
The nursery school program addressed child development holistically, including
physical, cognitive, and psychosocial development. They offered a safe place to be
during the day, nutritious meals, a sanitary environment and good hygiene practices,
and stimulation, interaction, and affection within a child-centered curriculum in
which teachers were considered ‘‘facilitators’’ of child development. Children
learned letters, numbers, concepts, and how to interact with others by free play and
interactive, participatory activities that were appropriate to their age and develop-
ment, rather than the rote learning typical of traditional schools (Karen Buddhist
Dhamma Dhutta Foundation 2010; Phan and Lewis 2010).
Although the program addressed many facets of child development, the women’s
organization and the NGO considered that the primary aim of the nursery program
was to meet children’s psychosocial needs. The teachers provided emotional and
social support to children, promoted socialization and cooperation, and used
positive discipline. This was considered especially important for development of a
sense of security and to deter the cycle of violence for children growing up in a
community exposed to violence.
Quality and sustainability of the program was ensured by a curriculum manual
and training by an international NGO. Topics covered included the principles of
early childhood development, guidelines for caregivers, concepts of child-centered
nursery school management, child hygiene and nutrition, teaching methodology,
and self-evaluation. Creative activities were also promoted by use of activity
booklets created with the participation of children and teachers, and included songs,
stories, games, picture drawing, and role playing using local materials and themes
(e.g., depicting children’s home life). Parents also received booklets describing
early child development, parenting, and child learning activities and were
encouraged to get involved in nursery schools through parent–child meetings and
volunteering at nursery schools.

123
286 A. Tanaka

Method

Development of Psychosocial Checklist

A brief checklist to assess psychosocial development of children which could be


completed quickly by nursery school teachers was needed. Based on a review of
psychosocial development literature and existing instruments to assess psychosocial
development of young children, a list of 92 items was compiled reflecting 17
categories (e.g., peer relationships, self-esteem, following rules, externalizing,
internalizing) describing developmentally appropriate and worrisome behaviors and
characteristics. Although the focus of the program and the evaluation was on the
psychosocial development of children, items describing possibly concerning
behaviors were included because of literature suggesting a high prevalence of
mental health and behavioral issues among refugee children, and it was considered
important for the program to be aware of behavioral or emotional issues of the
children they served.
When the initial list of items was available in English, six to eight nursery school
teachers from each camp gave feedback on them. The teachers who provided the
input were selected by a Karen Teacher Trainer and were experienced teachers in
the camps. Because of the large number of items, teachers at each camp gave
feedback on different sets of 30–31 items. The purpose of the checklist explained to
the teachers was to increase their knowledge of the children’s behavior. The items
were presented to the teachers in a group setting, with the Karen NGO staff orally
translating them into Karen. The items intended to measure positive psychosocial
development were rated with regard to their importance to children’s success in the
classroom and in life. The items intended to measure potential problem behaviors
were rated with regard to how much teachers would be concerned if the children
showed them.
Based on these ratings, and to reduce redundancy, 29 items were selected for the
first draft of the checklist (20 items describing positive psychosocial development,
nine items of possible concern). At least two items were selected from six categories
reflecting positive psychosocial development (group activities, social skills/leader-
ship/politeness, self-esteem, following rules, self-control, and emotions). Only one
item (appropriately asks for help if she/he needs it) was selected from the
assertiveness category, and none from the peers category was selected because of
redundancy with items in other categories (e.g., group activities) despite some high
ratings. No item was selected from the ‘‘other’’ category. Three items concerned
with cooperation with peers, following rules in games, and following teachers’
directions received ratings that were slightly lower than other items selected but
were added on the basis of a recommendation by Karen NGO staff that they were
important to the Karen culture.
Of the items describing behaviors of concern, at least one item was selected from
categories reflecting externalizing behaviors, internalizing behaviors, negative self-
esteem, passivity, and peer relationship problems. The highest rated items did not
include any from the self-regulation, autistic, and physical/physiological/develop-
mental problem categories.

123
Psychosocial Development of Karen Refugee Children 287

The checklist asked the teachers to rate the children on the basis of their behavior
in the past month. For the response choices, ‘‘True’’ (Yes) and ‘‘False’’ (No) were
preferred for their simplicity. In addition, for the items describing positive
psychosocial development, a middle category ‘‘Sometimes true, somewhat true, or
requires assistance or encouragement’’ was added to capture the stage of
development when the skills were unstable and displayed only occasionally or
with prompts or assistance by adults. For the item, ‘‘Follows rules when playing
games with others’’, a response choice of ‘‘The child does not participate in many
games with rules’’ was added, which was treated as missing in the analysis. For
more worrisome behaviors, ‘‘Sometimes or somewhat true’’ as a middle rating was
added to capture display of some difficult or negative emotions or behaviors which
were not frequent or intense enough to be of concern.
The checklist was translated into Sgaw Karen, one of the Karen dialects spoken
by most Karen, and back-translated into English. The original checklist and the
back-translated version were compared, issues were discussed, and revisions were
made.
The resulting checklist was pilot tested in each camp (a total of 50 checklists
were completed) by convenience samples of nursery school teachers. On the basis of
the review of the data and feedback from the Karen-speaking NGO staff and
teachers, six items were deleted and three items were added to create the final
26-item checklist. The items were deleted based on teachers’ difficulty understand-
ing the items (e.g., seeking approval and praise, compromising, negative self-
esteem) and redundancy with other items (e.g., having quick temper). Three added
items described presentations of depression symptoms (sadness, irritability, physical
symptoms) that initially did not make the cut but were considered important enough
because of literature suggesting high prevalence of depression among traumatized
children. Wording was modified between versions, most often for ease of translation
into Karen.
The final checklist included 17 items asking about behaviors reflecting positive
psychosocial development and nine items of possible concern. The final version also
included a section that asked the child’s age, gender, birth date, the date the child
started attending the nursery school, and the date the checklist was completed.
Once the final version of the checklist was developed, the checklists were
completed by nursery school teachers during teacher training. Each teacher
completed checklists for two children selected randomly from their classroom.
Random selection was ensured by each teacher choosing two numbers between one
and the total number of children in the classroom, and then looking at a numbered
list of children in her classroom. The checklists were completed for the children
corresponding to the numbers the teacher chose. As the teachers completed the
checklist, Karen-speaking NGO staff also read aloud the items and response choices
verbatim in front of the group. Three hundred and fifty checklists were completed
across 35 nursery schools. When administration was complete, feedback on the
checklist was again sought from the nursery school teachers. On the basis of the
feedback, two items intended to measure psychosocial development (showing pride
in self, resisting peer pressure) were found to be of questionable validity in Karen
language, and were thus excluded from the analysis.

123
288 A. Tanaka

Once the checklists were completed by the teachers, the data were entered into a
database. Data analysis for this paper was conducted using SPSS version 19.0.0.

Results

Data Analysis

First, demographics of the children were examined, including gender distribution


and age (distribution and mean). Children’s age at the time of the assessment, age
when children started attending nursery school, and length of time the children
attended the nursery school were calculated using the birthdate, the date of checklist
completion, and the date of entry into the nursery school. For the length of time in
nursery school, no adjustment was made for two months of vacation per year when
the nursery schools were closed. This analysis was first conducted including all the
children for whom the psychosocial checklist had been completed, to show the full
range of the children in the nursery schools. In addition, results are also presented
restricting the sample to only two to five year olds, because the number of children
younger than two or older than five was small, and they were excluded from
subsequent analysis.
Second, ratings on psychosocial development and behaviors of possible concern
were analyzed by age to examine the validity of the checklist and to describe
psychosocial development of children and prevalence of potential behaviors of
concern. Higher levels of psychosocial development were expected by age, thus
higher scores on these items with increased age indicated validity of the checklist.
No such hypothesis was in place regarding behaviors of possible concern.
The analysis included two steps. One was to examine the overall psychosocial
development and possible concern scores, and the second was to show the results for
individual items.
For the overall analysis, the checklist items were factor analyzed using principal-
components analysis and varimax rotation to develop composite scales. For all
items, ‘‘False’’ was coded as 1, ‘‘Sometimes true, somewhat true, or requires
assistance or encouragement’’ or ‘‘Sometimes or somewhat true’’ was coded as 2,
and ‘‘True’’ was coded as 3. Multiple rounds of exploratory factor analysis were run
specifying different numbers of factors. Among those, a solution with two factors
representing psychosocial development and possible concern yielded the best fit
(factor loadings are presented in Table 1). One item, ‘‘Easily gets angry, gets upset,
or cries’’ did not load highly on either of the factors (factor loading \0.1), so was
excluded from analysis using these composite scales but was included in analysis of
individual items. The alpha coefficient for the scales created from these factors were
acceptable, with 0.747 for psychosocial development and 0.730 for possible
concern.
Scale scores were created by averaging scores from the items falling in each
scale. The plan was to exclude children who had missing values for more than half
of the items for each scale. However, the maximum number of missing values was
four for psychosocial development and three for possible concern (with one child

123
Psychosocial Development of Karen Refugee Children 289

Table 1 Factor loadings of checklist items


Factors

Psychosocial Possible
development concern

Follows school and classroom rules 0.577 -0.195


Follows rules when playing games with others 0.560 -0.013
Follows teachers’ directions 0.550 -0.044
Greets adults when she/he sees someone she/he knows 0.537 0.181
Apologizes (says ‘‘I am sorry’’) when she/he does something wrong 0.533 -0.078
Participates in games or group activities with other children 0.530 -0.073
Can talk about the kind of feelings she/he is experiencing 0.515 0.138
(e.g., ‘‘I am mad’’ ‘‘I am sad’’)
Puts away materials or toys after using them 0.512 0.094
Appropriately asks for help if she/he needs it 0.508 0.082
Shows empathy and caring for others 0.485 -0.176
Shares with other children 0.470 -0.290
Participates in activities with other children that require cooperation and 0.463 -0.196
working together (e.g., pretend play with other children, building something
with blocks together)
Says good words, not arguing, yelling, or shouting 0.288 -0.224
When she/he cannot get something she/he wants or is not allowed to do what 0.287 0.141
she/he wants to do, she/he is able to control her/his anger or hurt feelings
Does not ask questions or statements that might embarrass or hurt others 0.280 0.176
Is rejected by other children 0.048 0.653
Has difficulty separating from parents -0.007 0.645
Has fears 0.023 0.610
Frequently has headaches, stomach ache, or other physical symptoms (showing 0.083 0.596
signs that she/he does not feel good)
Is aggressive toward people or things -0.139 0.541
Has difficulty staying with and concentrating on activities -0.040 0.511
Seems sad -0.040 0.506
Does not express feelings or emotions 0.074 0.487
Easily gets angry, gets upset, or cries 0.068 0.042

N = 270

missing all items under this scale), so none of the children were excluded for
missing values. Two scale scores were analyzed using analysis of variance
(ANOVA) using age (four levels ranging from two to five year olds) and gender
(two levels) as factors and including the interaction between the two.
For individual item-level analysis, the percentage rated as ‘‘True’’ was examined
to provide a more descriptive picture of the children at each age. This metric was
chosen because it was regarded as more intuitive to interpret statements such as
‘‘60 % of five year olds shared with other children whereas only 30 % of three year

123
290 A. Tanaka

olds did’’, compared with average scores for each item, which may vary by decimal
points across age and be difficult to interpret in terms of how they are reflected in
the actual behaviors of children. For psychosocial development items, the
percentage ‘‘True’’ reflects the percentage of children who engage in each behavior
consistently and independently, rather than those who are in the process of
mastering it. Focusing on those who were rated as ‘‘True’’ was especially important
for behaviors under the possible concern scale, to exclude occasional negative
behaviors or feelings that most children exhibit. Statistical significance of the
difference by age was examined by use of the chi-squared test.
A third analysis examined the effect of the nursery schools on the psychosocial
development of children. The hypothesis was that nursery schools would have a
positive effect in promoting the psychosocial development of children. No such
hypothesis was in place for behaviors of concern, because the teachers were not
trained to handle or reduce behavioral or mental health issues among children.
Because all checklists were completed for children attending the nursery schools, it
was impossible to directly examine the effect of attending nursery schools on
psychosocial development. However, an attempt was made to indirectly answer this
question by comparing the scores for children who had attended nursery schools for
different lengths of time. Analysis of covariance (ANCOVA) with the age of the
children as covariate and length of time in nursery school (two levels, up to one year
vs. more than one year in nursery school) as a factor was conducted. The cut-off of
one year for length of time in nursery schools was selected so there would be
sufficient sample size in each group for all age groups.
A decision was made to conduct individual item-level analysis if the length of
time in nursery school was statistically significant in ANCOVA using composite
scales. For each age, the percentage endorsing the item was compared by length of
time in nursery schools. The chi-squared test and Fisher’s exact test were used to
examine statistical significance of differences.

Description of the Children

Descriptive information for the 350 children for whom the psychosocial checklists
were completed by their teachers, and for 331 children who were two to five years
old at the time of checklist completion, are listed in Table 2. The number of children
assessed in each camp approximately reflected the comparative number of children
served in nursery schools in each camp. Slightly more than half of the children were
boys. Most of the children were two to five years old, but there were some children
who were under one year old and some as old as seven. The largest number of
children entered nursery schools when they were two (41 %) or three (36 %) years
old. Eight percent of the children had entered nursery schools when they were
younger than two years old, and 5% did not enter nursery schools until they were
five years old.
The length of time the children attended nursery schools is listed by age of
children in Table 3. The length of time generally increased with age for children up
to five years old, but six and seven year olds tended to have attended nursery schools
for a shorter time (most commonly 12–24 months).

123
Psychosocial Development of Karen Refugee Children 291

Table 2 Basic descriptive information about the children


Overall 2–5 year olds

N %/Mean (SD) N %/Mean (SD)

Camp 350 100.0 % 331 100.0 %


Mae La 226 64.6 % 215 65.2 %
Umpiem Mai 84 24.0 % 81 24.5 %
No Po 40 11.4 % 35 10.6 %
Gendera 349 100.0 % 330 100.0 %
Girls 163 46.7 % 153 46.4 %
Boys 186 53.3 % 177 53.6 %
Ageb 342 4.33 (0.96) 331 4.29 (0.88)
0 year old 1 0.3 % Na Na
One year old 1 0.3 % Na Na
Two years old 30 8.8 % 30 9.1 %
Three years old 95 27.8 % 95 28.7 %
Four years old 122 35.7 % 122 36.9 %
Five years old 84 24.6 % 84 25.4 %
Six years old 8 2.3 % Na Na
Seven years old 1 0.3 % Na Na
Age starting nursery schoolb 342 3.13 (0.91) 331 3.10 (0.86)
0 year old 4 1.2 % 3 0.9 %
One year old 20 5.8 % 19 5.7 %
Two years old 139 40.6 % 139 42.0 %
Three years old 122 35.7 % 120 36.3 %
Four years old 40 11.7 % 38 11.5 %
Five years old 17 5.0 % 12 3.6 %
Number of years in nursery schoolb 342 1.20 (0.89) 331 1.19 (0.89)
a
One checklist was missing gender information
b
Eight checklists had missing or invalid dates

Psychosocial Development and Behaviors of Possible Concern by Age


of Children

Average scale scores by age and gender of children are listed in Table 4. Statistical
testing using ANOVA showed that psychosocial development scores increased with
age, as expected (F(3,322) = 6.091, p = 0.000). Scores on the possible concern
scale were also indicative of significant effects of age (F(3,321) = 2.877,
p = 0.036), with mean scores decreasing with age. Effects of gender and interaction
between age and gender were not statistically significant for either scale.
Item-level results by age of children, showing the percentage of children who
were rated as ‘‘True’’ for each item, are listed in Table 5. As expected, of the items
falling under psychosocial development, the percentage of children rated as ‘‘True’’

123
292 A. Tanaka

Table 3 Number of years in nursery school, by age of children at the time of assessment
Age Time in nursery school Total

Up to More than More than More than More than


1 year 1–2 years 2–3 years 3–4 years 4–5 years

0 year old 1 0 0 0 0 1
One year old 1 0 0 0 0 1
Two years old 23 5 2 0 0 30
Three years old 64 29 2 0 0 95
Four years old 34 60 23 5 0 122
Five years old 19 21 34 8 2 84
Six years old 0 5 3 0 0 8
Seven years old 0 1 0 0 0 1
Total 142 121 64 13 2 342

Nursery schools are closed for approximately two months each year. These two months were included in
the time of attendance

increased with age (the percentage being higher for five year olds than for two year
olds) for 13 out of the 15 items. The difference across age groups was statistically
significant for five items. The two items that did not show better results for five year
olds compared with two year olds were ‘‘Appropriately asks for help if she/he needs
it’’ and ‘‘Says good words, not arguing, yelling or shouting.’’
Items with the largest percentage of five year olds rated as ‘‘True’’ include
‘‘Participates in games or group activities with other children’’ (69 %), ‘‘Follows
rules when playing games with others’’ (56 %), ‘‘Participates in activities with other
children that require cooperation and working together’’ (54 %), ‘‘When she/he
cannot get something she/he wants or is not allowed to do what she/he wants to do,
she/he is able to control his/her anger or hurt feelings’’ (51 %), and ‘‘Appropriately
asks for help if she/he needs it’’ (50 %). The smallest percentage of five-year old
children were rated ‘‘True’’ on ‘‘Apologizes when she/he does something wrong’’
(22 %), ‘‘Follows school and classroom rules’’ (30 %), and ‘‘Greets adults when he/
she sees someone she/he knows’’ (33 %).
Of the items that measure behaviors of possible concern, across all ages, more
than 30 % of the children were rated by their teachers as seeming sad and as easily
getting angry, upset, or crying. The prevalence of these items for five year olds was
more than 25 %. Across all ages, 26 % of the children had difficulty separating from
their parents, with more than 30 % among three year olds, and more than 20 % for
five year olds. When change by age was examined, the percentage of children
frequently showing signs they were ill was 40 % for two year olds but much lower
for older age groups (12 % for five year olds), a statistically significant decrease.
Another item that was statistically significantly different by age was having fears,
which seemed to peak at three years of age.

123
Psychosocial Development of Karen Refugee Children 293

Table 4 Scale scores by age and gender of children


Overall Boys Girls

N Mean SD n Mean SD n Mean SD

Psychosocial developmenta,b
Overall 330 2.21 0.31 177 2.21 0.33 153 2.21 0.30
Two years old 30 2.01 0.30 19 1.98 0.30 11 2.06 0.31
Three years old 94 2.16 0.29 44 2.13 0.33 50 2.19 0.26
Four years old 122 2.23 0.31 64 2.22 0.30 58 2.24 0.32
Five years old 84 2.31 0.31 50 2.35 0.31 34 2.25 0.32
Possible concernc,d
Overall 329 1.78 0.43 177 1.78 0.43 152 1.78 0.43
Two years old 30 1.88 0.36 19 1.84 0.37 11 1.95 0.34
Three years old 93 1.86 0.45 44 1.84 0.41 49 1.88 0.48
Four years old 122 1.74 0.42 64 1.74 0.44 58 1.75 0.39
Five years old 84 1.72 0.43 50 1.77 0.45 34 1.64 0.39
a
Response choices were 1 = False, 2 = Sometimes true, somewhat true, or requires assistance or
encouragement, and 3 = True. Higher scores indicate higher level of development
b
Effects of age F(3,322) = 6.091, p = 0.000, sex F(1,322) = 0.190 (n.s.), age 9 sex F(3,322) = 1.235
(n.s.)
c
Response choices were 1 = False, 2 = Sometimes or somewhat true, and 3 = True. Higher scores
indicate higher levels of behavior/characteristics that are of possible concern
d
Effects of age F(3,321) = 2.877, p = 0.036, sex F(1,321) = 0.022 (n.s.), age 9 sex F(3,321) = 0.889
(n.s.)

Psychosocial Development by Length of Time in Nursery Schools

Average scores for the two scales by length of time in nursery schools (up to
one year compared with longer than one year), controlling for the age of children at
the time of the assessment, are listed in Table 6. Children who attended nursery
schools for a longer period of time had significantly higher scores on the
psychosocial development scale (F(1,328) = 4.070, p = 0.044). The scores on the
possible concern scale did not differ by the length of time in the nursery school.
Because ANCOVA results were statistically significant for the psychosocial
development scale, individual checklist items that belonged to the scale were also
examined for the effects of length of time the children were in nursery schools. For
each age (2–5 year olds), the percentage of children who were given a ‘‘True’’
rating was compared between those who attended the nursery schools up to one year
and more than one year. The items and age groups that showed more than a 10%
point difference by the length of time in nursery schools are presented in Table 7.
The children who had attended nursery school for a longer period of time were
rated as more likely to follow rules when playing games with others (two and
five year olds), participate in activities with other children that required cooperation
(four year olds), be able to talk about feelings they were experiencing (five year

123
Table 5 Percentage of children whose teachers gave them the highest rating (‘‘True’’), by age of the children at the time of the assessment
294

Two years old Three years old Four years old Five years old Overall

n n n n N

123
Item % Rated % Rated % Rated % Rated % Rated
‘‘True’’ ‘‘True’’ ‘‘True’’ ‘‘True’’ ‘‘True’’

Psychosocial developmenta
Participates in games or group activities with other 30 40.0 95 43.2 122 62.3 83 68.7 330 56.4
children***
Follows rules when playing games with othersb,** 29 31.0 88 29.5 115 38.3 81 55.6 313 39.6
Participates in activities with other children that 30 33.3 95 41.1 121 49.6 84 53.6 330 46.7
require cooperation and working together (e.g.,
pretend play with other children, building
something with blocks together)
When she/he cannot get something she/he wants 27 22.2 94 38.3 120 32.5 82 51.2 323 38.1
or is not allowed to do what she/he wants to do,
she/he is able to control her/his anger or hurt
feelings*
Appropriately asks for help if she/he needs it 29 51.7 95 44.2 120 45.8 84 50.0 328 47.0
Follows teachers’ directions 28 32.1 93 39.8 119 45.4 84 47.6 324 43.2
Can talk about the kind of feelings she/he is 30 26.7 94 28.7 122 24.6 84 45.2 330 31.2
experiencing (e.g., ‘‘I am mad’’ ‘‘I am sad’’)*
Says good words, not arguing, yelling, or shouting 29 55.2 95 50.5 121 39.7 84 42.9 329 45.0
Puts away materials or toys after using them 30 16.7 93 34.4 121 33.1 84 41.7 328 34.1
Shows empathy and caring for others 30 26.7 94 29.8 119 35.3 83 38.6 326 33.7
Shares with other children 30 16.7 95 34.7 122 37.7 83 36.1 330 34.5
Does not ask questions or statements that might 30 23.3 95 29.5 119 30.3 84 34.5 328 30.5
embarrass or hurt others
Greets adults when she/he sees someone she/he 30 10.0 94 29.8 120 30.8 84 33.3 328 29.3
knows
A. Tanaka
Table 5 continued

Two years old Three years old Four years old Five years old Overall

Item n % Rated n % Rated n % Rated n % Rated N % Rated


‘‘True’’ ‘‘True’’ ‘‘True’’ ‘‘True’’ ‘‘True’’

Follows school and classroom rules** 30 16.7 91 17.6 116 37.9 81 29.6 318 28.0
Apologizes (says ‘‘I am sorry’’) when she/he does 29 17.2 95 9.5 119 18.5 83 21.7 326 16.6
something wrong
Possible concernc
Seems sad 29 27.6 91 30.8 121 37.2 84 28.6 325 32.3
Easily gets angry, gets upset, or criesd 30 16.7 93 38.7 122 29.5 83 28.9 328 30.8
Has difficulty separating from parents 29 24.1 92 32.6 119 26.1 83 20.5 323 26.3
Is aggressive toward people or things 29 24.1 94 25.5 121 18.2 84 13.1 328 19.5
Frequently has headaches, stomach ache, or other 30 40.0 93 19.4 121 14.9 84 11.9 328 17.7
physical symptoms (showing signs that she/he
Psychosocial Development of Karen Refugee Children

does not feel good)**


Has difficulty staying with and concentrating on 28 21.4 94 20.2 121 9.1 84 21.4 327 16.5
activities
Does not express feelings or emotions 28 21.4 90 14.4 122 15.6 84 16.7 324 16.0
Has fears* 30 16.7 94 22.3 120 10.8 84 9.5 328 14.3
Is rejected by other children 30 10.0 93 18.3 117 16.2 82 8.5 322 14.3

Difference by age statistically significant * p \ 0.05, ** p \ 0.01, *** p \ 0.001


a
Response choices were ‘‘False,’’ ‘‘Sometimes true, somewhat true, or requires assistance or encouragement,’’ and ‘‘True’’
b
This item had an additional choice ‘‘The child does not participate in many games with rules.’’ If this choice was selected, the item was counted as missing
c
Response choices were ‘‘False,’’ ‘‘Sometimes or somewhat true,’’ and ‘‘True’’
d
This item is not a part of the possible concern scale
295

123
296 A. Tanaka

Table 6 Adjusted mean scores by length of time in nursery school (adjusted by age)
Time in nursery school

Up to one year More than one year

n Mean SE n Mean SE

Psychosocial developmenta,b 140 2.17 0.03 191 2.24 0.02


Possible concernc,d 139 1.75 0.04 191 1.80 0.03
a
Response choices were 1 = False, 2 = Sometimes true, Somewhat true, or requires assistance or
encouragement, and 3 = True. Higher scores indicate higher level of development
b
Effect of time in nursery school F(1,328) = 4.070, p = 0.044
c
Response choices were 1 = False, 2 = Sometimes or somewhat true, and 3 = True. Higher scores
indicate higher levels of behavior/characteristics that are of possible concern
d
Effect of time in nursery school F(1,327) = 0.997 (n.s.)

olds), show empathy and caring for others (two and five year olds), not say things
that may embarrass or hurt others (three and five year olds), and put away materials
or toys after using them (three, four, and five year olds). The difference between
groups was statistically significant at the p \ 0.05 level for not embarrassing others
for three year olds, and for three to five year olds for putting away materials after
using them.
On the other hand, a larger percentage of the children who were newer in the
nursery schools were found to be able to control their feelings when they could not
get their way (two and five year olds), follow teachers’ directions (two and
four year olds), greet adults that they knew (three year olds), follow school and
classroom rules (four year olds), and apologize when they did something wrong
(two year olds). The difference was statistically significant for following teachers’
directions for four year olds.

Discussion

Validity of the Checklist

The purpose of this study was to conduct a culturally appropriate assessment of


psychosocial development of children in Karen refugee camps in Thailand. As the
assessment tool was being developed, care was taken to ensure the behaviors it
assessed were culturally meaningful and that the items were understandable by the
nursery school teachers, who were refugees themselves. The first step in the analysis
was to examine the validity of the data on the basis of whether or not expected age-
related increases in positive psychosocial skills were observed.
The results showed that scores on the psychosocial development scale increased
with age across both genders, which demonstrates the validity of the checklist. All
but two of the individual items also showed improvement by age. One item that did
not follow the expected age-related pattern, ‘‘Appropriately asks for help if she/he
needs it,’’ showed an increase from three to five year olds, but the percentage rated

123
Psychosocial Development of Karen Refugee Children 297

Table 7 Percentage of children whose teachers gave them the highest rating (‘‘True’’), by length of time
in nursery school (psychosocial development items for which the difference was 10 % or more)
Item Items for which outcomes were better for Items for which outcomes were worse for
those with longer attendance those with longer attendance

Age Time in nursery school Age Time in nursery school

Up to One year or Up to One year or


one year more one year more

n % n % n % n %
Rated Rated Rated Rated
‘‘True’’ ‘‘True’’ ‘‘True’’ ‘‘True’’

Follows rules when Two years 22 27.3 7 42.9


playing games with old
othersa Five years 18 44.4 63 58.7
old
Participates in activities Four years 33 42.4 88 52.3
with other children old
that require
cooperation and
working together
(e.g., pretend play
with other children,
building something
with blocks together)
Can talk about the kind Five years 19 31.6 65 49.2
of feelings she/he is old
experiencing (e.g., ‘‘I
am mad’’ ‘‘I am sad’’)
Shows empathy and Two years 23 21.7 7 42.9
caring for others old
Five years 19 26.3 64 42.2
old
Does not ask questions Three years 64 21.9 31 45.2
or statements that old*
might embarrass or Five years 19 26.3 65 36.9
hurt others old
Puts away materials or Three years 62 27.4 31 48.4 Two years 23 21.7 7 0.0
toys after using them old* old
Four years 34 17.6 87 39.1
old*
Five years 19 21.1 65 47.7
old*
When she/he cannot get Two years 20 25.0 7 14.3
something she/he old
wants or is not Five years 19 68.4 63 46.0
allowed to do what old
she/he wants to do,
she/he is able to
control her/his anger
or hurt feelings
Follows teachers’ Two years 21 38.1 7 14.3
directions old
Four years 34 61.8 85 38.8
old*

123
298 A. Tanaka

Table 7 continued
Item Items for which outcomes were better for Items for which outcomes were worse for
those with longer attendance those with longer attendance

Age Time in nursery school Age Time in nursery school

Up to One year or Up to One year or


one year more one year more

n % n % n % n %
Rated Rated Rated Rated
‘‘True’’ ‘‘True’’ ‘‘True’’ ‘‘True’’

Greets adults when she/ Three years 63 33.3 31 22.6


he sees someone she/ old
he knows
Follows school and Four years 33 45.5 83 34.9
classroom rules old
Apologizes (says ‘‘I am Two years 22 22.7 7 0.0
sorry’’) when she/he old
does something
wrong
Participates in games or Three years 64 39.1 31 51.6 Two years 23 43.5 7 28.6
group activities with old old
other children
Appropriately asks for Three years 64 39.1 31 54.8 Two years 22 54.5 7 42.9
help if she/he needs it old old
Five years 19 36.8 65 53.8
old
Says good words, not Two years 22 50.0 7 71.4 Four years 34 50.0 87 35.6
arguing, yelling, or old old
shouting Five years 19 52.6 65 40.0
old

Response choices were ‘‘False,’’ ‘‘Sometimes true, somewhat true, or requires assistance or encouragement,’’ and
‘‘True’’
Difference between time in nursery school categories was statistically significant at * p \ 0.05 level. The chi-squared
test was used for most comparisons. Fisher’s exact test was used when more than one cell had n of less than 5
a
This item had an additional choice ‘‘The child does not participate in many games with rules.’’ If this choice was
selected, the item was counted as missing

‘‘True’’ for two year olds was higher than for five year olds. This pattern may be
because of a shifting standard for what is ‘‘appropriate’’ by age of children, and the
item would benefit from clearly stating behaviors that are considered ‘‘appropriate.’’
The result for the second item, ‘‘Says good words, not arguing, yelling or
shouting,’’ showed that two year olds had the highest percentage rated ‘‘True’’, with
a decreasing trend up to four year olds. This is surprising, because, as expressed in
the phrase ‘‘terrible twos’’ in the Western world, two year olds are not known for
politeness, but for outbursts and tantrums, considered a reflection of a developing
need for autonomy. However, the appearance of ‘‘terrible twos’’ is not culturally
universal (Mosier and Rogoff 2003), and this may not be a predominant
characteristic for two year olds in Karen culture (M. Thitiphanawan, personal
communication, August 18, 2011).
There was no a-priori hypothesis for the relationship of behaviors of concern to
age. In the young age range examined, the overall results showed that scores on the

123
Psychosocial Development of Karen Refugee Children 299

possible concern scale decreased with age. This may be because, although
consistent exhibition of these behaviors is of concern, young children are still
developing skills to manage such behaviors (e.g., separation anxiety, being able to
control feelings).
The number of children with frequent illnesses decreased with age. This item was
originally included as a potential measure of depression in childhood, and it did load
on the possible concern factor. However, the clear pattern of age-related decrease
seen for this item, especially between two and three year olds, was not observed for
other items related to depression or emotional distress. This suggests that this item
actually reflected physical health issues and developing immunity as children grew
older. The item ‘‘Easily gets angry, gets upset, or cries,’’ meant to measure a
possible emotional problem, did not statistically fit the possible concern scale. It
may be that the teachers did not interpret this as a negative behavior, or saw this as a
concern but its appearance did not correlate with other behaviors of concern.

Psychosocial Development

As discussed above, the scale level analysis showed that as their age increased,
children in the nursery schools displayed more proficiency in behaviors reflecting
positive psychosocial development. In reviewing the results on individual items, it
should be noted that the percentage rated as ‘‘True’’, including only the children
who exhibited behaviors consistently without assistance or encouragement, was
examined. Using that stringent criterion, half to two-thirds of five-year old children
were found to be proficient in participating in games, group activities, and activities
that require cooperation, following rules when playing games, controlling self when
frustrated, and appropriately asking for help. This seems to indicate that children in
the nursery schools were developing well in cooperation with other children, which
the curriculum actively promotes. In contrast, a third or fewer of five year olds had
mastered behaviors that involve politeness (apologizing, greeting adults) and
following school and classroom rules. For all these behaviors, results were worse for
children who were in nursery schools for a longer period of time than for children
who were newer to the schools, as discussed in the section on effect of nursery
schools on psychosocial development.

Behaviors of Concern

The scores for the possible concern items showed that more than 30% of the
children seemed sad or got angry/upset or cried easily. Again, these are the children
for whom teachers rated these characteristics as ‘‘True’’, not sometimes or
somewhat true. Thus, sadness or emotional outbursts can be regarded as pervasive
characteristics of these children. Anger, upset, and crying may also be a reflection of
irritability, often a symptom of depression in young children (American Psychiatric
Association 2000; Office of the Surgeon General 1999) or a sign of general stress.
Sadness, depression, and emotional problems have been found at higher prevalence
among refugee children, although evidence has mostly been with older children
(Bronstein and Montgomery 2011; Mollica et al. 1997). These results indicate that

123
300 A. Tanaka

depressive symptoms are also common among younger children in long-term


refugee situations.
Difficulty separating from parents was common, with 26 % of all children having
difficulty, peaking at 33 % for three year olds and 21 % even among five year olds.
Although separation anxiety is common among young children, it typically does not
continue for a prolonged period in familiar settings, for example day care or
preschool where children attend every day. However, cultural differences in the
display of separation anxiety have been found, with young children from cultures
where closer proximity and physical contact to parents are the norm exhibiting more
anxiety and distress than children from western cultures (Miyake et al. 1985). This
may also be the case in Karen culture. Examination of turnover and work shifts of
teachers at nursery schools may also shed light on whether children are having the
opportunity to develop secondary attachment to their teachers (Bowlby 2007).
On the other hand, trauma experience in childhood has been found to heighten
anxiety reactions in later years (Heim and Nemeroff 2001); thus, heightened
separation anxiety may be a reflection of an effect of trauma in their lives. Parents
with unresolved trauma or loss may also behave in ways that make it difficult for
their children to develop secure attachment to them and to feel safe while being
away from them (Cassidy and Mohr 2001; Schuengel et al. 1999). Thus, even if the
children have not been exposed to traumatic experiences themselves, they can be
indirectly affected by their parents’ experiences.

Effect of Nursery Schools on Psychosocial Development

The adjusted mean score on the psychosocial development scale was higher for
children who attended the nursery schools for a longer period of time than for
children who had attended nursery schools for one year or less. This result is
consistent with the idea that attending nursery schools facilitated psychosocial
development of the children. More positive findings for children who attended
nursery schools longer were found for following rules when playing games,
participating in activities that require cooperation, being able to talk about feelings,
showing empathy and caring for others, not saying things that may embarrass or hurt
others, and putting away materials or toys.
In refugee camps, children live in poverty in a crowded environment, may not be
supervised adequately or may be exposed to behaviors of adults around them who
suffer from the stress of a long-term stay in refugee camp and effects of violence
which led to their flight. The nursery schools provide a safe place for children to be,
where psychosocial development is promoted by positive interaction with teachers
and group play and cooperation in a peaceful environment. These results, showing
increased cooperativeness with peers and awareness of their own and others’
feelings for the children who had attended nursery schools for a longer period of
time, are consistent with the focus of the curriculum. These outcomes are
particularly significant in the context affected by violence in that these skills can be
building blocks for peaceful conflict resolution (Sandy and Cochran 2000).
On the other hand, some items were more frequently recorded for children who
were relatively new to the nursery schools. Children who had attended nursery

123
Psychosocial Development of Karen Refugee Children 301

schools for one year or less were found to be more likely to greet adults that they
knew, apologize when they did something wrong, follow school and classroom
rules, follow teachers’ directions, and be able to control their feelings when
frustrated. These items have a theme of politeness, obedience, and constriction. It
may be that these children are still in the process of getting comfortable in the
nursery school environment and cannot express their feelings freely. In addition,
they may be accustomed to teachers as authoritarian figures from their experience in
Burma or with adults in the camp community in general, compared with more
authoritative environments of child-centered nursery schools. Thus, they are more
likely to behave politely, not act out their feelings, and follow rules and teachers’
directions.
The skills covered by these items were considered important for the children’s
success in life and classroom by teachers or Karen NGO staff, who gave input on
the items based on this dimension during the checklist development process. It may
be that although the Karens who gave input believed that these were important skills
for success, the western-influenced curriculum that they followed did not promote
them, at least not as much as it did other skills, or as much as more traditional
cultural values emphasized them. These results raise an important question of
whether the nursery school curriculum adequately supports all important aspects of
psychosocial development of Karen children. The nursery school program may
benefit from a discussion between the teachers and the NGO, which provides
guidance on the schools’ curriculum, around the skills that children who were in the
program for a longer time did not show increased proficiencies in. The discussion
should focus on the importance of these behaviors in the Karen community and for
the future success of children, whether they are given enough attention in the
classrooms, and ways to promote these behaviors in the curriculum.
A longer stay in nursery schools was not related to a reduction in problem
behaviors. Although secure, warm, and supportive environments can have a positive
effect on children with emotional and behavioral issues, the program was not
intended to treat mental health or behavioral problems. However, this type of
program in a refugee camp setting would benefit from training to increase teachers’
awareness of and skills to work with children who have behavioral/emotional issues.
In addition, they should have consultative access and relationships with mental
health specialists in the camps so they can refer or seek support for children who
exhibit serious behavioral issues or signs of mental health problems.
In interpreting these results based on the length of attendance at nursery schools,
some limitations and confounding factors should be noted. Differences observed
among children may be because of factors other than attendance at nursery schools.
We must ask the question: why did some children not start attending nursery schools
until recently, whereas other children have been attending for a long time? It may be
their parents were not aware of the existence of nursery schools or their benefits,
which may indicate that the parents may be less connected to community resources
in general or less educated about children’s needs. It may be that the children’s
parents were not working so they did not feel the need to bring their children to
others’ care, which may mean that the family had enough financial support from
extended family or they were more economically deprived than families with

123
302 A. Tanaka

working parents. It may be that the children who just started attending were new to
the camp. In that case, trauma that led to their flight from their home villages was
more recent.
Unfortunately, because of the program’s need to collect information from
teachers with the least amount of burden and without access to children’s parents,
additional information such as arrival date at the camp and family situations could
not be collected. However, different effects of attendance on psychosocial
development (an area that teachers were trained to promote) and emotional/
behavioral issues (an area that teachers were not trained to affect), supports the
program theory and is likely to indicate that nursery schools did have an intended
positive effect on children. In addition, characteristics reflected in the items that
were rated higher for children who attended nursery schools for a longer period of
time versus those rated higher for the children who were relatively new were
consistent with the effect of the nursery school curriculum versus the effects of
larger culture in Burma and the camps.

Limitations of the Study

As discussed above, restriction of data collection to a short and simple questionnaire


that could be completed by teachers limited the scope of information that could be
collected for this study. Information on trauma history, date of arrival at camp, and
family environment would have greatly deepened interpretation of the results, but
was beyond the capacity of the program to collect. Items asking about commonly
observed symptoms among children who had experienced trauma, for example night
terrors, also could not be included in the checklist, because teachers did not have
reliable information on such incidents without asking parents. Sample size also
restricted the number of comparisons and variables that could be included in
statistical testing. Increasing the number of checklists collected to more than two per
teacher would have increased their burden and not been feasible.
In addition, other studies examining reports of children’s behavior have shown
different patterns of results based on type of informant (Kerestes 2006). Thus,
children’s behaviors described in this study should be understood as reflecting
observations by teachers in the nursery school setting, which may be different from
their behavior at home as seen by parents, or in other settings.
The development of a psychosocial checklist through iterations of input from
Karen NGO staff and nursery school teachers is a strength of this study in terms of
ensuring cultural appropriateness. At the same time, use of a unique instrument
without a norm or comparison data limits the interpretation of the results in that it
does not tell how much of the positive development and the possible concerns are
due to typical child development. Comparison with Karen children who live in
settings not affected by conflict would be a worthwhile project for the future.
Similarly, it should be noted that use of the highest rating (‘‘True’’) for items on the
possible concern scale does not indicate the behaviors are of clinical significance.

123
Psychosocial Development of Karen Refugee Children 303

Conclusion

Refugee children face many risks of compromised psychosocial development,


including their trauma history and parents who are potentially affected by their own
trauma, mental health issues, and the stress of lack of control over their lives. With
two-thirds of the refugees in the world living in protracted exile, more information
on psychosocial development of young refugee children in refugee camps, and
primary prevention programs to support their healthy development is needed.
This study examined the psychosocial development of children in the nursery
school program in Karen refugee camps in Thailand. Results of the assessment using a
psychosocial checklist developed for the Karen population demonstrated that the
psychosocial development of two to five-year-old children in nursery schools
increased with increasing age. Despite potential unmeasured confounders, nursery
schools are likely to be making a positive difference in the psychosocial development
of young children in refugee camps, especially in the areas of cooperation with peers
and awareness of their own and others’ feelings. A significant number of children had
emotional issues, and there was no evidence that attendance in nursery schools
ameliorated behaviors of potential concern. Because nursery schools can be an ideal
setting for early identification, referral, and intervention for potential concerns,
additional teacher training in mental health and establishment of connections with
mental health specialists in the camps is recommended.

Acknowledgments The nursery school program and program evaluation were supported by the Bernard
van Leer Foundation. I would like to thank the staff of Taipei Overseas Peace Service for supporting the
development of the psychosocial checklist and collecting input and final data from the teachers. Special
appreciation goes to Muetae Thitiphanawan for providing information on the program, camp situation,
and Karen culture. I also express gratitude to all the nursery school teachers for their great work with
children and for completing the checklist for the evaluation. I would also like to thank Dr Charles Oberg
for his support and feedback and Anne Maertz for her review and copy editing of the manuscript.

References

Alaimo, K., Olson, C. M., & Frongillo, E. A, Jr. (2001). Food insufficiency and American school-aged
children’s cognitive, academic, and psychosocial development. Pediatrics, 108(1), 44–53.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders
(revised) (4th ed.). Washington, DC: American Psychiatric Association.
Anders, K. E. (1980). Deference for the elders and control over the younger among the Karen in
Thailand. Retrieved May 11, 2012, from http://www.ibiblio.org/obl/docs/KirstenEA-Deference.pdf
Bassuk, E. L., & Rosenberg, L. (1990). Psychosocial characteristics of homeless children and children
with homes. Pediatrics, 85, 257–261.
Bombay, A., Matheson, K., & Anisman, H. (2009). Intergenerational trauma: Convergence of multiple
processes among First Nations people in Canada. Journal of Aboriginal Health, 5, 6–47.
Bowlby, R. (2007). Babies and toddlers in non-parental daycare can avoid stress and anxiety if they
develop a lasting secondary attachment bond with one carer who is consistently accessible to them.
Attachment & Human Development, 9(4), 307–319.
Bronstein, I., & Montgomery, P. (2011). Psychological distress in refugee children: A systematic review.
Clinical Child and Family Psychology Review, 14, 44–56.
Bureau of Democracy, Human Rights, and Labor. (2007). 2006 country reports on human rights
practices: Burma. Retrieved July 23, 2011, from http://www.state.gov/g/drl/rls/hrrpt/2006/
78768.htm

123
304 A. Tanaka

Bureau of Democracy, Human Rights, and Labor. (2011). 2010 country reports on human rights
practices: Burma. Retrieved July 23, 2011, from http://www.state.gov/documents/organization/
160450.pdf
Burmese Border Consortium. (2004). Twenty years on the border. Bangkok, Thailand: Burmese Border
Consortium.
Cardno Agrisystems Limited. (2009). Vulnerability analysis in burmese refugee camps in Thailand.
Retrieved May 2, 2012, from http://www.tbbc.org/resources/2009-10-livelihoods-vulnerability-
analysis.pdf
Cassidy, J., & Mohr, J. J. (2001). Unsolvable fear, trauma, and psychopathology: Theory, research, and
clinical considerations related to disorganized attachment across the life span. Clinical Psychology:
Science and Practice, 8(3), 275–298.
Dekel, R., & Goldblatt, H. (2008). Is there intergenerational transmission of trauma? The case of combat
veteran’s children. American Journal of Orthopsychiatry, 78(3), 281–289.
Fazel, M., Doll, H., & Stein, A. (2009). A school-based mental health intervention for refugee children:
An exploratory study. Clinical Child Psychology and Psychiatry, 14(2), 297–309.
Flisher, A. J., Kramer, R. A., Hoven, C. W., Greenwald, S., Algeria, M., Bird, H. R., et al. (1997).
Psychosocial characteristics of physically abused children and adolescents. Journal of American
Academy of Child and Adolescent Psychiatry, 36(1), 123–131.
Goldstein, R. D., Wampler, N. S., & Wise, P. H. (1997). War experiences and distress symptoms of
Bosnian children. Pediatrics, 100(5), 873–878.
Habir, E. E., Marriage, K., Littlefield, L., & Pratt, K. (1994). Teachers’ perceptions of maladaptive
behaviour in Lebanese refugee children. Australian and New Zealand Journal of Psychiatry, 28(1),
100–105.
Heim, C., & Nemeroff, C. B. (2001). The role of childhood trauma in the neurobiology of mood and
anxiety disorders: Preclinical and clinical studies. Biological Psychiatry, 49, 1023–1039.
Hindstrom, H. (2012). Attacks continue despite Karen ceasefire. Democratic voice of Burma. Retrieved
June 15, 2012, from http://www.dvb.no/news/attacks-continue-despite-karen-ceasefire/20059
Human Rights Documentation Unit. (2007). Burma human rights yearbook 2006. Retrieved May 9, 2012,
from http://www.ncgub.net/NCGUB/mediagallery/downloadd538.pdf?mid=20070702190012811
Human Rights Documentation Unit. (2009). Burma human rights yearbook 2008. Retrieved July 23,
2011, from http://www.ncgub.net/NCGUB/mediagallery/downloadc516.pdf?mid=20091123192
152709
Karen Buddhist Dhamma Dhutta Foundation. (2010). The Karen people: culture, faith and history.
Retrieved May 11, 2012, from http://www.karen.org.au/docs/Karen_people_booklet.pdf
Karen Human Rights Group. (2012). Abuse since the DKBA and KNLA ceasefires: Forced labour and
arbitrary detention in Dooplaya. Retrieved June 15, 2012, from http://www.khrg.org/khrg2012/
khrg12f2.html
Kerestes, G. (2006). Children’s aggressive and prosocial behavior in relation to war exposure: Testing the
role of perceived parenting and child’s gender. International Journal of Behavioral Development,
30, 227–239.
Kiernan, K. E., & Huerta, M. C. (2008). Economic deprivation, maternal depression, parenting and
children’s cognitive and emotional development in early childhood. The British Journal of
Sociology, 59(4), 783–806.
Kinzie, J. D., Sack, W. H., Angell, R. H., Manson, S., & Rath, B. (1986). The psychiatric effects of
massive trauma on Cambodian children: I. The children. Journal of the American Academy of Child
Psychiatry, 25(3), 370–376.
Lamphear, V. S. (1985). The impact of maltreatment on children’s psychosocial adjustment: A review of
the research. Child Abuse and Neglect, 9, 251–263.
Lustig, S. L., Kia-Keating, M., Knight, W. G., Geltman, P., Ellis, H., Kinzie, J. D., et al. (2004). Review
of child and adolescent refugee mental health. Journal of the American Academy of Child and
Adolescent Psychiatry, 43(1), 24–36.
Macksoud, M. S., & Aber, J. L. (1996). The war experience and psychosocial development of children in
Lebanon. Child Development, 67, 70–88.
Milner, J. & Loescher, G. (2011). Responding to protracted refugee situations: Lessons from a decade of
discussion. Forced migration policy briefing 6. Retrieved August 21, 2012, from http://
repository.forcedmigration.org/pdf/?pid=fmo:5937

123
Psychosocial Development of Karen Refugee Children 305

Miyake, K., Chen, S.-J., & Campos, J. J. (1985). Infant temperament, mother’s mode of interaction, an
attachment in Japan: An interim report. Monographs of the Society for Research in Child
Development, 50(1/2), 276–297.
Mollica, R. F., Poole, C., Son, L., Murray, C. C., & Tor, S. (1997). Effects of war trauma on Cambodian
refugee adolescents’ functional health and mental health status. Journal of American Academy of
Child and Adolescent Psychiatry, 36(8), 1098–1106.
Mosier, C. E., & Rogoff, B. (2003). Privileged treatment of toddlers: Cultural aspects of individual choice
and responsibility. Developmental Psychology, 39(6), 1047–1060.
Office of the Surgeon General. (1999). Mental health: a report of the surgeon general. Retrieved July 29,
2011, from http://www.surgeongeneral.gov/library/mentalhealth/home.html
Osofsky, J. D. (1995). The effects of exposure to violence on young children. American Psychologist,
50(9), 782–788.
Phan, Z., & Lewis, D. (2010). Undaunted: my struggle for freedom and survival in Burma. New York,
NY: Free Press.
Raboteg-Saric, Z., Zuzul, M., & Kerestes, G. (1994). War and children’s aggressive and prosocial
behaviour. European Journal of Personality, 8, 201–212.
Riley, A. W., Coiro, M. J., Broitman, M., Colantuoni, E., Hurley, K. M., Bandeen-Roche, K., et al.
(2009). Mental health of children of low-income depressed mothers: Influences of parenting, family
environment, and raters. Psychiatric Services, 60(3), 329–336.
Sandy, S. V., & Cochran, K. M. (2000). The development of conflict resolution skills in children:
Preschool to adolescence. In M. Deutsch & P. T. Coleman (Eds.), The handbook of conflict
resolution: Theory and practice (pp. 316–342). San Francisco, CA, USA: Jossey-Bass.
Schuengel, C., Bakermans-Kranenburg, M. J., & Van IJzendoorn, M. H. (1999). Frightening maternal
behavior linking unresolved loss and disorganized infant attachment. Journal of Consulting and
Clinical Psychology, 67(1), 54–63.
South, A. (2011). Burma’s longest war: anatomy of the Karen conflict. Amsterdam, The Netherlands:
Transnational Institute and Burma Center Netherlands.
Thailand Burma Border Consortium. (n.d.). Mae Sot area. Retrieved May 2, 2012, from http://
www.tbbc.org/camps/mst.htm
Thailand Burma Border Consortium. (2006). Burmese border refugee sites with population figures:
September 2006. Retrieved July 23, 2012, from http://www.tbbc.org/camps/map-2006-09-sep-
ccsdpt-tbbc.pdf
Thailand Burma Border Consortium. (2007). Programme report: July to December 2006 including
revised funding appeal for 2007. Retrieved May 2, 2012, from http://www.tbbc.org/resources/2006-
6-mth-rpt-jul-dec.pdf
The Karen Women’s Organization. (2007). State of terror: The ongoing rape, murder, torture and forced
labour suffered by women living under the Burmese military regime in Karen state. Mae Sariang,
Mae Hong Son, Thailand: The Karen Women’s Organization.
UNHCR. (2006). 2005 UNHCR statistical yearbook: Thailand. Retrieved July 23, 2011, from http://
www.unhcr.org/4641bec911.html
UNHCR. (2007). 2006 UNHCR global report: Thailand. Retrieved July 23, 2011, from http://www.
unhcr.org/publ/PUBL/4666d2680.pdf
Vaage, A. B., Thomsen, P. H., Rousseau, C., Wentzel-Larsen, T., Ta, T. V., & Hauff, E. (2011). Paternal
predictors of the mental health of children of Vietnamese refugees. Child and Adolescent Psychiatry
and Mental Health, 5(2). Retrieved August 19, 2011, from http://www.capmh.com/content/5/1/2
Yehuda, R., Halligan, S. L., & Grossman, R. (2001). Childhood trauma and risk for PTSD: Relationship
to intergenerational effects of trauma, parental PTSD, and cortisol excretion. Development and
Psychopathology, 13, 733–753.

123
Copyright of International Journal of Early Childhood is the property of Springer Science &
Business Media B.V. and its content may not be copied or emailed to multiple sites or posted
to a listserv without the copyright holder's express written permission. However, users may
print, download, or email articles for individual use.