A Holistic Exploration of Social Support Among Former Child Soldiers in Northern Uganda: A Mixed Methods Approach (WITHDRAWN)

Background and Purpose: This study employed an explanatory sequential mixed methods approach to examine how former child soldiers living in Northern Uganda experience social support in their community. This sequential explanatory design triangulates initial quantitative results (Phase 1) with a follow-up qualitative phase (Phase 2). The two sets of findings are then synthesized within a single discussion section.

Phase 1 relied on data from the Survey of War Affected Youth (SWAY). SWAY data were collected in Northern Uganda from July 2005-August 2007. SWAY assessed well-being, trauma and emotional symptoms, psychological distress, war experience, and social support. Analysis of Phase 1 data provided the context for Phase 2 data collection and analysis. In Phase 2, data were collected via face-to-face, semi-structured interviews with former child soldiers and service providers living in Northern Uganda.

Methods: Through secondary data analysis, initial descriptive statistics of Phase 1 data of abducted youth (n=1118) showed 33% were female, 47% had between 5-7 years of education, 5% were a forced wife, 39% passed through a reception center, and 79% returned to their family after their time in armed conflict. Using SPSS, assumptions were tested and met, and a multiple linear regression was calculated to predict social support based on 15 independent variables. Through an established community partner in Northern Uganda, Phase 2 proceeded with snowball and convenience sampling to recruit participants. Participants for Phase 2 included 20 former child soldiers (50% female) and 7 service providers (3 female). For the returnees, abduction lengths varied from 8 months to 19 years and ages ranged from 20-39 years old. All interviews were recorded for later transcription and analysis. Using Atlas.ti and a phenomenological approach to data analysis, horizontalization of the data, thematic clustering, and a composite description of the essence of the experience occurred.

Results: In regards to Phase 1, a significant regression equation was found (F(15, 559)= 28.437, p<.000), with an R² of .42. Significant correlates of social support were violence inflicted on family, forced to harm family or friend, being abducted more than 2 weeks, gender (being female), receiving amnesty, being employed, and educational attainment. All significant correlates were associated with an increase in social support. Phase 2 revealed several themes of importance: the unique situation of female former child soldiers and their children, the persistence and effects of income poverty, the importance of family acceptance, the presence of stigma and shame, and an overall desire for access to education.

Conclusions and Implications: When synthesized, findings from Phase 1 and Phase 2 suggest that more research is needed to better understand the unique situation of females in this context. While quantitative analysis shows increased access to social support due to being female or a child mother, qualitative data reveals extensive hardship among this group. With research showing the link between social support and health, it is imperative that social workers working with this or similar populations center social support as one way to influence positive life outcomes.