43 for the younger cohort \u03c7 2 1 N 569 009 p 076 Finally racialethnic

43 for the younger cohort χ 2 1 n 569 009 p 076

This preview shows page 6 - 8 out of 15 pages.

4.3 % for the younger cohort, χ 2 (1, N 0 569) 0 0.09, p 0 0.76). Finally, racial/ethnic differences were also evident in both the likelihood for exposure to violence and the likelihood of PTSD. Recent violence was more common among African- American youth (57 %) compared to adolescents who iden- tified as Latino (43 %), White (33 %), or Other racial/ethnic groups (33 %), χ 2 (3, N 0 1240) 0 38.19, p <0.001. In con- trast, among those who experienced violence, African- American adolescents had lower rates of PTSD (2 %) than Latino (5 %) and White (8 %) adolescents. Next, five MANOVAs were run to identify Time 1 var- iables that differed for the three groups (no violence expo- sure, violence exposure without PTSD, violence exposure with PTSD) in order to include these factors in discriminant function analysis. Race, gender, and age were used as cova- riates. Dependent variables were grouped to reflect commu- nity risk (Multivariate F (4, 2310) 0 2.32, p 0 0.06, η 2 0 0.005), family context (Multivariate F (8, 2375) 0 3.32, p <0.01, η 2 0 0.01), adolescent behavioral adjustment (Multivariate F (10, 2372) 0 8.25, p <0.001, η 2 0 0.03), cognitive vulner- abilities (Multivariate F (8, 2316) 0 5.83, p <0.001, η 2 0 0.02), and interpersonal problems (Multivariate F (10, 2378) 0 11.83, p <0.001, η 2 0 0.05). Multivariate F values were significant (with bonferroni correction of p 0 0.01) in all MANOVAs, except for community risk factors. Tables 1 and 2 present group mean scores and results from follow-up univariate F tests. As shown, the following Time 1 variables differed significantly by group: previous interpersonal violence, parent report of family adversity and family conflict, adolescent report of family support, CBC and YSR Externalizing, YSR Internalizing, EASI Impulsivity, CBC and YSR Thought Disorder, CBC and YSR Social Problems, and involvement with deviant peers. These variables, as well as age, gender, and African-American race, were entered into a discriminant function analysis to determine if they could jointly distinguish the three groups. 344 J Abnorm Child Psychol (2013) 41:339 353
Image of page 6
Both discriminant functions generated in analyses were statistically significant, indicating that the included set of variables could significantly distinguish adolescents in all three groups. Together, the two functions correctly classified Table 1 Group differences in baseline demographic, community and family factors Characteristic: No violence Exposure ( n 0 673) Violence exposure, No PTSD ( n 0 546) Violence exposure, PTSD ( n 0 23) Univariate test of difference Demographic Factors Age 13.35 ( SD 0 1.5) 13.66 ( SD 0 1.5) 13.72 ( SD 0 1.5) F (2, 1238) 0 6.66** Gender, % χ 2(2, N 0 1242) 0 7.68* -Female 58 % 40 % 2 % -Male 50 % 48 % 2 % Race/ethnicity, % χ 2(2, N 0 1242) 0 44.78** -African-American 43 % 56 % 1 % -Latino 57 % 40 % 3 % -White 67 % 30 % 3 % -Other 67 % 33 % 0 % Community Context Community SES 1.96(0.77) 1.81(0.75) 1.91(0.79) F (2, 1238) 0 2.15 Observed neighborhood quality 1.86(0.59) 2.01(0.65) 2.03(0.73) F (2, 1169) 0 4.09 Family Factors Family SES composite 0.07(1.44) 0.14(1.40) 0.44(1.41) F (2, 1238) 0 0.86 Family adversity 1.46(1.76) 1.77(1.81) 2.04(1.55) F (2, 1220) 0 5.63** Family conflict 46.87(10.05) 48.63(10.58) 50.13(10.60) F (2, 1198) 0 4.21* Perceived Family Support 2.74(0.29) 2.68(0.33) 2.56(0.52) F (2, 1218) 0 8.49** * p <0.01 ** p <0.001 Table 2 Group differences in behavioral adjustment, cognitive vulnerabilities, and interpersonal problems Characteristic: No violence Exposure ( n 0 673) Violence exposure, No PTSD ( n 0 546) Violence exposure, PTSD ( n 0 23)
Image of page 7
Image of page 8

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture