Psychometric properties of the Czech version of the Personality Inventory for DSM-5: internal consistency, validity and discrimination capacity of the measure
Abstract
Objectives. In the Alternative DSM-5 Model for Personality Disorders, the diagnosis of specific personality disorders is obtained through Personality Inventory for DSM-5 (PID-5) assessment of dimensional personality traits associated in 25 "lower order" facets and 5 "higher order" domains.
The aim of the study was to test PID-5 psychometric properties (internal consistency, validity and discrimination capacity). Participants and setting.
PID-5 was administered to a community volunteers sample (n = 351) and a clinical sample of psychiatric patients (n = 143) individually and in a group setting using pen-paper method and online data collection. 33 respondents completed the inventory twice to check test-retest reliability. Hypotheses.
Evidence will be found to support internal consistency and convergent validity of the PID-5 personality trait domains, as well as their stability in time. Moreover, significant mean-level differences will be found between the two samples.
Statistical analyses. The data were analyzed with parallel analysis, Pearson's correlation coefficient, t-tests, and ROC analysis.
Results. The study results confirmed excellent internal consistency and satisfactory reliability of all five PID-5 domains in time.
Unidimensional structure of the measure was confirmed for 22 of 25 facets. Significant correlations of all trait facets of the tested domains (r = 0.53 - 0.65 for the community sample, and r = 0.43 - 0.66 for the clinical sample) confirmed convergent validity.
Significant differences between the two samples in the test scores of four of the five tested domains (absolute value of effect size d = 0.74 - 1.34 within 95% CI) show good discrimination capacity of the method. Study limitations.
Two different forms of administration; prevalence of students in the community sample; different test-retest time periods.