EXPOSURE TO POLITICAL VIOLENCE AND PSYCHOLOGICAL WELL-BEING IN

Exposure to Political Violence and Psychological Well-being in Bosnian Adolescents: A Mixed Method Approach LYNNE JONES Cambridge University, UK

KONSTANTINOS KAFETSIOS Anglia Polytechnic University, Cambridge, UK

A B S T R AC T Previous research on political violence has shown an inverse relationship between overall exposure and psychological well-being. The aim of this study was to examine this relationship in more detail by exploring the impact of specific types of war-related events upon psychological well-being, as well as the role of social and political context in moderating these outcomes. The psychological well-being of 337 Bosnian adolescents living in two towns on opposite sides of the war was assessed using the Hopkins Symptom Checklist-25 (HSCL-25) and the Harvard Trauma Questionnaire (HTQ). Based on the combined symptom scores, and indepth interviews regarding life history, war experiences, and subjective experience of psychological well-being a gender-matched sub-sample of 40 adolescents was selected and completed a 45-item trauma event scale. Results from the quantitative and qualitative analyses showed that the relationship between exposure, displacement and well-being varied significantly depending on the community in which the adolescents lived. Specific meanings given to different types of war events were important in moderating their effect. Living in a neglected, isolated and depressed community, worry about school performance, missing friends and family breakdown could have as significant an effect on well-being as exposure to war-related events. The findings demonstrate the need to take social context and meaning of events into account when examining the impact of war exposure on psychological well-being. K E Y WO R D S adolescents, Bosnia-Herzegovina, exposure, political violence, psychological well-being

Clinical Child Psychology and Psychiatry Copyright © 2005 SAGE Publications (London, Thousand Oaks and New Delhi) Vol 10(2): 157–176. DOI: 10.1177/1359104505051209 www.sagepublications.com 157

CLINICAL CHILD PSYCHOLOGY AND PSYCHIATRY 10(2)

I heard some shooting when I woke up one morning. I asked my mum ‘what’s happening?’ and she told me that the war had started . . . When it got dark, me and my family ran away to a safer place. When we got there my father had to go and fight against the Chetniks. Then a few days later I was told that my father had been killed. That was so hard for me, and also the house we lived in did not have any windows or doors. . . . There were 31 of us living there. Eight shells hit the roof of that house. We stayed four months and then we came back home. The war went on for another four and a half years. Peace coming made no difference to me. (14-year-old boy, Bosnia-Herzegovina) I T I S well accepted that exposure to political violence may have a negative effect on psychological well-being. One of the problems in understanding exposure is that wars are complex chronic affairs with enduring impact on all aspects of life long after the conflict has ended, as the above story illustrates. This article explores the nature and significance of exposure to specific kinds of war-related events from the perspective of young adolescents, examines their differing impact on post-war recovery, and how this is influenced by the social and political context. A growing body of research suggests a ‘dose–effect’ relationship between exposure and well-being: the greater the exposure to political violence, the more psychological problems are experienced (Almqvist & Brobert, 1999; Bryce, Walker, Ghorayeb, & Kanj, 1989; Farwell, 1999; McCallin, 1998; Mghir, Freed, Raskin, & Katon, 1995; Mollica, Poole, Son, Murray, & Tor, 1997; Paardekooper, de Jong, & Hermanns, 1999; Qouta, Punamäki, & Sarraj, 1995b; Thabet & Vostanis, 1999). There is also consensus that although the effects may diminish with length of time from exposure (Foy, Madvig, Pynoos, & Camilleri, 1996), some are remarkably enduring (Mollica et al., 1997; Sack, Clarke, & Seeley, 1996; Sack, Him, & Dickson, 1999; Savin, Sack, Clarke, Meas, & Richart, 1996). However, a small study of Central American immigrants and their children found no relationship between the amount of violence witnessed and symptoms experienced (Locke, Southwick, McCloskey, & Fernandez-Esqier, 1996). Ziv and Israeli (1973) found that children from bombarded kibbutzim were no more anxious than children from non-bombarded ones. Such discrepancies have drawn attention to the factors influencing the relationship between exposure and response. The way in which a young person makes sense of events may be one of the most significant. One group of studies has explored the child’s

LY N N E J O N E S , O B E , M A , M B C H B , M R C P S Y C H , P H D is a Child and Adolescent Psychiatrist working for International Medical Corps, an international humanitarian relief organization, in conflict and post-conflict areas. She is also a senior research associate at the Centre for Family Research, Cambridge University, working on children’s understanding of political violence. She is currently engaged in establishing a pilot community-based mental health service in Sierra Leone. C O N TA C T : Lynne Jones, Centre for Family Research, Social and Political Sciences Faculty and Department of Developmental Psychiatry, Cambridge University, Free School Lane, Cambridge University, CB2 3RF, UK. E-mail: [email protected] K O N S TA N T I N O S ( K O S TA S ) K A F E T S I O S is a Social Psychologist with research interests in theoretical and applied areas of emotion, close relationships and well-being. He is currently at the Psychology Department at the University of Crete and has previously held appointments at Anglia Polytechnic University and the University of Cambridge.

158

JONES & KAFETSIOS: EXPOSURE TO POLITICAL VIOLENCE

interpretations of the whole event and has found that stronger ideological convictions and greater participation may contribute to greater exposure but lessen its impact through the development of active coping skills, at least in low-level conflicts (Dawes & De Villiers, 1987; Kostelny & Garbarino, 1994; Protacio-Marcelino, 1989; Punamäki, 1996; Punamäki & Suleiman, 1990; Qouta, Punamäki, & El Sarraj, 1995a; Straker, Mendelsohn, Moosa, & Tudin, 1996). Also, studies have examined the subjective significance of specific types of exposure. Several authors have demonstrated that specific war-related events have specific outcomes, both in terms of enduring symptoms and psychosocial adaptation. Posttraumatic symptoms tended to associate with memories of specific war traumas and may endure many years after the event. Depressive symptoms have been linked with loss, separation from parents and more recent stressful events such as refugee camp life or the threat of repatriation (Macksoud & Aber, 1996; Sack et al., 1996, 1999; Savin et al., 1996). Most of these studies used a quantitative approach to demonstrate the strength of particular relationships between specific events and specific outcomes. Many authors conducted the analyses by asking their respondents to identify their worst traumatic experience, and when it occurred. They then explored the impact of this event (Macksoud & Aber, 1996; Sack et al., 1999; Savin et al., 1996). Although these studies have drawn out the relative subjective importance of various events, at least as they relate to psychopathology, they cannot provide the respondents’ view of why one event is so significant, how it fits in the context of their whole lives, or the young person’s understanding of the relation between event and outcome. The authors provide interpretations as to why these relationships might exist, but the child’s voice is missing. Social and political contexts have also received some attention in most studies of exposure. Some studies involved area comparisons between communities with different levels of exposure (Elbedour, Ten Bensel, & Bastien, 1993; Walton, Nuttall, & Nuttall, 1997; Ziv & Israeli, 1973), or different demographic characteristics (Adjukovic & Adjukovic, 1993; Macksoud & Aber, 1996). Other studies have compared displaced with resident families (Kuterovac, Dyregrov, & Stuvland, 1994), encamped refugees with resettled ones (Savin et al., 1996). Straker et al. (1996) compared black youth at three different points in a black township’s history and found the meaning given to specific events shifted as the political context changed. Violence was only cited as the worst problem in periods when it was primarily intracommunal with no political objective. There have been few other studies comparing the impact of political context on perceptions of exposure in this manner, and very few comparing neighbouring adolescents on opposing sides of a war (Apfel & Simon, 2000). The present study is part of a larger ethnographic investigation of Bosnian adolescents living in two neighbouring cities on opposite sides of the conflict. A previous article from this study demonstrated that the amount of exposure to war-related events related to adolescent well-being, and to the degree to which they sought meaning for the conflict, depending on the context (Jones, 2002). The aim of this article is to expand on the role of exposure in three ways: first, to explore the complexity of political violence in the Bosnian war and create a typology of the specific experiences of political violence; second, to examine the role of social and political context in generating specific meanings and outcomes (i.e. do adolescents on opposite sides interpret similar events in different ways?); and third, to examine the relationship between specific types of war-related events and psychological well-being from the perspective of the adolescent. The methodological stance is primarily qualitative but the data gathered have also made it possible to carry out quantitative analyses. This mixed method approach, admittedly rare (e.g. Tashakkori & Teddlie, 1998), allows for an effective exploration of 159


contextual mediators of the relationship between political violence and psychological well-being.

Background on the conflict The war in Bosnia-Herzegovina began in 1992 and ended with the 1995 Dayton Peace Agreement. Approximately one million people were displaced and 250,000 were killed, the majority of them civilians. Foča and Goražde are neighbouring towns in south-east Bosnia. Prior to the war, both had ethnically mixed communities with good relations.

Focˇa Serb military forces took over Foča town in 10 days and the non-Serb population was forced to flee or was detained under appalling conditions (Human Rights Watch, 1998; The International Criminal Tribunal for the Former Yugoslavia, 1999). Most Serbian adolescents were out of town during the fighting and returned within the year. Material living conditions were harsh and school was intermittent. In 1995, 3 days of NATO air strikes destroyed half of the town’s bridges but caused no civilian casualties in this area. Shortly after the Dayton Peace Agreement, many Bosnian Serbs, believing they would not be secure in territory given to the other side, came to Foča. Because (in 1997) the local authorities continued to oppose the return of non-Serbs to the town, and some indicted war criminals remained publicly active, there was little input of international aid, and little rehabilitation.

Gorazˇde Most of the Muslim adolescents in the study were in the town throughout the war, under siege for almost 4 years, suffering intermittent shelling and sniping and dependent on airdrops of humanitarian aid or what could be procured locally. The Dayton Peace Agreement kept Goražde as a pocket of land belonging to the Federation, almost entirely surrounded by Republika Srpska. At the time of the study, feelings of geographical and social isolation were still present and economic difficulties persisted, although with the help of international donors, much of the town was being rebuilt.

Methods The study began 2 years after the military conflict finished. It was conducted in parallel in both cities between October 1997 and July 1998. Ethical approval for the project was granted by Cambridge University Psychology Research Ethics Committee.

Sample Participants The main sample consisted of 337 adolescents (54.3% from Goražde) aged between 13–15 years, from four schools (one high school and one primary school in each town). Headmasters in all the schools gave permission and allocated classes according to timetable convenience. There were no substantial differences with other classes of the same grade. All the adolescents were physically well and apart from one girl, had not been in any psychological support programme. Measures Well-being self-report checklists

All participants completed the Hopkins Symptoms Checklist (HSCL-25 composed of 10 questions on anxiety, and 15 items on depression) 160


and the 16-item trauma symptom component of the Harvard Trauma Questionnaire (HTQ; Mollica et al., 1991). The scales were chosen because they had already been shown to be meaningful to a population exposed to long-term political violence in Indochina (Mollica et al., 1992). The range and expression of symptoms matched those presented to the primary researcher when she worked as a clinical psychiatrist in Goražde in the 6 months preceding the study. They were relevant to these communities and had content validity. The scales had already been translated and back-translated for a Bosnian population. Local interpreters in both cities back-translated them again to ensure semantic validity. As a result of a pilot study with 100 adolescents, the question relating to loss of sexual interest or pleasure was eliminated. The adolescents completed the questionnaire in supervised class time. All participated anonymously and any that wished could abstain. None did. One adolescent’s questionnaire was discarded as unusable.

Story writing All the children were asked to write stories entitled ‘What War Means to Me and How it Has Affected My Life.’

Sub-sample Selection

After the checklist scores had been summated, a gender-matched sub-sample of 40 was selected (20 from each town consisting of the 10 highest and 10 lowest scoring adolescents). Each selected adolescent and their parents received an explanatory letter and invitation and met with the researcher to have the study explained. The school psychologists assisted with the selection to enable the primary researcher to remain blind to the adolescent’s symptom score until the end of the study. If a child did not wish to participate, the child of the appropriate gender in that school with the next score up or down was invited. In all there were 47 refusals, 39 of whom were low scorers, and the majority of whom were boys (29), in both cities. The adolescents who refused to participate were somewhat less symptomatic than the group finally selected but this was not a significant difference (t(1,44) = 2.8, ns). Both the 20 high and the 20 low scorers of the final sub-sample fell within the highest and lowest quartiles of the main sample distribution.

Measures The Exposure to War-Related Events Questionnaire (EWRE) In order to cover the full range of events experienced by adolescents in Bosnia, a 45-item questionnaire was devised, drawing on both Macksoud’s Childhood War Trauma questionnaire (Macksoud, 1992) and the trauma event component of the HTQ (Mollica et al., 1991). The main source of events derived from reading literature on the Bosnian war, the primary researcher’s personal experience of the conflict, and interviews conducted by her over the previous 6 years with the Bosnian community in the UK and the Balkans. Each of the 45 items were coded on three dimensions as demonstrated in Table 1; specifically: (a) whether or not the event occurred; (b) frequency: how many times a particular event occurred; (c) intensity: derived from scoring the longest time period an adolescent had to endure a particular event, such as being bombed, or having father absent. For the small number of events such as the death or injury of a person, where duration can provide no indication of intensity, adolescents were scored on whether they witnessed the event or not. Each adolescent’s scores were then summated for each dimension to provide measures 161


Table 1. Dimensions used for scoring exposure to war-related events

Type of event

(a) Event occurred or not

(b) Frequency: number of times event occurred

(c) Intensity: Longest duration of event or whether a death/injury witnessed

e.g. Bombing from the air

No = 0 Yes = 1

<2=0 >2=1 > 10 = 2 > 50 = 3 > 100 or ‘so often that could not remember’ = 4

< 2 days = 0 > 2 days = 1 > 6 months = 2 > 1 year = 3 > 2 years = 4 OR Not witnessed = 0 Witnessed = 1

of the number of events to which a child was exposed, the overall frequency of events, and their overall intensity. The dimensions of frequency and intensity were added to the commonly used event scale because it was thought that there might be significant differences between adolescents shelled 1 or 50 times. Adolescents who had endured particularly intense experiences (i.e. weeks in a shelter at one time or actually witnessing a death) might differ from those whose experiences had been brief or distant from them. The EWRE was administered to the 40 adolescents in the sub-sample, usually towards the end of their first interview. In order to ensure full understanding, the primary researcher asked the questions and filled in results.

School marks The adolescent’s end of year grade average was taken as a measure of social function (5 = Excellent, 1 = Failure).

Interviews with adolescents

Each adolescent was interviewed twice for approximately 2 hours on each occasion. The protocol used was adapted over the course of the fieldwork in response to the developing analysis. In all cases, the topics covered included life history, the war experiences of self and family, and subjective perceptions of well-being. In order to allow the adolescents to frame their responses to the war in their own terms, there was no formal clinical interview. The interview also included open-ended questions on well-being followed up with more in-depth clinical exploration if this seemed warranted.

Interviews with parents Shorter interviews were conducted with two thirds of the parents in the family home to assess their perceptions of the impact of the war on their child’s well-being and their understandings of events. The lifeline

The adolescents drew a lifeline as a non-verbal method of exploring their past and present subjective sense of psychological well-being in relation to various life events. This is an adaptation of an instrument developed by Gergen (1988). A grid was constructed with the x-axis representing calendar dated years in the adolescent’s life and the y-axis representing overall psychological well-being expressed as a percentage (100% represented ‘completely well’ and 0% represented ‘completely unwell’). They placed dots on the grid showed how well they remembered feeling for each year of their lives, from 5 years old. These were then joined to produce a lifeline. This allowed the less articulate adolescents to show the relative differences in the impact of various kinds of events including pre- and post-war events. They were also scored according to the position of 162


the most recent dot, in relation to the best year of their lives. They received a score of 1 if it was in the same position. A score of 2 indicated a less than 25% drop from their best year. A score of 3 indicated a more than 25% drop from their best year.

Participant observation

In addition to the formal interviews, the primary researcher was engaged in the life of both communities for a year. This allowed for participation in all the activities of normal life, and observation of the adolescent in their own environment. Time was also spent observing in school.

Data analysis Qualitative data analysis followed the basic principles of grounded theory (Glaser & Strauss, 1967; Strauss & Corbin, 1997). ‘Rather than working from a priori assumptions’ (Filstead, 1971, p. 6) theory was ‘generated in the course of close inspection and analysis’ (Henwood & Pidgeon, 1992, p. 103). Analysis of the data began while in the field. On leaving the field all the materials were transcribed, checked and tapes were listened to again. Notes were made as to emotional tone and narrative style as well as to content. There was constant comparative analysis: that is coding the data with a continuous alertness to the similarities and differences that exist between instances. This allows recurrent themes and clusters of responses to emerge leading to the creation of typologies and case examples to illustrate particular theories (Mastnak, 1995; Orona, 1997). For example, initial readings of transcripts showed the importance of exposure in one town but not the other, yet there seemed to be equal numbers of less well adolescents in each city. This necessitated a return to text to look at the specific interpretations of events that occurred in both cities, such as shelling and loss. Computer software for qualitative analysis (Nud*ist 4) was used. Psychological well-being Adolescents were classified as psychologically ‘well’ or ‘less well’ on the basis of four qualitative assessment measures: • • • •

the adolescent’s subjective assessment non-verbally expressed by the lifeline; the primary researcher’s impression drawn from observation and interview; parents’ and/or teachers’ view; school marks as a measure of function.

Particular importance was given to a negative subjective view. If adolescents gave themselves an end point of 3 on the lifeline, they were categorized as ‘less well’ regardless of other measures. If they gave themselves an endpoint of 1, they were only categorized as ‘less well’ on the basis of at least two of the other three measures. Thus poor school marks, parental anxieties or researcher impression, on their own, were not enough to classify an adolescent as ‘less well’. The primary purpose of using symptom rating scales in this study was to identify adolescents for the sub-sample. Because the symptom scales had not been tested for criterion validity with this population, and because of concerns about the specificity and sensitivity of such scales when used on their own (American Association of Child and Adolescent Psychiatry, 1998; Perrin, Smith, & Yule, 2000; Richman, 1993; Sack, Seeley, & Clarke, 1997) they were not intended as a means of assessing well-being. Recent analysis of the scales has shown that the two symptom checklist items have good internal consistency and discriminant validity. In 9/40 of cases the reported presence or absence of symptoms did not correspond to the well-being of the adolescent assessed by qualitative means. However, 76.2% of adolescents identified as ‘well’ by qualitative means had low scores on the combined anxiety/depression and trauma checklists; 76.5% of adolescents identified as ‘less well’ had high scores on the combined lists. This suggests 163


Table 2. Numbers of adolescents reporting war-related events (EWRE) ranked by order of frequency Event

N

Event

N

Separated from father Leave home – proximity of fighting Missing school Live in shelter Sniper fire at close distance Extended family member/friend killed Bombing from air Air raid Extended family member/friend injured Shelling at close distance Home attacked or shelled Separated from mother Separated from important relative/friend School attacked or shelled Siege Leave for another country Close family member injured Extended family member intimidated/kidnapped

39 36 32 29 27 27 25 23 23 19 18 17 17 16 16 14 14

Pet killed 11 Live in the open 7 Live in shelter – w/o parents 7 Suffered serious physical injury 7 Deprived of basic foods 7 Leave country w/o parents 6 Family member intimidated/kidnapped/detained 6 Leave home – forced by others 5 Victim of intimidation/arrest/kidnap 5 Cold 5 Close family member killed 5 Extended family member/friend tortured/beaten/abused 4 Armed combat at close distance 3 Leave home – damaged 1 Held in detention 1 Witnessed stranger injured 1 Close family member tortured/beaten/abused 1

13

that these particular symptom checklists when used in combination do have some use as a public health measure (Jones & Kafetsios, 2002). We therefore felt it was worth examining the relationship between exposure and symptom scores, as well as between exposure and the qualitative assessment of well-being.

Results Main sample Area comparison of symptom scores in the main sample Adolescents’ scores on the anxiety, depression and trauma scales were not significantly different in the two towns. However, girls scored higher than boys on all three symptom checklists (anxiety F(1,328) = 15.61, p < .001; depression F(1,328) = 19.07, p < .001; trauma F(1,328) = 10.05, p < .001). There were no interactions between town and gender. Examination of results using the cut-off point scores method suggested by Mollica et al. (1992) did not reveal any statistically significant differences between the two towns in anxiety (␹2 = 2.1, ns), depression (␹2 = 0.61, ns), or trauma (␹2 = 2.6, ns).

Sub-sample Exposure to violence, number and type of war-related events experienced

The number of war-related events reported ranged from 5 to 20, with an average of 12.17 events (SD 4.07). Thirty-two per cent of the adolescents experienced 5–9 events, 40% experienced 10–15 events and 28% experienced 16–20 events. Table 2 presents the number of adolescents reporting particular war-related events (EWRE) ranked by order of frequency. Compared with their counterparts in Foča, adolescents in Goražde reported a higher mean number of events (F(1,34) = 12.1, p < .001), events occurring with greater 164


Table 3. Exposure to war-related events in the two towns

Number of events Frequency Intensity

Gorazˇde —————————— Mean SD

Focˇa —————————— Mean SD

F(1,34)

14.10 11.30 16.50

10.20 3.45 7.50

12.1 16.6 31.23

4.12 7.60 4.82

2.96 4.02 5.25

Table 4. Correlations between exposure indices and symptom scores by town Gorazˇde Focˇa —————————————— ——————————————— Anxiety Depression Trauma Anxiety Depression Trauma Number of events Frequency Intensity

.19 .32 .31

.12 .20 .31

.47* .48* .30

.03 –.27 –.26

–.24 –.31 –.15

–.21 –.27 –.22

Displacement

.24

.14

.44*

–.40

–.65**

–.43*

N = 20 participants in each town. Non-parametric. * p < .05; ** p < .01.

frequency (F(1,34) = 16.6, p < .001), and events of greater intensity (F(1,34) = 31.23, p < .001. The means are shown in Table 3. There were no significant differences between males’ and females’ exposure scores, a finding that reinforces the idea of EWRE reports as valid indicators of exposure. There was some, albeit not significant, relationship with their living situation (rs(40) = .21, ns) in that displaced adolescents in both towns had higher exposure scores.

Exposure and psychological well-being Quantitative analyses Overall, exposure scores and symptom scores in the sub-sample were not significantly related. However, there was a significant interaction with town. In Goražde, the number of events, the frequency and the intensity of experienced events were positively related to symptoms (see Table 4). In Foča, symptom scores were negatively related to exposure, but did not reach conventional significance levels. There was no significant overall association between exposure scores and the qualitative measure of psychological well-being (well/less well). However, as with the symptom scores, there was a different pattern in each town. Exposure was negatively related to well-being in Goražde adolescents; the more exposed adolescents were less well (number of events rs(20) = –.25; frequency rs(20) = –.40, p < .05; intensity rs(20) = –.32). In Foča exposure and well-being were positively associated: more exposed adolescents were well (number of events rs(20) = .24; frequency rs(20) = .44, p < .05; intensity rs(20) = .01). As in the main sample females reported significantly higher levels of anxiety/depression symptoms (F(1,334) = 13.65, p < .0001) and trauma symptoms (F(1,334) = 23.8, p < .000) and there was no interaction between gender and town of origin. However, there were no differences between boys’ and girls’ levels of well-being as revealed by the qualitative measures of feeling less well/well (␹2(2) = 2.48, ns). 165


Table 5. Current worries identified by children

School War returning again Parents Adolescent concerns None

Well (N = 21)

Less well (N = 17)

11 3 1 6 6

13 7 4 1 1

Qualitative analyses In spite of their different conflict experiences there were equal numbers of adolescents who felt ‘less well’ in Goražde and Foča. In Goražde those adolescents who spent the war in quieter parts of the countryside or were evacuated out of the country altogether felt well. The narratives of the less-well adolescents included accounts of seeing wounded and dead bodies, seeing other adults and children killed in front of them, being imprisoned in their homes, losing close friends or relatives; as well as repeated attacks on their homes and school, and being cold and hungry. In Foča feeling less well showed no relationship to exposure. It was related to personal loss of a close relative, missing friends, a feeling of discomfort and unhappiness with the current situation, and stress related to school. Almost all the adolescents stated that their biggest current worry was their performance in school. However, less-well adolescents had more current worries (see Table 5). They expressed concern about war coming again, and about their parents. Well adolescents had fewer worries, and these were the normal concerns of adolescence such as looks, relationships and their personal future.

Displacement Quantitative analyses

Displacement also affected well-being in the two towns in a similar fashion to exposure (see Table 4). Displaced adolescents in Goražde had higher symptom scores (especially trauma) and were less well. In Foča displaced adolescents had lower scores and were more likely to be well. The interaction effect of displacement and town remained even when controlling for gender and exposure (F(1,34) = 8.6, p < .01). Similarly, the interaction effect of exposure (high–low) and town remained when controlling for displacement and gender (F(1,34) = 5.6, p < .01). Both exposure and displacement interacted with town independently to affect symptoms of well-being. A three-way analysis of variance (town, exposure and displacement) was not significant (F(1,34) = 2.1, p < .08 the trend is due to displaced adolescents from Goražde having higher exposure scores and higher trauma symptoms).

Qualitative analyses Just over a quarter of the adolescents in both communities were still displaced in 1998 (no longer living in their own homes) but the experience varied greatly in its meaning and impact. Displaced Goražde adolescents had fled from fighting or been ethnically cleansed from their communities. All had had their homes destroyed and regarded displacement as unpleasant but did not see it as the worst of their problems. All wanted to go back to their old homes even though some were in what had become Republika Srpska. In Foča, displaced adolescents had fled from fighting, or moved deliberately with the encouragement of their own authorities in order to remain in a Serbian area after the Dayton Peace Agreement. Most had ambivalent views about 166


displacement. All of them had hated having to leave their homes, but none wanted to return to what they saw as ‘Muslim’ territory. Two of the rural adolescents regarded living in town as a great improvement on village life.

Specific types of exposure Quantitative analyses Specific types of events were related to different patterns of symptoms: Witnessing the death or injury of a family member or close friend was strongly associated with symptoms in all three scales (Anxiety r(40) = .32, p < .05; Depression r(40) = .44 and Trauma r(40) = .37, p < .01). Similarly, witnessing the death of someone close was positively associated with anxiety, depression and trauma symptoms (Anxiety r(40) = .35, Depression r(40) = .35 and Trauma r(40) = .36, p < .05). Being injured produced high anxiety and trauma scores (Anxiety r(40) = .33, p < .05; Trauma r(40) = .28, p < .07). Missing school and exposure to siege for a long duration, and repeated exposure to hunger, were related to higher trauma scores (r(40) = .32, p < .05 in both cases). Experiencing bombing and having your home attacked or shelled were related to all three scales if they happened repeatedly (Anxiety r(40) = .33, Depression r(40) = .33 and Trauma r(40) = .35, p < .05).

Qualitative analyses Losses: Twenty-seven adolescents suffered the death of a relative or close friend. Five adolescents lost someone in their immediate family. Four of these were still actively grieving between 2 and 6 years after the event. More than half the adolescents (27/40) regarded the death or injury of friends or family as their worst experience. Losses came up repeatedly in the adolescent’s written stories. Shelling: Adolescents in Foča all experienced far less shelling than adolescents in Goražde but almost half, regarded it as their worst experience (7/18), compared with only two in Goražde (2/13). In both towns the majority of those who experienced shelling found it frightening at the time (24/31). Being frightened or not, depended on how personally threatened adolescents felt. Foča adolescents who believed that NATO bombs were only aimed at military targets and their homes were in secure positions, were not scared: S4 was not scared. . . . We didn’t think that they would bomb civilian targets, I knew that I wouldn’t be hurt because they didn’t have anything to shoot at here. They were just destroying bridges and some military buildings. In contrast, S12, who lived close to a bridge thought ‘that the building could be destroyed or they might kill us all’. The reactions of Goražde adolescents, who were exposed to more persistent shelling, followed a common pattern: initial curiosity and bewilderment, changing to fear when they saw what shelling could do. This was followed by habituation, which is defined as a combination of skill in determining the level of threat, and boredom. I didn’t know what it was, at all. And later when I was a bit more grown up, we wanted to go out, we just didn’t want to stay in the house. But when some people died in our basement, then we got scared. (G6) Later she acquired the confidence to play outside: When we heard the sound of the grenade flying above us, we just kept playing because we knew it was flying over us [. . .]. but when they fired the grenades – you 167


hear when they fire it – and when you don’t hear that sound any more you just lie down, because you don’t know then where it’s going to fall. G2 enjoyed ‘counting the grenades’ in the third year of the war. ‘It was boring playing cards all the time, so it was quite a new game for us, to count grenades.’ In contrast, some adolescents were always frightened and did not habituate. GE had been told terrifying stories of imprisonment and rape by an aunt who escaped from Foča, and lived close to the front line in Goražde. She saw the bombing as extremely personally threatening: ‘Every night we were really frightened because we thought that they could come to our basement to kill us, to slaughter us . . . nothing was as bad as that’. Witnessing injury or death: Those who experienced this in either community regarded it as one of the worst experiences and remembered it in vivid detail years after the event. It was much more common in Goražde. G6 was sitting in her basement when two grenades exploded in front of it injuring three people sheltering just outside: they got stuck on the door because they were rushing forward looking for help, so there were three of them, you know, on the door. And my mother just got close. The man she was laying down next to my brother, his face was totally gone. And when he started bleeding, the blood came right out on us. [. . .] And he was asking for water and my mother wanted to give him water. She got lost again because she didn’t know how to give him water because this man didn’t have a mouth any more. Other experiences: In other respects the two communities had very different war experiences. In Foča, evacuation was organized in advance of the fighting, usually with mother so few adolescents were separated from both parents. In Goražde, Muslim families were taken by surprise, and more adolescents were separated from both parents because of last-minute evacuation, or because one parent might have gone to search for food while the other was fighting. Many adolescents regarded shelling when at school as one of the most frightening experiences because no family was present. Goražde adolescents who experienced detention first hand found this particularly terrifying. Media stories of events such as the massacre at Srebrenica meant that many Goražde adolescents believed they would all be killed (massacred) during the last offensive to try and take the town. In Foča, few appeared to feel threatened in these particular ways, but the ethnic cleansing of the non-Serb population from the town did leave its mark. Almost a quarter said that missing friends who had been sent away, was one of their worst experiences. Post conflict life: In Foča, the lifelines and narrative discussion showed that some adolescents found post-war life worse than the experience of war. This was related variously to the changing conditions in the town, and transition to high school with its much tougher work regime. FE, expressed this eloquently in her written story: It was very difficult to part with my best and dearest friends who had to leave their homes and go somewhere else. Even now, after the war is over it’s not so nice or great. We all miss many things. We’re all, somehow depressed, tense, and nervous. Domestic conflict and violence: Adolescents also made clear that certain domestic life events that involved abuse, or loss through divorce were more painful than war. G8 lived with a violent and alcoholic father and ran away frequently as a small child. War was something of a relief because father was absent. For G8, 1994, when Goražde was exposed to the worst offensive, was ‘the best in my life because my father on the Front Line [. . .] I was together with my mum and I really felt really nice’. In addition, the family was dependent on her resourcefulness in finding food. At the end of the war her father 168


returned. There was further domestic violence and she took an overdose. Now that her parents are divorced things have begun to improve. Adolescents also described specific responses to different kinds of events. They made clear distinctions between ‘worst’ experiences, which usually involved loss, and their most ‘frightening’ ones. S12 (quoted earlier) explained how specific kinds of events had different effects: ‘the most sad thing was death of those people and most frightening is bombing’. Being frightened during the war did not relate to later well-being, in that equal numbers of well and less-well adolescents discussed being scared at various times during the conflict. Narrative interviews revealed two main clusters of responses: sadness and depression were more common in those adolescents who had lost a close relative or friend through either death or divorce. G4, for example, still felt very sad about his father, killed in the first year of the war (8 years previously). He still cried often, and was doing badly at school. It also occurred in adolescents native to Foča who complained of changes in the town (such as FE quoted earlier). Most adolescents who had witnessed death and injury at close proximity described experiencing persistent intrusive recurrent images, but not all associated such symptoms with feeling less well. In well adolescents these images were becoming less frequent with time. In the less-well adolescents the images were more persistent and troubling and associated with other problems such as poorer concentration, and irritability. Adolescents who had both experienced a loss and witnessed death, not surprisingly felt the least well and had both kinds of responses. Overall, the narratives revealed that although most adolescents in Goražde experienced a greater quantity of political violence, adolescents in both communities suffered a wide range of disturbing events over long periods. Exposure to different types of events affected adolescents in quite different ways. The relationship to well-being did not show a straightforward dose–effect, and the significance and impact of different events was clearly mediated by the social context.

Discussion This study found a significantly higher number of events (mean 12) than other studies in areas of chronic conflict such as the Middle East (Macksoud & Aber, 1996). This indicates the particular severity of this war. Given the conflict histories of the two communities, it is not surprising that Goražde adolescents were exposed to more events, more frequently, and of a greater intensity than those in Foča. Thus the similar means for symptom scores in the two communities in the main sample and the existence of equal numbers of less-well children in both communities in the sub-sample requires some explanation.

The effect of context The results from the sub-sample seem to suggest that a certain level of exposure is necessary before a dose–effect occurs. In the more-exposed community of Goražde, adolescents who felt ‘less well’ and had more symptoms, were more likely to have experienced a greater number of war-related events and to be displaced. However, the sub-sample results also suggest that in particular political contexts there are some things that can be more detrimental than exposure to political violence. In the less-exposed community of Foča more exposed, displaced adolescents felt better than less exposed, local ones. Not surprisingly, displaced adolescents in both towns had been exposed to more events, as in most cases it was greater exposure to danger that had precipitated displacement. However, the independent interactions between displacement and town, and exposure and town on well-being suggest that being displaced is not simply a marker for high exposure, but is a specific experience with an independent impact. 169


Salama, Spiegel, van Dyke, Phelps, and Wilkinson (2000) looked at psychiatric morbidity in Serbian adults who had remained in post-conflict Kosovo, and found a high prevalence of severe social dysfunction and severe depression, particularly in women. The scores were much higher than population scores for other European countries and comparable with those of the Kosovar Albanian population, who had been far more exposed to severe and multiple war-related events (Cardozo, Vergara, Agani, & Gotway, 2000; Salama et al., 2000). Salama et al. postulated that the social isolation of this group, sudden change in socio-economic status, feelings of physical insecurity and lack of hope for the future all contributed to this morbidity. Similar mechanisms may be at work in Foča. The complete disruption of the town, the arrival of large numbers of new people, combined with a change in the name, left many local families feeling like strangers in their own community. Native-born young people made clear their sensitivity to this mood. Many spontaneously identified living in a neglected, isolated and depressed community, worry about school marks and missing friends as their worst problems. Possibly these factors interact in a particular way. Adolescents worried more over school in Foča than Goražde because it compensated parents who had lost out in the war, and was a way out (through university) of an unhappy living situation. Half the adolescents in Foča wanted to leave for another country (Jones, 2002). In contrast, the more-exposed, displaced adolescents in Foča, in spite of being upset by losing their home, had not witnessed this transformation of their own community through ethnic cleansing. The new town name did not upset them. Some had actually improved their living circumstances by moving to a more spacious urban house. Others wanted to emphasize their adaptation to their new life. They underlined their lack of interest in returning to what they regarded as enemy territory by stressing their lack of symptoms and current well-being.

Specific events and specific outcomes In keeping with other research (Sack et al., 1999; Savin et al., 1996) there is some evidence that specific events have specific outcomes, and that this varies with their frequency, intensity and the context in which events occur. This is best exemplified by the contrast between the impact of loss, witnessing injury or death, and the impact of shelling. Both the quantitative and qualitative results make clear that loss and injury affecting friends or family were the worst things that could happen, regardless of the community in which they occurred, and that their impact was enduring and exacerbated by witnessing the event. Shelling made the majority of adolescents frightened. Fear was less if the bombardment could be interpreted as non-threatening. Interestingly, this was more likely in Goražde where shelling persisted for long enough to allow habituation, and occurred in the context of multiple other threatening events. Similar findings were reported during the Blitz in various English cities in World War II (Bodman, 1941; Carey-Trefzer, 1949). Being frightened at the time of bombardment was not necessarily associated with persistent symptoms, or feeling less well, in the long term. It was more likely to be so for those who endured more than one day of NATO bombing in Foča, or where there was a direct attack on an adolescent’s home in Goražde. Not surprisingly, the duration of the siege and of missing school had an impact on wellbeing in Goražde. Both were markers for high exposure. The longer period of siege they endured, the more exposed an adolescent was, and the more school they missed. There may also have been a direct effect in that adolescents who missed school had less contact with friends, and less daily activities to distract them from the war. Although it was not possible to identify a clear-cut relationship between loss events and depressive symptomatology, or traumatic events and trauma symptoms on the basis 170


of the symptom checklists alone, narrative interviews did reveal specific patterns of responses. Witnessing injury or death was likely to produce responses associated with post-traumatic stress, such as intrusive imagery, fearfulness and irritability. Losses were more likely to result in persistent feelings of sadness and depression. All of these results confirm the importance of subjective psychological distance from events compared with the actual quantity of events (Jensen & Shaw, 1993). Aerial bombardment that is intense and unexpected may feel much more threatening, than intermittent shelling that has become the background to daily life, unless one is reminded of one’s vulnerability by a direct hit. Actually witnessing injury and death, even of a stranger, brings home to the child their own vulnerability in a very direct way. This concept is particularly useful in distinguishing frightening from less-frightening events. When looking at post-conflict recovery, an additional explanatory concept might be the subjective perception of personal disruption caused by an event. To what degree does an adolescent feel their personal world; the networks of family and friends and landscape that composed it, have been disrupted. This concept helps explain the enduring effects of loss, and why displaced Goražde adolescents who long to return to their old homes, or local Foča adolescents who want to escape from homes they no longer recognize, have done less well than those determined to adapt to new circumstances. Having to start completely afresh in a new community may be easier than returning to an old one that is changed. The disrupted past is closer, because the reminders in the form of uninhabited ruined buildings and empty spaces, can be seen. The concept also clarifies why divorce and domestic violence, which completely disrupt the immediate world of the child or adolescent, appear to be more disturbing than war.

Limitations There are some limitations to the study design that were unavoidable in this social context. The schools and allocated classes were not randomly chosen, but there were no significant differences with the other classes of adolescents of a similar age group in the two communities. With the sub-sample it was not possible to completely randomize the selection of low- and high-scoring adolescents because of the intensive cooperation required with their families. The primary school adolescents who refused to participate in the qualitative part of the study did have particular characteristics: the majority of refusals came from low-scoring boys in both primary schools, and from one particular primary class in Goražde. Young boys who see themselves as ‘symptom free’ may feel they have better things to do with spare time than talk to a female interviewer about feelings and experiences. A few of these boys were identified by staff as having behaviour problems in school. In Foča, some of the refusals are likely to have come from families hostile to a researcher from a NATO country. In Goražde there was the added problem of one class having an influential teacher opposed to psychological studies. All the respondents in the sub-sample came from within the highest and lowest quartiles when looking at the full range of scores. However, the study sample may be slightly biased towards the more symptomatic group in the population, including those who felt psychological well-being worth discussing, and away from those who lacked symptoms or denied them. The small sample size also means that both the qualitative and quantitative findings should be applied to other contexts with great caution.

Conclusions The findings from this study have implications for the methods and focus of research exploring the effects of exposure to political violence on psychological recovery. There 171


are also implications for clinical practice. First, there are advantages to using mixed methods. Culturally specific trauma event scales can give a detailed person-specific record of exposure in a way that simple area comparisons cannot. Adding the dimensions of frequency and intensity provides information on aspects of exposure that have subjective significance. Exploring pre- and post-conflict life history and circumstances allows war-related events to be set in the context of other life experiences and their impact compared. Qualitative methods, in combination with clinical rating scales, allow for a more comprehensive assessment of well-being, and make it possible to understand the subjective significance of particular events and particular symptoms. Second, the study demonstrates the wide range of events to which young people can be exposed in a conflict. It suggests that the type of event, and the context in which it occurs, are more important than the overall quantity of exposure. It demonstrates that adolescents can make subtle distinctions between different kinds of events and between different kinds of responses, all of which have implications for their long-term wellbeing. It appears to be the degree to which an event disrupts the adolescent’s personal network, and their personal sense of security that is most significant in determining its longer term impact. Loss, whether through death, or divorce; and witnessing the death or injury of a close relative or friend have the strongest association with later poorer psychological well-being, regardless of the context in which they occur. The impact of shelling appears to depend on the degree to which it is perceived as personally threatening, which may have a paradoxical relationship with quantity, where habituation leads to less feeling of threat. This depends on the context created by political community and other war-related events.

Implications for clinical practice There are a number of practical implications for those involved with providing psychosocial and mental health assistance to children and young people after wars. First, those planning post-conflict interventions should not ignore those communities with relatively low levels of exposure to political violence, particularly when they are perceived by humanitarian workers as being on the side of the perpetrator. Children in such communities may have specific difficulties related to the isolation and neglect of their communities, and the perceived role in the conflict, that are worthy of attention. Second, the identification of children in need of support should be child focused rather than event focused. The level of fear generated by an event is not a good indicator of longer term distress. This means that therapies such as ‘debriefing’ that may be offered on the basis of exposure to trauma rather than clinical need (Raphael & Wilson, 2000) would not be appropriate and might pathologize children unnecessarily. Some generalizations are possible regarding events: those who have experienced losses of family or friends, particularly if witnessed, are likely to be vulnerable and in the greatest need of support. Third, the form of the support offered needs to take into account the nature and interpretation of the events experienced, and children’s own wishes regarding treatment. The first author has found that group work emphasizing shared experience and peer support is effective with pre-teenagers who have lost one or both parents, both in Kosovo and Bosnia. Family-focused interventions are likely to be the most effective with those children suffering from family dysfunction or domestic abuse. These should include the involvement of protection agencies if necessary (Jones, Rrustemi, Shahini, & Uka, 2003). Seven ‘less-well’ children in this study were offered psychological assistance in the form of individual counselling. One accepted. Another, suffering from disturbing 172


traumatic symptoms that left him with a persistent anxiety and fear, elected to have the fear ‘removed’ by an indigenous practitioner. The treatment involved ritual, but in contrast to western psychotherapeutic approaches, there was no exploration of the events generating the fear or ventilation of feeling. The others refused but identified a variety of social and political changes that would improve their well-being, in particular the return of friends, more sport and recreation activities, classroom reform, the rebuilding of houses, being able to return home, and the arrest of profiteers and war criminals. Such interventions may require clinical practitioners to devote more time to advocacy than individual therapy but they have the advantage of addressing the problem at a communal level, and thus helping to the repair social world in which the young person lives (Jones & Kafetsios, 2002).

Implications for future research Jensen and Shaw (1993) called for the development of a typology of environments to understand the dissimilar effects of similar events in war. Future research in this area may benefit from the inclusion of ethnographic methods and should include both sides of a conflict to increase our understanding of the role of political context. Research comparing the impact of non-war and war-related life events would also be of value.

Acknowledgements The authors would like to thank the William T. Grant foundation for funding the research on which this article is based, Professor Richard Mollica for permission to use the Bosnian translations of the HTQ and HSCL-25, Professor Martin Richards and unknown reviewers for helpful comments on earlier drafts and Sally Roberts for assistance in preparation of the manuscript.

References Adjukovic, M., & Adjukovic, D. (1993). Psychological well-being of refugee children. Child Abuse and Neglect, 17, 843–854. Almqvist, K., & Brobert, A.G. (1999). Mental health and social adjustment in young refugee children 31⁄2 years after their arrival in Sweden. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 723–730. American Association of Child and Adolescent Psychiatry. (1998). Practice parameters for the assessment and treatment of children and adolescents with post traumatic stress disorder. Journal of the American Association of Child and Adolescent Psychiatry, 37(10, Suppl.), 5s–26s. Apfel, R.J., & Simon, B. (2000). Mitigating discontents with children in war: An ongoing psychoanalytic inquiry. In A.C.G.M. Robben & M.M. Suarez-Orozco (Eds.), Cultures under siege: Collective violence and trauma (pp. 102–130). Cambridge: Cambridge University Press. Bodman, F. (1941). War conditions and the mental health of the child. British Medical Journal, 286–288. Bryce, J.W., Walker, N., Ghorayeb, F., & Kanj, M. (1989). Life experiences, response styles and mental health among mothers and children in Beirut, Lebanon. Social Science & Medicine, 28, 685–695. Cardozo, B.L., Vergara, A., Agani, F., & Gotway, C.A. (2000). Mental health, social functioning, and attitudes of Kosovar Albanians following the war in Kosovo. Journal of the American Medical Association, 284, 569–577. Carey-Trefzer, C.J. (1949). The results of a clinical study of war damaged children who 173


attended the Child Guidance Clinic, The Hospital for Sick Children, Great Ormond Street, London. Journal of Mental Science, 400, 35–44. Dawes, A., & De Villiers, C. (1987). Preparing children and their parents for prison: The Wynberg seven. In D. Hanson (Ed.), Mental health in transition. Cape Town, South Africa: OASSSA Second National Conference Proceedings. Elbedour, S., Ten Bensel, R., & Bastien, D.T. (1993). Ecological integrated model of children of war: Individual and social psychology. Child Abuse and Neglect, 17, 805–819. Farwell, N. (1999). After liberation: Psychosocial well-being of Eritrean youth. Dissertation Abstracts International Section A: Humanities and Social Sciences, 59(8-A), 3208. Filstead, W.J. (Ed.). (1971). Qualitative methodology: First hand involvement with the social world. Chicago: Markham. Foy, D.W., Madvig, B., Pynoos, R.S., & Camilleri, A.J. (1996). Etiologic factors in the development of post-traumatic stress disorder in children and adolescents. Journal of School Psychology, 34(2), 133–145. Gergen, M. (1988). Narrative structures in social explanation. In C. Antaki (Ed.), Analysing everyday explanation: A casebook of methods (pp. 94–112). Sage: London. Glaser, B.G., & Strauss, A.L. (1967). The discovery of grounded theory: Strategies for qualitative research. New York: Aldine. Henwood, K., & Pidgeon, N. (1992). Qualitative research and psychological theorising. British Journal of Psychology, 83, 97–111. Human Rights Watch. (1998). Bosnia and Herzegovina, ‘a closed, dark place’: Past and present human rights abuses in Foča. New York: Human Rights Watch. Jensen, P.S., & Shaw, J. (1993). Children as victims of war: Current knowledge and future research needs. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 697–708. Jones, L. (2002). Adolescent understandings of political violence. A qualitative study. Social Science & Medicine, 55, 1351–1371. Jones, L., & Kafetsios, K. (2002). Assessing adolescent mental health in war-affected societies: The significance of symptoms. Child Abuse and Neglect, 26, 1059–1080. Jones, L., Rrustemi, A., Shahini, M., & Uka, A. (2003). Mental health services for war affected children: Report of a survey in Kosovo. British Journal of Psychiatry, 183, 540–546. Kostelny, K., & Garbarino, J. (1994). Coping with the consequences of living in danger: The case of Palestinian children and youth. International Journal of Behavioral Development, 17, 595–611. Kuterovac, G., Dyregrov, A., & Stuvland, R. (1994). Children in war: A silent majority under stress. British Journal of Medical Psychology, 67, 363–375. Locke, C.J., Southwick, K., McCloskey, L.A., & Fernandez-Esqier, M.E. (1996). The psychological and medical sequelae of war in Central American refugee mothers and children. Archives of Pediatric and Adolescent Medicine, 150, 822–828. Macksoud, M.S. (1992). Assessing war trauma in children: A case study of Lebanese children. Journal of Refugee Studies, 5(1), 3–15. Macksoud, M.S., & Aber, J.L. (1996). The war experiences and psychosocial development of children in Lebanon. Child Development, 67, 70–88. Mastnak, L. (1995). The process of engagement in nonviolent collective action: Case studies from the 1980s. Unpublished PhD dissertation, University of Bath, UK. McCallin, M. (1998). Community involvement in the social integration of child soldiers. In P.J. Braken & C. Petty (Eds.), Rethinking the trauma of war (pp. 60–75). London: Free Association Press. Mghir, R., Freed, W., Raskin, A., & Katon, W. (1995). Depression and post-traumatic stress disorder among a community sample of adolescent and young adult Afghan refugees. Journal of Nervous and Mental Disease, 183, 24–30. 174


Mollica, R., Caspi-Yavin, Y., Bollini, P., Truong, T., Tor, S., & Lavelle, J. (1992). The Harvard Trauma Questionnaire: Validating a cross-cultural instrument for measuring torture, trauma, and post-traumatic stress disorder in Indochinese refugees. Journal of Nervous and Mental Diseases, 180, 111–116. Mollica, R.F., Caspi-Yavin, Y., Lavelle, J., Tor Svang, Y.-T., Chan, S., Pham, T., Ryan, A., & De Marneffe, D. (1991). The Harvard Trauma Questionnaire, manual: Cambodian, Laotian and Vietnamese versions. Boston: Harvard Program In Refugee Trauma. Mollica, R.F., Poole, C., Son, L., Murray, C.C., & Tor, S. (1997). Effects of war trauma on Cambodian refugee adolescent’s functional health and mental health status. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 1098–1106. Orona, C.J. (1997). Temporality and identity loss due to Alzheimer’s disease. In A. Strauss & J. Corbin (Eds.), Grounded theory in practice (pp. 171–196). London: Sage. Paardekooper, B., de Jong, J.T.V.M., & Hermanns, J.M.A. (1999). The psychological impact of war and the refugee situation on South Sudanese children in refugee camps in Northern Uganda: An exploratory study. Journal of Child Psychology and Psychiatry, 40, 529–536. Perrin, S., Smith, P., & Yule, W. (2000). Practitioner review: The assessment and treatment of post traumatic stress disorder in children and adolescents. Journal of Child Psychology and Psychiatry, 41, 277–289. Protacio-Marcelino, E. (1989). Children of political detainees in the Philippines: Sources of stress and coping patterns. International Journal of Mental Health, 18, 71–86. Punamäki, R.L. (1996). Can ideological commitment protect children’s psychosocial well-being in situations of political violence? Child Development, 67, 55–69. Punamäki, R.L., & Suleiman, R. (1990). Predictors and effectiveness of coping with political violence among Palestinian children. British Journal of Social Psychology, 29, 67–77. Qouta, S., Punamäki, R.L., & El Sarraj, E. (1995a). The impact of the peace treaty on psychological well-being: A follow-up study of Palestinian children. Child Abuse and Neglect, 19, 1197–1208. Qouta, S., Punamäki, R.L., & El Sarraj, E. (1995b). The relations between traumatic experiences, activity, and cognitive and emotional responses among Palestinian children. International Journal of Psychology, 30, 289–304. Raphael B., & Wilson, J. (Eds.). (2000). Psychological debriefing: Theory, practice and evidence. Cambridge: Cambridge University Press. Richman, N. (1993). Annotation: Children in situations of political violence. Journal of Child Psychology and Psychiatry, 34, 1286–1302. Sack, W.H., Clarke, G.N., & Seeley, J. (1996). Multiple forms of stress in Cambodian adolescent refugees. Child Development, 67, 107–116. Sack, W.H., Him, C., & Dickson, M.A. (1999). Twelve-year follow-up study of Khmer youths who suffered massive war trauma as children. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 1173–1179. Sack, W.H., Seeley, J.R., & Clarke, G.N. (1997). Does PTSD transcend cultural barriers? A study from the Khmer Adolescent Refugee Project. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 49–54. Salama, P., Spiegel, P., van Dyke, M., Phelps, L., & Wilkinson, C. (2000). Mental health and nutritional status among the adult Serbian minority in Kosovo. Journal of the American Medical Association, 284, 578–584. Savin, D., Sack, W.H., Clarke, G.N., Meas, N., & Richart, I. (1996). The Khmer Adolescent Project: III. A study of trauma from Thailand’s Site II refugee camp. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 384–391. Straker, G., Mendelsohn, M., Moosa, F., & Tudin, P. (1996). Violent political contexts and the emotional concerns of township youth. Child Development, 67, 46–54. Strauss, A., & Corbin, J. (Eds.). (1997). Grounded theory in practice. London: Sage. 175


Tashakkori, A., & Teddlie, C. (1998). Mixed methodology: Combining qualitative and quantitative approaches. London: Sage. Thabet, A., & Vostanis, P. (1999). Post-traumatic stress reactions in children of war. Journal of Child Psychology and Psychiatry, 40, 385–391. The International Criminal Tribunal for the Former Yugoslavia. (1999). Indictment of Milorad Krnojelac: Case No: IT-97–25-I. Walton, J.R., Nuttall, R.L., & Nuttall, E.V. (1997). The impact of war on the mental health of children: A Salvadoran study. Child Abuse and Neglect, 21, 737–749. Ziv, A., & Israeli, R. (1973). Effects of bombardment on the manifest anxiety level of children living in Kibbutzim. Journal of Consulting and Clinical Psychology, 40, 287–291.

176

EXPOSURE TO POLITICAL VIOLENCE AND PSYCHOLOGICAL WELL-BEING IN

Recommend Documents