A Latent Structural Comparison Between the Circumplex Scales of Interpersonal Problems (CSIP) and the Inventory of Interpersonal Problems (IIP-C).

Abstract

This study investigates differences in measuring interpersonal problems using two established instruments: the Circumplex Scales of Interpersonal Problems (CSIP) and the Inventory of Interpersonal Problems (IIP-C). We developed a German translation of the CSIP and compared it to the German IIP-C in a community sample representative of the German general population regarding age and gender (NT1 = 677, NT2 = 514). To compare the two instruments, we used the extension to the circular stochastic process model for the circumplex (SPMC-E). This method makes it possible to map the scales from one instrument onto the latent circumplex structure of the other and to examine how clinically relevant covariates (i.e., dysphoria and personality dysfunction) relate to the latent circumplex as assessed by either instrument. Both instruments demonstrated good fit to a latent quasi-circumplex, close alignment with theoretically expected scale locations, and high one-month test-retest reliability (r > .70). However, notable differences emerged: the CSIP covered more maladaptive variants of dominant behavior than the IIP-C, and dysphoria and personality dysfunction were associated with a submissive-cold style in the CSIP, but not the IIP-C. These findings reflect meaningful differences in item content and show how the two questionnaires could complement each other in measuring interpersonal problems.