Exposure to Political Conflict and Violence and Posttraumatic Stress in Middle East Youth: Protective Factors

Abstract

We examine the role of family- and individual-level protective factors in the relation between exposure to ethnic-political conflict and violence and posttraumatic stress among Israeli and Palestinian youth. Specifically, we examine whether parental mental health (lack of depression), positive parenting, children's self-esteem, and academic achievement moderate the relation between exposure to ethnic-political conflict/violence and subsequent posttraumatic stress (PTS) symptoms. We collected three waves of data from 901 Israeli and 600 Palestinian youths (three age cohorts: 8, 11, and 14 years old; approximately half of each gender) and their parents at 1-year intervals. Greater cumulative exposure to ethnic-political conflict/violence across the first 2 waves of the study predicted higher subsequent PTS symptoms even when we controlled for the child's initial level of PTS symptoms. This relation was significantly moderated by a youth's self-esteem and by the positive parenting received by the youth. In particular, the longitudinal relation between exposure to violence and subsequent PTS symptoms was significant for low self-esteem youth and for youth receiving little positive parenting but was non-significant for children with high levels of these protective resources. Our findings show that youth most vulnerable to PTS symptoms as a result of exposure to ethnic-political violence are those with lower levels of self-esteem and who experience low levels of positive parenting. Interventions for war-exposed youth should test whether boosting self-esteem and positive parenting might reduce subsequent levels of PTS symptoms.

Acknowledgments

This research was supported by a grant from the National Institute of Child Health and Human Development (Grant No. HD047814; L. Rowell Huesmann, Principal Investigator).

Notes

¹Parents of children in the 8-year-old cohort provided reports of their children's exposure to ethnic-political conflict in each wave, but children in the older cohorts (11- and 14-year-olds in Wave 1) provided self-reports. We followed this strategy because our Institutional Review Board had concerns about the 8-year-olds’ emotional reactions to reporting on their own exposure. Also, given time constraints on interviews with young children, having parents report on these items decreased the length of the interview for 8-year-olds. To examine the comparability of children's and parents’ reports of children's exposure to political conflict/violence, we administered the measures to both children and parents of the youngest cohort in Wave 3 and found them to be highly correlated (r = .68).

²Following La Greca et al. (2010), as an alternative to FIML to estimate missing data, we replicated the correlation and regression analyses using a multiple imputation procedure in Mplus, generating 10 imputed data sets. This procedure did not affect the significance of the correlations or the regression results.

³As an alternative approach to dealing with nonnormality, we used the bootstrapping procedure in AMOS, requesting 200 bootstrapped samples and bias-corrected confidence intervals for each of the parameter bootstrap estimates. This procedure did not affect the conclusions based on the robust maximum likelihood estimation results.

Note: A three-way (Age × Sex × Ethnic Group) analysis of variance was computed for each variable; Ns range from 1,238 (Posttraumatic Stress W3) to 1,501 (Posttraumatic Stress W1) across analyses. Post hoc multiple comparison (least significant differences) tests were computed between means of subgroups defined by age cohort, sex, and ethnic group. Within a comparison, means with different subscripts are significantly different at p < .05. Pol Viol. = political conflict/violence; W1 = Wave 1; Pal = Palestinians; Isr-J = Israeli Jews; Isr-A = Israeli Arabs; W 1 & 2 = Waves 1 and 2 averaged; W3 = Wave 3.

Note. W1 = Wave 1; Pol. Viol. = political conflict/violence; W 1 & 2 = Waves 1 and 2 averaged; W3 = Wave 3.

*p < .05. **p < .01.

^a 0 = female, 1 = male.

^b 1 = Palestinian, 0 = not Palestinian.

^c 1 = Israeli Arab, 0 = not Israeli Arab.

*p < .05. **p < .01.