Korean versions of thePerceivedStressScale (PSS‐14, 10 and 4): psychometric evaluation in patients with chronic disease

Background The P erceived S tress S cale ( PSS ) is a representative instrument used to measure stress. The original PSS comprises 14 items ( PSS ‐14) in two subscales, but 10‐ and 4‐item versions are also available ( PSS ‐10 and 4, respectively). The target populations of psychometric studies using the PSS have far mainly comprised college students, and the underlying constructs of the PSS versions are controversial: one factor vs. two factors and first order vs. second order. Objective The aim of this study was to evaluate the psychometric properties of the K orean versions of the PSS ‐14, 10 and 4 (designated KPSS ‐14, 10 and ‐4, respectively) in patients with chronic disease. Methods The PSS ‐14, 10 and 4 were translated into K orean using forward and backward translation. Factorial construct validity was tested using both exploratory and confirmatory factor analyses. Item convergent validity and item discriminant validity were tested. Concurrent validity was examined using the C enter for E pidemiologic S tudies– D epression scale. Known‐groups validity was analysed using t ‐test and effect size. Reliability was tested using C ronbach's alpha and the intraclass correlation coefficient. Results Exploratory factor analysis supported a two‐factor model for all K orean versions of the PSS , and confirmatory factor analysis indicated that the model fit the KPSS ‐10 well and the KPSS ‐4 only marginally. The testing of item convergent and discriminant validity revealed a 100% scaling success. As expected, all scores in the KPSS ‐14, 10 and 4 were moderately correlated with depression scores and differed significantly according to gender. The C ronbach's alpha for the KPSS ‐14 and 10 exceeded the criterion of 0.70. The intraclass correlation coefficient values of all three K orean versions were satisfied. Conclusions The KPSS ‐10 exhibited a first‐order, two‐factor construct, and excellent reliability and validity were established for K orean patients with chronic disease. The psychometric properties of the shortest version, KPSS ‐4, were only marginally acceptable.