Psychosocial Support for Children in the Republic of South Sudan: an Evaluation Outcome

Eiling et al.

Psychosocial support for children in the Republic of South Sudan: an evaluation outcome

Ellen Eiling, MarianneVan Diggele-Holtland, TomVanYperen & Frits Boer

This paper describes an exploratory outcome evalu- Nations Children’s Fund (UNICEF), 2009). ation of War Child Holland’spsychosocial support Exposure to the disruption, loss and violence intervention I DEAL, a life skills intervention associated with armed con£ict, and its after- aimed at improving the ability ofchildren and young math, negatively a¡ects the psychosocial people a¡ected by armed con£ict in the Republic of wellbeing of children (Jordans et al., 2010; South Sudan to ‘deal’ with their daily lives. The Tolet al., 2011a; Attanayake et al., 2009). speci¢c objectives were to assess whether I DEAL At the individual and interpersonal levels is consistent with local perceptions of wellbeing, psychosocial support can rebuild broken and to explore the outcomes, as well as the factors relationships and promote nonviolence that in£uence outcomes of this intervention. The (Wessels & Monteiro, 2006). This article research was conducted using mixed methods, focuses on the evaluation of a psychosocial with a focus on qualitative and participatory support intervention that can contribute to methods: group exercises, individual goal setting nonviolent behaviours and other peace- (N ¼110)and interviews (N ¼ 62).Totriangulate building processes at the interpersonal level ¢ndings from children, interviews were held with in the Republic of South Sudan. teachers (N ¼ 7),facilitators (N ¼ 5),and parents Thirty-eight years of civil war, between the (N ¼11).It was found that the content of the inter- Sudan and South Sudan, has displaced 4.9 vention was consistent with children’s perceptions million people and killed more than 2 of wellbeing. Speci¢c outcomes that were reported million. The vital infrastructure of the by children and con¢rmed by facilitators, teachers Republic of South Sudan hasbeen essentially and parents included decreased ¢ghting and destroyed, and there is an almost complete improvedrelationships withpeersandparents. Find- lack of basic social services. The population ings suggest that I DEAL positively a¡ect is estimated at 9 million, with more than half children’ssocial coping skills and has the potential living below the poverty line, especially in to improve children’s emotional coping skills and rural areas (World Bank, 2013). The popu- classroom performance. lation below the age of 18 is around 51% (Central Bureau of Statistics, 2009). The Keywords: armed con£ict, children, parti- country is highly rural, and is characterised cipatory outcome evaluation, psychosocial by isolated and inaccessible communities. support, Republic of South Sudan The high costs of primary education, the necessity for children to contribute to the family income, as well as long distances to Introduction schools, all contribute to severely low school An estimated one billion children and enrolment levels. At 8%, South Sudan’s young people under the age of 18 live in female literacy rate is the lowest in the world areas a¡ected by armed con£ict (United (World Bank, 2012).

61 Copyright © War Trauma Foundation. Unauthorized reproduction of this article is prohibited. Psychosocial support for children in the Republic of South Sudan: an evaluation outcome Intervention 2014,Volume 12, Number 1, Page 61 - 75

After South Sudan’s independence in 2011, is implemented in combination with inter- ongoing con£ict in some border regions, ventions promoting education andchild pro- intercommunal con£icts, food insecurity, tection. I DEAL aims to support children seasonal £ooding and internal displacement (11^15 years of age) to deal better with the continue to a¡ect communities (United aftermath of armed con£ict, by strengthen- Nations, 2012). In addition, the presence of ing determinants of resilience and psycho- seven armed groups, the recent outburst of social wellbeing, such as self-con¢dence and violence, as well as more generalised vio- social andemotional coping skills. Important lence all continue to negatively impact determinants for resilience include having the psychosocial development of children. an internal locus of control and a sense of This is often demonstrated by violent beha- self-e⁄cacy (Betancourt & Kahn, 2008). To viour and weak social skills, as observed increase children’s con¢dence that they can during situation analyses and byWar Child’s make a di¡erence (internal locus of control) sta¡. and to achieve their goals (self-e⁄cacy), the Eastern Equatoria State, where the evalu- intervention uses participatory methods, to ation of I DEAL was conducted, was create opportunities for the children to particularly a¡ected by the civil war from express themselves and to reduce their fear 1983^2005. Until today, the region experi- of speaking within group situations. ences insecurity caused by landmines, many Other essential determinants of resilience civilians owning weapons, and shattered are social and emotional coping skills social and community relations. Cattle (Betancourt & Kahn, 2008), and supportive rustling, armed robbery, and banditry are social relationships with peers, parents and endemic. In many areas of the state there is other adults (Benzies & Mychasiuk, 2008). little security presence.Violent con£icts over I DEAL’s aims of strengthening children’s cattle and land are common, which are social and emotional coping skills contrib- further exacerbated by the return of refugees utes to improving social relationships, and the arrival of internally displaced civi- through group discussions and game based lians. In addition, Eastern Equatoria experi- activities, focused on collaboration and ences chronic food insecurity, a lack of con£ict solving. basic services, and few economic opportu- The themes addressed in the six modules of nities (United Nations Development Pro- I DEAL (each consisting of 3^4 sessions), gramme (UNDP), 2012). War Child has are identity, dealing with emotions, peer worked in Eastern Equatoria since 2009, relations, relationships with adults, con£ict implementing child protection, education and peace, and the future. Participants and psychosocial support programming. actively contribute to the intervention by The ongoing con£icts, generalised violence selecting the themes to be addressed, and and the recent outburst of ethnic violence call through participatory monitoring and for continued support in all three areas, as evaluation exercises (Claessens et al., 2012). well as e¡ective peacebuilding, to promote The intervention consists of a maximum of children’s rights and their healthy develop- 19 sessions of 1.5 hours each, implemented ment. over a period of 4 to 6 months, depending War Child follows an ecological approach on local circumstances and modules (Bronfenbrenner,1979) to improve children’s selected. Each session combines creative rights and healthy development, through and participatory techniques, such as role strengthening protective factors at various play, drawing, games and group discussions levels in society (Betancourt & Kahn, to stimulate active learning (Kalksma-van 2008). Therefore, War Child Holland’s Lith, 2007). The groups have a maximum psychosocial support intervention ‘IDEAL’1 of 25 participants and are facilitated by

Copyright62 © War Trauma Foundation. Unauthorized reproduction of this article is prohibited. Eiling et al.

community workers (hereafter facilitators) a. the achievement of participants’ in schools, after school hours. Previously personal goals, and; published and non published evaluations in b. social and emotional coping skills other countries have shown positive short- 3. What factors in£uence the outcomes of term outcomes of I DEAL; mitigating reac- the interventions? tions to violence, such as aggression and improving relations with adults and peers Methods (Claessens et al., 2012). Design Despite growing evidence of the e¡ective- The exploratory evaluation followed a ness of psychosocial interventions for child- mixed method, nonrandomised, pre and ren and young people (Flament et al., 2007; post test design. As it was intervention Jordans et al., 2010), there is an urgent need driven, the research included a convenience for a stronger evidence base on the e¡ective sample in schools where the intervention approaches for children a¡ected by armed took place, and did not include a control con£ict (Tol et al., 2011b). As complex group. The pre test was conducted in April psychosocial interventions call for practice 2012, the post test in November 2012. The driven research, process oriented and parti- research is in line with international ethical cipatory approaches are required, before standards and guidelines for the evaluation moving towards more rigorous research of psychosocial programming (UNICEF, designs (Claessens et al., 2012). Building 2011; Inter-Agency Standing Committee on a pilot study conducted in Uganda (IASC), 2007). Respondents gave informed (Claessens et al., 2012) in 2012, War Child consentverbally.Con¢dentiality wasensured Holland, in collaboration with the Univer- to all respondents, and it was explained that sity of Amsterdam, the Netherlands Youth all identity information was collected for Institute, and HealthNet TPO, conducted analysis only. an exploratory outcome evaluation and To ensure data quality, nine community a process evaluation of I DEAL in the workers received three days of training Republic of South Sudan. and conducted the interviews in local lang- The primary purpose of this research is uages (Juba Arabic and Lotuho), with to further strengthen the intervention, in structured topic lists. In order to build trust, addition to facilitating future e¡ect evalu- the teacher introduced the researchers and ations, by exploring locally relevant out- the research purpose to the children, and come measures. The study also aims to researchers used child friendly approaches contribute to current knowledge gaps in and games. the ¢eld of psychosocial support, through addressing three prioritised, international research areas for psychosocial support Sample interventions (Tolet al., 2011b): e¡ectiveness The evaluation took place in Eastern Equa- of school based psychosocial support; local toria State, where a total of 11 groups (152 perceptions of psychosocial problems; and boys and 105 girls) participated in I DEAL. the extent to which current interventions Out of the 11 groups, ¢ve (from di¡erent address these needs. This paper addresses villages) were selected to participate in the the following questions: research. Teachers selected children from grades 3 and 4, and in some cases they 1. Does I DEAL ¢t with participants’ per- included girls from another grade to improve ceptions of wellbeing? gender balance. Consequently, the age of 2. What are the outcomes of I DEAL participants ranged from 8 to 16 years old regarding: (see Table 1 for demographic information).

Copyright © War Trauma Foundation. Unauthorized reproduction of this article is prohibited.63 Psychosocial support for children in the Republic of South Sudan: an evaluation outcome Intervention 2014,Volume 12, Number 1, Page 61 - 75

Table 1. Demographic characteristics of participating children School Mean age (SD) Female N Male N Total N

112.38(1.69)111021 212.17(1.09)131124 311.32(1.72)82028 410.33(1.04)111627 511.91(1.44)61622

The selection of the research groups was sampling. However, due to practical con- based on the security and accessibility of straints, this resulted in a sample of 11 the area, and the starting dates of the inter- parents from two locations. vention. The ¢ve groups included a total of 122 children: 73 boys and 49 girls. Research instruments The total dropout rate of the intervention Local perceptions of wellbeing was 32%, resulting in a smaller research The participatory research tool, Wellbeing sample at the post test stage. See Table 2 for Exercise (Stark et al., 2012; Hart et al., an overview of the sample sizes used for ¢nal 2007), was used as described by Stark et al. analysis, per research instrument. For each (2012) to compare the content of I DEAL instrument, the selection of respondents with local perceptions of wellbeing, as well was based on presence during data collec- as its relevance and cultural ¢t. During tion, age and gender. Each child (N ¼122) group exercises, children’s local perceptions participated in at least one of the research of wellbeing were explored through re£ec- methods. From each location, at least one tion of what determines the wellbeing of a teacher from grade three or four was inter- child within their community, and what viewed. It was an aim to include parents behaviours they associated with peers who from each location through snowball are ‘doing well’.

Table 2. Overview of sample per research instrument Research Mean Female Male Total instruments age (SD) N N N

Baseline Wellbeing exercise 5 groups 12 (1.65) 15 26 41 of 8 Personal goals 11.6 (1.56) 44 66 110 Post test Personal goals 11.4 (1.59) 30 44 74 Evaluation interview ^ 26 36 62 Group discussions 5 groups ^^^56 of 11 Interview parents 2 locations ^ 8 3 11 Interview teachers 5 locations ^ 1 5 6 Interview facilitators ^ 1 2 3 Two researchers had been facilitators in the past and also provided input.

Copyright64 © War Trauma Foundation. Unauthorized reproduction of this article is prohibited. Eiling et al.

Personal goals asked: ‘has there been any change (positive/nega- Children set individual goals at the tive) after ¢nishing I DEAL, for you personally?’ beginning of the intervention, which is Answer options were: a) deterioration, b) included as the ‘Personal Goal’ exercise in the no change, c) some improvement, or d) intervention’s monitoring and evaluation2. much improvement, followed by an open Formulating child led and participatory question to obtain an explanation. indicators of success supports facilitators to Group discussions address the needs speci¢ed by the children A group discussion was held in each location themselves. Further, children’s participation in order to evaluate the modules. During a in planning, monitoring and evaluation group exercise, based on a structured topic can, in itself, facilitate psychosocial growth list, each child used a sticker to indicate the (Karki, Kohrt & Jordans, 2009). To formu- module they liked most (green), the one late realistic and achievable goals, the they liked least (red) and the module where personal goals were based on self-identi¢ed they learned the most (orange). In addition, strengths and di⁄culties, and set within children were asked, within small groups, the themes of the intervention; personal to formulate suggestions regarding time, set- traits and relationships with peers and ting, themes and facilitation. adults. Progress towards reaching personal goals was measured on a visual analogue Interviews with teachers, facilitators, project sta¡ line of 10cm, a continuous line without and parents numbers, using a smiling and a sad face to Totriangulate the ¢ndings, structured inter- indicate the positive and negative ends, views with teachers, facilitators, project sta¡ respectively (Baker & Hall,1994). In clinical and parents were held using structured topic settings, similar types of ‘Goal Attainment lists, including both closed and open ended Scoring’ have shown positive therapeutic questions. Respondents were, for example, value in encouraging patients to reach their asked if they had observed any changes goals (Durrant et al., 2007). There is grow- in children’s behaviour over the past six ing evidence that this way of measuring months, and what they thought had caused has greater sensitivity over standard these changes. Interviews with facilitators measures, potentially avoiding £oor and and War Child sta¡ included questions on ceiling e¡ects (Turner-Stokes et al., 2006). the quality of implementation, and other These e¡ects occur when respondents factors that could have in£uenced the inter- choose the most positive or negative answer vention’s outcomes. option, which inhibits the measurement of individual di¡erences. Analysis Quantitative data was entered and analysed Evaluation interviews using Excel and SPSS 20 software. To Based on a client satisfaction questionnaire compare the di¡erence between the mean (Healthnet TPO & Centre for Trauma scores at pre and at post test in relation to Psychology,2010),structured interviews were the variation in the data, t-tests were held at the end of the intervention to assess used to assess progress towards achieving children’s satisfaction with the intervention. children’s personal goals. Qualitative data Using both closed and open ended questions, was analysed, combining grounded theory the interviews assessed motivation to parti- and content analysis. The combination was cipate, reasons for missing one or more chosen to allow for new insights provided sessions, level of enjoyment and children’s by the data, while keeping general theoretic perceptions of the bene¢ts of the interven- themes in mind. First, intensive case level tion. To explore outcomes, children were analysis was done, followed by extensive

Copyright © War Trauma Foundation. Unauthorized reproduction of this article is prohibited.65 Psychosocial support for children in the Republic of South Sudan: an evaluation outcome Intervention 2014,Volume 12, Number 1, Page 61 - 75

analysis combining and comparing types of Personal goal scores informants, topics and crosscutting themes, Atotal of 110 personal goals were formulated in order to discover patterns emerging from in the ¢ve locations. Most children (45%) the data (Russell Bernard, 2002; Silverman, formulated goals to improve their pro social 2011). Two researchers separately conducted behaviour, including ‘sharing’, ‘helping’, ‘being the coding, in order to increase the objecti- forgiving’ and ‘having respect for others’. T h i r t y vity of data analysis. Consensus was reached four percent stated goals relating to relation- where needed. ships with others: to improve friendships or make more friends,‘socialise’and ‘play together’ with peers. Nearly 1/5 of the children (19%) Results wanted to improve on‘workinghard’and being Participants’ perceptions of wellbeing ‘disciplined’. Half of them mentioned this in Participants’ perspectives of what deter- relation to their parents or home environ- mines a child’s wellbeing can be summarised ment. Two percent of the children’s personal in ¢ve main categories (in the order of goals combined discipline and pro social frequency): material welfare; access to edu- behaviour. One child wanted to improve cation; good relations with peers and adults; self-esteem and become more courageous, religious beliefs; and a safe environment. which was categorised as ‘emotional wellbeing’. When discussing behaviours associated with Before I DEAL, the average score was wellbeing, most responses were related to: 3.0 (SD ¼1.757, N ¼ 74), in relation to achiev- behaviour towards peers, behaviour towards ing their personal goals. None of the child- adults, general conduct, expressing emotions ren showed lower scores at post test. Post and behaviour motivated by future goals. test scores within the same group were Participants strongly associated wellbeing signi¢cantly higher (t(73) ¼ 16.26, p ¼ 0.00), with being friendly and not ¢ghting, listen- resulting in an average score of 6.9 ing well to adults, and being respectful. See (SD ¼1.914). No signi¢cant di¡erences were Table 3 for the complete overview of speci¢c found between locations, age or gender. Boys behaviours mentioned. showed slightly more improvement than girls, with their average improvement rates at 4.2 and 3.5, respectively. These di¡er- Outcomes of I DEAL ences were not statistically signi¢cant During evaluation interviews (N ¼ 62), (t(72) ¼1.548, p ¼ 0.126). Quotes below 48% of the respondents stated that they provide examples of personal goals and the noticed signi¢cant personal improvement, progress towards achieving them. while 30% noticed some improvement. A 13 year-old girl’s personal goal stated The main types of improvement are listed ‘I would like to improve playing with my friends’. inTable 4. None of the participants reported She indicated that she progressed from 4 to negative changes, and 14 (22%) children 6 on the visual scale (0^10), and explained: did not report any changes during the inter- ‘I was afraid of playing with my peers, but now at view. Two of them also showed very little least I talk and play with them’. progress towards their personal goals.They A 9 year-old boy’s personal goal stated: indicated that they did not like the games ‘I would like to try to be more forgiving when and thought that I DEAL was too di⁄cult. someone has done something wrong’. After I Four children dropped out of the inter- DEAL he progressed from 3 to 7: ‘I no longer vention, and therefore missed a signi¢cant ¢ght in revenge,but Icontrolmytemperand taketime amount of sessions. The other eight did to decide. For example, someone slapped me and I report positive improvement regarding did not react. But after a day, I asked him [about their personal goals. it] and he asked for forgiveness from me, and I

Copyright66 © War Trauma Foundation. Unauthorized reproduction of this article is prohibited. Eiling et al.

Table 3. Behaviour associated with well being by participants Behaviour of child Behaviour of child Category that is doing well N that is not doing well N

Behaviour towards Friendly to others 7 Lonely and isolated 5 peers (no friends) Sharing / helping others 5 Bullying 2 Playing peacefully 4 In£uences others 1 negatively Playing with friends 3 Behaviour towards Listens / does what 10 Disobedient 8 parents/caregivers is told Respectful 5 Lack of respect 8 Helpsathome 3 Doesn’thelpathome 3 Loving relationship 3 Runs away / wanders 3 around Other 3 Does not accept 2 punishment General conduct No ¢ghting 8 Fighting 8 No stealing or being 3 Rude and insulting 4 greedy Humble / graceful 2 Greedy/doesn’t share 3 Other 2 Steals and makes trouble 2 Prays a lot 1 Not loved/beaten by others 2 Doesn’t go to church 1 Emotional Always happy 6 Short tempered / restless 3 wellbeing No worries 1 Unhappy 1 Future goals Tries to achieve good 4 Doesn’t go to school 6 results in school Hard working 1 Lazy 4 Doesn’t care for himself 1 Accepts punishments/appreciates gifts/does not go out of home without permission. Not using bad words/visiting people.

did. [But] when someone has seriously hurt me, it conduct problems, improved pro social takes time to forgive’. behaviour, relationships with peers, and relationships with adults. Thirty-nine percent (N ¼ 24) of children interviewed Social coping skills reported that they ¢ght less and are less During the evaluation interviews, most chil- aggressive after having participated in I dren reported changes regarding social skills DEAL. As one 11year-old boy stated: ‘I have and relationships. The following changes noticed some changes in me. For example, I used are described in more detail below: reduced to ¢ght a lot, but since I was taught about dealing

Copyright © War Trauma Foundation. Unauthorized reproduction of this article is prohibited.67 Psychosocial support for children in the Republic of South Sudan: an evaluation outcome Intervention 2014,Volume 12, Number 1, Page 61 - 75

Table 4. General changes reported in evaluation interviews Type of change % Total (N) Girls (N) Boys (N)

Less ¢ghting and less aggressive 39 24 10 14 Improved friendships and social skills 32 20 8 12 Improved relation with parents and 19 12 4 8 other adults Improved sharing / helping / respecting 11 7 4 3 others Less worries and better control over 64 3 1 own emotions No more stealing 3 2 1 1 Participant described one type of change 39 24 10 14 Participant described more than one 39 24 10 14 type of change Participant could not mention any 22 14 5 9 relevant changes Total 100 62 25 37

with emotions I could forgive [others] and control Nineteen percent (N ¼12) of the children myemotions’.Also, two children said they used interviewed reported improvement in their to steal, but no longer do so.Teachers in three relationship with adults.The type of changes locations and a facilitator con¢rmed a children described intheir relationships with decrease in ¢ghting, bullying and other their parents related mostly to respect and aggressive behaviour. obedience, as illustrated in these quotes: Eleven percent (N ¼ 7) described more general improvements in sharing, helping and ‘Whatever my parent tells me to do Ialways do respecting others (pro social behaviour). in time’.(Girl,11years old) This was also observed by four teachers, who ‘I always apologise to my parents for the bad stated that some children started helping things I do to them’.(Girl,13 years old) their peers with their schoolwork and were showing more respect for others. According Seven parents also described these changes. to three teachers in two locations, girls and One 14-year-old boy described a di¡erent young children in particular displayed form of improvement in his relationship with improvement in their general behaviour his parents, saying ‘I can go for advice to my after I DEAL. One facilitator mentioned parents now’. that participants became better at sharing Two teachers also reported that children during I DEAL. became more obedient, with girls in particu- One-third (N ¼ 20) of the children indicated lar showing more respect towards teachers that they strengthened their friendships and after the intervention. See Table 4 for an improved their social skills, learning how to overview of changes reported by children. make friends and ‘play well’ together. This change was also described by two teachers Emotional coping skills who observed children working well together, In the evaluation interviews, four out of 62 helping each other, behaving more socially children reported that they worried less than and becoming more united as a group. before, or could control their fears better.

Copyright68 © War Trauma Foundation. Unauthorized reproduction of this article is prohibited. Eiling et al.

One boy implicitly mentioned emotional ‘I DEAL children are performing better than coping skills as his reason for participating those who [did not participate]. In the way in I DEAL: they behave, for example they have learned about respecting one another and their parents ‘I wanted to participate in I DEAL because of and teachers. Such improvement has prevented playing gameslikeboom-chika.They make me their learning being interrupted because of stronger and physically fit in the mind, as well psychological issues’. as making me learn to let go of other problems, According to another teacher, some children like the memories of bad things like death, performed better because they were more poverty, lack of food’. attentive, and others because they had become more con¢dent and participatory One teacher commented that I DEAL in class: ‘It has also empowered some children to reduces children’s problems because ‘playing actively participate in the class, they have now makes them no longer think of it’. realised their right to talk in public or class’. Two facilitators observed that children became more con¢dent and active during I Factors contributing to intervention DEAL sessions over time. Teachers (N ¼ 4) outcomes also reported that children in I DEAL had In order to identify factors in£uencing become more con¢dent in class and parti- the outcomes, a process evaluation was cipated more. However, both a facilitator conducted to assess: appreciation of the and a teacher mentioned that improved intervention; attendance; the quality of self-con¢dence sometimes resulted in de¢ant implementation; characteristics of facilita- and disruptive behaviour from a few child- tors; and contextual factors (security, ren. The male facilitator stated: illnesses, and accessibility of locations). As appreciation and attendance may have had ‘[...]some children can also become more the more direct in£uence on the individual stubborn, they start making fun of it [the outcomes, these are described below. activity],because when you learn to express yourself you get this freedom, you’re free at Appreciation heart and free in your mind’. Findings from the group discussion indicate that children liked the module addressing Another facilitator noted that some children social relationships with peers most. The remained very shy, hardly participating in module on solving con£icts was the least the group. Only two children mentioned liked by the children, who also considered changes in con¢dence, related to their it, and the module on dealing with emotions, personal goal. to be the most useful. None of the participants thought that any of the modules Academic performance or sessions should be removed, however. All teachers interviewed (N ¼ 7) reported Participants provided suggestions on how positive changes in their pupil’s academic to improve certain exercises, and recom- performance after participating in I DEAL. mended including local games in the ses- ‘Some children started studying together in groups, sions. The majority (84%) of the children even without being told’, explained one teacher. that were interviewed (N ¼ 62) stated that Another teacher explained that the increase they ‘enjoyed the programme a lot’, 15 % of respect and increased concentration has responded ‘quite a bit’ and 1% ‘a little’. M o s t contributed to the children’s academic per- children indicated that they felt good in the formance: group during sessions, with the exception of

Copyright © War Trauma Foundation. Unauthorized reproduction of this article is prohibited.69 Psychosocial support for children in the Republic of South Sudan: an evaluation outcome Intervention 2014,Volume 12, Number 1, Page 61 - 75

nine children who stated they did not always intervention was culturally relevant. Results enjoy being in the sessions, as illustrated by showed that the sessions address most of the following quotes: the locally de¢ned determinants of wellbeing; having positive social relationships ‘For me, I did not want to participate because with peers and adults, through being I do not want to be embarrassed by friends, respectful, being friendly, preventing con- for example, during the introduction I do not £icts, and being able to cope with negative speak Arabic language very well. [...]some- emotions. The factors that children associ- times I feel bad to be in the group when I am ated with wellbeing that the intervention embarrassed’.(Girl, 9 years old) did not address were: religious practices; well ‘Idid not feelgood most of the time, because the disciplined behaviour; education; safety; gamesarehard to understand’.(Girl,11years and material welfare. Other War Child old) Holland programme components address safe environments and access to education. A signi¢cant positive relationship was found Material welfare and religion are currently between the level of enjoyment in I DEAL not addressed in War Child’s work. An and changes reported by children in the assessment of local resources and coping evaluation interviews (Kendall’s t ¼ 0.269, mechanisms would further increase the p ¼ 0.025). intervention’s relevance and cultural accep- The children who enjoyed the session less tance. Further tailoring the intervention (16%) were also those who reported the least to the context and target group could be improvement at the end of the intervention. realised through an assessment and consultation phase, as described by Jordans Attendance and colleagues (2013). Of the 122 participating children, 24% Secondly, the outcomes of the intervention attended all19 sessions,38% missed1^3 ses- were explored. Findings showed a signi¢cant sions, and 6% missed 4^7 sessions. The improvement towards achieving personal main reasons for missing a session were ill- goals. It can be argued that these ¢ndings ness or domestic chores, such as work in the may be biased, due to the subjective nature garden or cleaning. Thirty-two percent of of the self assessment. However, the results the children completely dropped out, of indicate that children seemed to score their which 29 (75%) dropped out during the last progress realistically, not scoring too high module, predominantly in the two locations at either the start, or the end. that had postponed the last sessions until The main changes reported by children, as after the Christmas holidays. Sometimes ses- well as by facilitators, teachers and parents, sions were cancelled or postponed due to were decreased ¢ghting and improved national holidays or festivities, or because relationships with peers and parents. community meetings were held in the same These outcomes are in line with previous location. In a few locations, children missed unpublished evaluations conducted in other sessions for security reasons related to inter- countries using I DEAL. The reported communal con£ict. One teacher indicated decrease in ¢ghting suggests improve- that some parents did not support their chil- ment in the ability to cope with the emotions dren’s participation because they perceived leading to aggressive behaviour. Facilitators the sessions as playtime. and teachers observed changes in con¢- dence and participation, which also suggest Discussion increased emotional coping skills. The fact Firstly, local perceptions of wellbeing that very few children mentioned coping were explored to assess to what extent the with emotions as a di⁄culty or change might

be caused by cultural or linguistic factors, that participants of psychosocial support or a di⁄culty in re£ecting on emotions. interventions should be selected based on These ¢ndings suggest that the intervention psychosocial distress screening (Jordans strengthens essentialcomponents ofchildren’s et al., 2008). Although this may further resilience: social and emotional coping skills, improve the outcomes, War Child Holland, and supportive relationships (Betancourt & aims to enhance social integrationby includ- Kahn,2008).However,whenadoptinga more ing all children within an existing group ecological conceptualisation of resilience, (i.e. class), in order to prevent stigmatisation the in£uence of children’s family and com- of those selected, or jealousy among those munitylifehastobetaken into account as well excluded. (Tol et al., 2009). In addition, by reducing Togain more insight into other contextually violence and strengthening relationships, I relevant factors a¡ecting children’s psycho- DEAL potentially contributes to peacebuild- social wellbeing, as well as broader peace ing processes (Wessels & Monteiro, 2001). building processes, it is essential to assess Furthermore, I DEAL seemed to have posi- community needs and build on available tively a¡ected pupils’ academic perform- resources. Addressing these factors of ‘ecologi- ance, in line with studies inWestern settings cal resilience’ can contribute to preventing (Gavala & Flett, 2005; Van der Merwe, re-occurrence of violent con£ict (Tol et al., 2005), as well as earlier ¢ndings from War 2009). Child Holland in Uganda (Claessens et al., 2012). This may be explained by the social Limitations and emotional coping skills gained, which Conclusions regarding attribution of out- have shown to positively a¡ect school per- comes have to be drawn with care, due to formance (Lam & Kirby, 2002; Zins et al., the non randomised sampling, and the lack 2004). of a control group. Demographic infor- Additionally, factors that could in£uence the mation, except gender and age, was not outcomes of the intervention were analysed. collected; therefore conclusions about repre- The ¢ndings indicate that the participants sentativeness cannot be drawn. Despite who enjoyed I DEAL less than the others, e¡orts made to balance gender in both the reported less improvement. Attendance intervention, as well as in the research, this could not be identi¢ed as a factor in£uencing turned out to be challenging due to low individual outcomes. However, enhanced school attendance of girls in the region. attendance and continuity of implementa- The dropout rate for personal goal measure- tion could signi¢cantly improve outcomes. ments was 33%, which may have biased the Further, the quality of implementation, results to some extent. However, the inter- characteristics of facilitators, and contextual vention driven evaluation provides strong factors (security, illnesses, and accessibility of qualitative evidence for outcomes, as the locations) may have in£uenced the outcomes. sample size was substantial, and responses In addition, although no signi¢cant di¡er- from children were triangulated with those ences were found in outcomes of younger or of parents, teachers and facilitators. older participants, the wide age range might have negatively in£uenced the outcomes, Conclusion as this can a¡ect the group dynamics (for Contributing to current gaps in knowledge example, younger children may not have felt on psychosocial support (Tol, et al., 2011b), comfortable and learn at a di¡erent pace). this study indicates that the psychosocial As not every child experiences behavioural, support intervention I DEAL is consistent social or emotional problems as a result of with local perceptions of wellbeing, posi- being exposed to con£ict, it can be argued tively a¡ects children’s social coping skills

Copyright © War Trauma Foundation. Unauthorized reproduction of this article is prohibited.71 Psychosocial support for children in the Republic of South Sudan: an evaluation outcome Intervention 2014,Volume 12, Number 1, Page 61 - 75

and potentially improves children’s emotio- Acknowledgments nal coping skills and classroom perform- Wewould like to express our deepest appreciation ance. Through reducing violence and to all who made the research possible: in the ¢rst strengthening relationships, I DEAL also place all participating children, head teachers has the potentialto contribute to peacebuild- and teachers, parents and caregivers, the facilita- ing processes. Findings indicate that enjoy- tors from partner organisations Sarra and Yopas, ment of participation in the intervention and all War Child Holland sta¡ in South Sudan. positively in£uences the outcomes of the We would also like to acknowledge the crucial intervention. role of the core research team in South Sudan (Grace Abalo, Umba Peter Bosco, Grace John, Recommendations Joyce Katala, Jane Mindraa, Stephen Lokol, Findings from this study should be used to and Paul Tongi), especially Umba Peter Bosco, further improve the e¡ectiveness of the psy- Joyce Katala and Mindraa Jane Silver. Much chosocial support intervention and inform thanks also go to Frits Boer, Mark Jordans and further research. The results lead towards Tom van Yperen for their guidance throughout the following recommendations: the research. Our thanks also go to all colleagues who provided input and support in completing Toe¡ectively promote children’s psycho- this paper: Annette van Andel, Elise Griede, social wellbeing and healthy develop- Lasu Joseph, Jessie Kolkman, Laura Miller, ment, psychosocial interventions should Arjen Mulder, Esther Obdam and Marten becombinedwithother interventionsthat Schoonman. Furthermore, we would like to address material wellbeing, health, edu- thank Adessium Foundation for ¢nancially sup- cation, and protection, and that build on porting this research, and enabling War Child local resources and coping mechanisms. Holland to conduct in depth evaluations and Tofoster consistent participation and sus- learn about how to further improve the psycho- tainability of outcomes, community sup- social support provided to children a¡ected by port for the intervention needs to be armed con£ict. increased. Building on this exploratory outcome References study, future evaluations should measure Attanayake,V., McKay, R., Jo¡res, M., Singh, S., outcomes by de¢ning more speci¢c and Burkle, F. & Mills, E. (2009). Prevalence of culturally relevant indicators of psycho- mental disorders among children exposed to social wellbeing. war: a systematic review of 7,920 children. Medi- Future research is needed to analyse the cine, Con£ict and Survival, 25,3-17. relation between speci¢c intervention components and outcomes using (waiting Baker, R. & Hall, J. (1994). A Review of the list) control groups, to be able to draw Applications of the REHAB Assessment System. conclusions about e¡ectiveness. Future Behavioural and Cognitive Psychotherapy, 22,211-231 studies should also analyse to what extent doi: 10.1017/S1352465800013084. psychosocial characteristics, age range Benzies, K. & Mychasiuk, R. (2008). Fostering and gender balance within the group, family resiliency: a review of the key protective attendance, quality of implementation factors. Child & Family Social Work, 14(1), 103-114 and contextual factors in£uence the e¡ec- do i: 10.1111/j.1365 -220 6.20 0 8.0 058 6. x. tiveness of the intervention. More research is needed to explore the Betancourt,T.S. &Khan,K.T.(2008).Themental longer term e¡ects (including on aca- health of children a¡ected by armed con£ict: demic performance) of IDEAL andother Protective processes and pathways to resilience. psychosocial interventions. International Review of Psychiatry, 20(3), 317-328.

Bronfenbrenner, U. (1979). The ecology of human Inter-agency Standing Committee (IASC). development. Cambridge: Harvard University (2007). IASC guidelines on mental health and Press. psychosocial support in emergency settings. Geneva, Switzerland: IASC. Central Bureau of Statistics (2009). Fifth Popu- lation Census of Sudan. Figures retrieved from Jordans,M.J.D.,Komproe,I.H.,Tol,W.A.&de http://ssnbs.org/storage/SPHC%202008%20tables. Jong, (2008). Screening for psychosocial distress pdf amongst war a¡ected children: Cross-cultural construct validity of the CPDS. Journal of Child Claessens, L. F., De Graa¡, D. C., Jordans, M. J. Psychology and Psychiatry, 50(4), 514 - 523 do i: 10.1111/ D., Boer, F. & VanYperen,T. A. (2012). Participa- j.1469-7610.2008.02028.x. tory evaluation of psychosocial interventions for children: a pilot study in Northern Uganda. Inter- Jordans,M.J.D.,Komproe,I.H.,Tol,W.A., vention, 10(1), 43-58 doi: 10.1097/WTF.0b013e328 Kohrt,B.A.,Luitel,N.P.,Macy,R.D.&de 35179b1. Jong, J.T.V.M.(2010). Evaluation of a classroom- based psychosocial intervention in con£ict- Durrant, C., Clarke, I.,Tolland, A. & Wilson, H. a¡ected Nepal: a cluster randomized controlled (2007). Designing a CBT Service for an Acute trial. Journal of Child Psychology and Psychiatry, 51, In-patient Setting: A pilot evaluation study. 818-826 doi: 10.1111/j.1469-7610.2010.02209. Clinical Psychology and Psychotherapy, 14,117-125 doi:10.1002/cpp.516. Jordans, M. J. D., Tol, W. A., Susanty, D., Ntamatumba, P., Luitel, N. P., Komproe, I. H. Flament,M.F.,Nguyen,H.,Furino,C.,Schachter, & De Jong, J.T.V.M.(2013). Implementation of a H., MacLean, C., Wasserman, D., Sartorius, N. Mental Health Care Package for Children in Areas & Remschmidt, H. (2007). Evidence-based of Armed Con£ict: A Case Study from Burundi, primary prevention programmes for the pro- Indonesia, Nepal, Sri Lanka, and Sudan. PLoS motion of mental health in children and Med, 10(1), e1001371 doi:10.1371/jour nal. pme d.1001371. adolescents: A systematic worldwide review. In: H. Remschmidt, B. Nurcombe, M. L. Belfer, Kalksma- van Lith, B. (2007). Psychosocial inter- N. Sartorius & A. Okasha (Eds.), The Mental ventions for children in war-a¡ected areas: the Health of Children and Adolescents: An Area of Global state of the art. Intervention, 5(1), 3-17 doi: 10.1097/ Neglect (65^136). West Sussex, England: Wiley WTF.0b013e3280c264cd. & Sons. Karki,R.,Kohrt,B.A.&Jordans,M.J.D.(2009). Gavala, J. R. & Flett, R. (2005). In£uential Child Led Indicators: pilot testing a child factors moderating academic enjoyment/moti- participation tool for former child soldiers in vation and psychological wellbeing for Maori Nepal. Intervention, 7(2),92-109. university students at Massey University. New ZealandJournal of Psychology, 34,52-57. Lam, L. T. & Kirby, S. L. (2002). Is Emotional Intelligence an Advantage? An exploration of Hart, J., Galappatti, A., Boyden, J. & Armstrong, the impact of Emotional Intelligence on individ- M. (2007). Participatory tools for evaluating ual performance. The Journal of Social Psychology, psychosocial work with children in areas of 142(1),133-143 doi:10.1080/00224540209603891. armed con£ict: a pilot in eastern Sri Lanka. Inter- vention, 5(1), 41- 6 0. Parker, J. D. A., Creque, R. E., Barnhart, D. L., Harris, J., Majeski, S. A., Wood, L. M., Bond, B. HealthnetTPO & Centre forTrauma Psychology: J. & Hogan, M. J. (2004). Academic achievement Monitoring and Evaluation Package - version II. in high school: Does emotional intelligence mat- RetrievedMay2010fromhttp://www.healthnettpo. ter? Personality and Individual Di¡erences, 37,1321- org/¢les/695/m-e-format.pdf. 1330 doi:10.1016/j.paid.2004.01.002.

Copyright © War Trauma Foundation. Unauthorized reproduction of this article is prohibited.73 Psychosocial support for children in the Republic of South Sudan: an evaluation outcome Intervention 2014,Volume 12, Number 1, Page 61 - 75

Russell Bernard, H. (2002). Research Methods in http://www.undp.org/content/dam/southsudan/ Anthropology: Qualitative and Quantitative Approaches. library/Documents/CSAC%20Reports/Eastern- Oxford: Altamira Press. %20Equatoria-Consult-May-12.pdf

Si lve r m a n, D. (2011). Interpreting Qualitative Data. UNICEF (2011)Inter-Agency Guide to the Evalu- London: SAGE. ation of Psychosocial Programming in Emergen- cies. NewYork: United Nations Children’s Fund. Stark, L., Wessells, M., King, D., Lamin, D. & Lilley,S.(2012).A grounded approach to the de¢nition UNICEF (2009). Machel study 10-year strategic of population-based, child protection and wellbeing review: children and con£ict in a changing world. outcome areas. London, England: Interagency NewYork: United Nations Children’s Fund. Learning Initiative on Community-Based Child Protection Mechanisms and Child Protection Van der Merwe, N. (2005).The Relationship between Systems. psychosocial wellbeing and academic performance of university students (Mini-dissertation). Potchefstroom, Tol, W. A., Barbui, C., Galappatti, A., Silove, D., South Africa: North-West University. Betancourt, T. S., Souza, R., Golaz, A. & van Ommeren, M. (2011a).Mental health and psycho- Wessells, M. & Monteiro, C. (2006). Psychosocial social support in humanitarian settings: linking Assistance for Youth: Toward Reconstruction for practice and research. Lancet, 378,1581-1591 Peace in Angola. Journal of Social Issues,62,121-139. doi:10.1016/S0140 - 6736(11)61094 -5. Wessels, M. & Monteiro, C. (2001). In: Christie, Tol,W. A., Patel,V.,Tomlinson, M., Baingana, F., D.J., Wagner, R.V., Winter, D.A., (Eds.). (2001). Galappatti, A., Panter-Brick, C., Silove, D., Peace, Con£ict, and Violence: Peace Psychology for the Sondorp, E., Wessells, M. & van Ommeren, M. 21st Century. Englewood Cli¡s, New Jersey: Pre- (2011b). Research priorities for mental health and ntice-Hall. psychosocial support in humanitarian settings. World Bank, (2012). Educationin the Republicof South PLoS Medicine, 8(9), 1001096 doi:10.1371/journal. Sudan: status and challenges for a new system.Africa pmed.1001096. human development series; Africa education Tol, W. A., Jordans, M. J. D., Reis, R. & de Jong, country status report. Washington, DC: World (2009). Ecological resilience: working with child Bank. related psychosocial resources in war-a¡ected World Bank, (2013). South Sudan Overview. communities. In: D. Brom, R. Pat-Horenczyk & Retrieved at http://www.worldbank.org/en/coun J.Ford (Eds.),Treatingtraumatizedchildren:risk,resili- try/southsudan/overview ence, and recovery. London: Routledge. Zins, J.E., Weissberg, R.P., Wang, M.C. & Turner-Stokes, L., Hurn, J., Kneebone, I. & Walberg, H.J. (Eds.) (2004). Building academic Cropley, M. (2006). Goal setting as an outcome success on social and emotional learning: What does measure: A systematic review. Clinical Rehabilita- the research say? New York, NY: Teachers College tion, 20(9),756-772 doi:10.1177/0269215506070793. Press. UN (2012). Report of the Secretary-General on South 1 Sudan 8 November 2012. Retrieved June 2013 from The I DEAL methodology can be found on the http://www.un.org/en/ga/search/view_doc.asp? WCH open-source platform www.warchildlear- symbol=S/2012/820&Lang=E ning.org. 2 See Monitoring and Evaluation Toolkit at UNDP (2012). Community consultation report www.warchildlearning.org/trainingtools/evalua Eastern Equatoria State South Sudan. Retrieved at tion.

Ellen Eiling, Drs, is Research, Monitoring and Evaluation Specialist at War Child Holland, the Netherlands. email: [email protected] MarianneVan Diggele-Holtland, MSc, isJunior Researcher atWarChild Holland, the Netherlands. TomvanYperen,PhD,isanExpertatYouthCareoftheNetherlandsYouthInstitute.HeisalsoProfessorof Monitoring and InnovationYouthCare at the State University, Groningen, the Netherlands. Frits Boer,MDPhD,is an Emeritus Professor of Child and Adolescent Psychiatry at the University of Amsterdam, the Netherlands.