Psychometric properties of a Korean version of the Perceived Stress Scale (PSS) in a military sample

Perceived stress reflects a person's feeling of how much stress the individual is under at a given time. The Perceived Stress Scale (PSS) is a popular instrument measuring the extent to which individuals perceive situations in their life as excessive relative to the ability to cope. Based on a literature review, however, several issues related to the scale remain: (a) the dimensionality is not established, (b) little information about the individual items exists, and (c) much research is based on university student samples. To address these, this study evaluated the psychometric properties of the Korean version of the Perceived Stress Scale (KPSS) using a military sample.

This study was conducted in South Korea with 373 military personnel, aged 19-30 years. Both classical test theory (CTT) and the Rasch rating scale model were used to examine the psychometric properties of the KPSS, including factor structure, concurrent validity, reliability, and item analyses.

Internal consistency reliability for the overall and negative/positive perception subscales was.85, .85 and .86, respectively. Based on Rasch reliability, person and item reliability were .82 and .98, respectively. Person and item separation were 2.13 and 7.19, respectively. Concurrent validity was established, with significantly positive association with the measures of depression and negative association with the measure of life satisfaction. Findings from the CFA suggested that a bifactor model with two group factors was the best fit to the observed data. The RSM showed that all but one item had acceptable infit and outfit statistics, and item difficulty ranged from -.73 to 1.22. Besides, the RSM showed positive and moderate inter-item correlations ranging from .42 to .75.

The results provided evidence that a 10-item Korean version of the Perceived Stress Scale was a reliable and valid scale to measure perceived stress in military samples.