Chapter 4 124 Correlations were calculated between all variables to test whether BPP is positively associated with alexithymia, SIB, depressive symptomatology, and dissociative experiences (hypothesis 1) and whether alexithymia is positively associated with SIB, depressive symptomatology, and dissociative experiences (hypothesis 2). Separate simple mediation analyses were conducted for all variables to test whether the associations between BPP and respectively SIB, depressive symptomatology, and dissociative experiences are mediated by alexithymia (hypothesis 3). As there were two measures of alexithymia (i.e., TAS-20 and TSIA) all analyses were conducted twice. Mediation analyses were performed using the PROCESS v3.2 add-on for SPSS (Hayes and Preacher 2014). To test for mediation the bias-corrected bootstrap method was used. PROCESS yields estimates of indirect, direct, and total effects. Complete mediations are inferred when the indirect effect is significant, and the direct effect is not significant; partial mediations are inferred when both indirect and direct effects are significant; no mediation is inferred when the indirect effect is not significant. Default settings of 95 % confidence intervals for indirect effects and 10000 samples for bootstrapping was used. For the indirect effects tests, CIs that do not include zero suggest significant mediation [114]. 4. Results 4.1 Simple correlations See table 3 for an overview of all correlations. Calculation of Spearman rank ordered correlations confirmed positive relationships between BPP and alexithymia, BPP and DEP, BPP and DIS, BPP and measures of direct SIB (H1). A positive relationship was not confirmed for BPP and indirect SIB, although it only just failed to reach the required significance level (rs = .22, p < .059). Regarding expected relationships between alexithymia and other clinical variables (H2), results of the analysis confirmed positive relationships between alexithymia measures and DEP. A positive relationship between alexithymia and DIS was confirmed when alexithymia was assessed by self-report, but not when assessed via clinical interview. A relationship between alexithymia and SIB was only confirmed for direct SIB in the past year and only when alexithymia was measured via clinical interview. 4.2 Mediation analyses Next, we examined whether the associations of BPP with DEP, and DIS and SIB are mediated by alexithymia. First the analyses using the TAS-20 total scores for alexithymia are presented. Results of the analyses for all variables can be found in Table 4.
RkJQdWJsaXNoZXIy MjY0ODMw