Psychometric analysis of the Ten-Item Perceived Stress Scale

Psychol Assess. 2015 Mar;27(1):90-101. doi: 10.1037/a0038100. Epub 2014 Oct 27.


Although the 10-item Perceived Stress Scale (PSS-10) is a popular measure, a review of the literature reveals 3 significant gaps: (a) There is some debate as to whether a 1- or a 2-factor model best describes the relationships among the PSS-10 items, (b) little information is available on the performance of the items on the scale, and (c) it is unclear whether PSS-10 scores are subject to gender bias. These gaps were addressed in this study using a sample of 1,236 adults from the National Survey of Midlife Development in the United States II. Based on self-identification, participants were 56.31% female, 77% White, 17.31% Black and/or African American, and the average age was 54.48 years (SD = 11.69). Findings from an ordinal confirmatory factor analysis suggested the relationships among the items are best described by an oblique 2-factor model. Item analysis using the graded response model provided no evidence of item misfit and indicated both subscales have a wide estimation range. Although t tests revealed a significant difference between the means of males and females on the Perceived Helplessness Subscale (t = 4.001, df = 1234, p < .001), measurement invariance tests suggest that PSS-10 scores may not be substantially affected by gender bias. Overall, the findings suggest that inferences made using PSS-10 scores are valid. However, this study calls into question inferences where the multidimensionality of the PSS-10 is ignored.

MeSH terms

  • Adult
  • African Americans / psychology*
  • African Americans / statistics & numerical data
  • European Continental Ancestry Group / psychology*
  • European Continental Ancestry Group / statistics & numerical data
  • Factor Analysis, Statistical
  • Female
  • Helplessness, Learned
  • Humans
  • Male
  • Middle Aged
  • Psychometrics*
  • Reproducibility of Results
  • Self Report*
  • Sex Factors
  • Social Perception*
  • Stress, Psychological / epidemiology*
  • United States / epidemiology