Page 109 - Personality disorders and insecure attachment among adolescents

starters group (M = 173.6, SD = 65.5) were also significant (f = 4.46, p = .015). Post Hoc test showed significant difference (p = .014) between the NSSI group and the no NSSI group. Comparison of NSSI groups’ coping skills Comparing the three NSSI groups using the CERQ at pre-treatment showed significant differences on three scales of the CERQ: Self-blame (NSSI group (M = 14.8, SD = 3.7), no NSSI (M =10.8, SD = 4.5), NSSI starters (M = 12.7, SD = 4.2), F = 8.72, p < .001). Positive refocusing (NSSI group (M = 8.7, SD = 4.0), no NSSI (M =11.1, SD = 4.2), NSSI starters (M = 10.4, SD = 3.3), F = 3.19, p = .046). Positive reappraisal (NSSI group (M = 8.6, SD = 3.2), no NSSI (M =11.8, SD = 4.9), NSSI starters (M = 12.7, SD = 3.5), F = 9.28, p < .001). Post Hoc test showed significant differences (p = <.001) between the NSSI and the no NSSI group on Self-blame and Positive reappraisal. Logistic regression The characteristics gender, number of personality disorders, total score of SCL-90 and the coping skills self-blame, positive refocusing and positive reappraisal differed significantly between the NSSI and the non-NSSI group and were entered into a logistic regression equation. The logistic regression analysis was performed to test the predictive value of the variables on (the dichotomous dependent variable) NSSI versus non-NSSI. Only gender, self-blame and positive reappraisal significantly predicted membership of the NSSI-group. The model shows the bivariate odds ratios. Three variables (gender, self-blame and positive reappraisal) significantly predicted membership of the NSSI-group. The model as a whole explained 38 % (Nagelkerke R square) of the variance in NSSI, and correctly identified 80.4 % of cases. Table 3 shows the results of this logistic regression analysis. Table 3: Predictors of NSSI B (SE) 95% CI for Odds Ratio Lower Odds Ratio Upper Constant Gender (female) Self-blame Positive reappraisal NSSI = Non-Suicidal Self-Injury -0.21* (0.07) 0.71 R2 = .38 (Nagelkerke). Model chi2 (3) = 30.76 * p < .05 -0.91 (1.32) 1.71* (0.80) 1.16 0.19* (0.07) 1.06 5.50 26.14 1.20 1.36 0.81 0.92 Frequency of NSSI in the NSSI group In the NSSI group, the percentage of participants that reported no NSSI in the last month significantly increased from 23.5% (n = 16) at pre-treatment to 63.2% (n = 43) at post-treatment (Mc 105