Using contemporary psychometric methods to construct a concise version of the psychological strain scale.

Area of Science:

• Psychological assessment
• Educational psychology
• Psychometrics

Background:

• The original Psychological Strain Scale (PSS-40) is a comprehensive measure.
• There is a need for a more concise yet reliable and valid assessment tool for psychological strain.

Purpose of the Study:

• To develop a shortened version of the PSS-40, named PSS-20.
• To integrate Classical Test Theory (CTT) and Item Response Theory (IRT) for scale refinement.

Main Methods:

• Applied PSS-40 to 10478 college students.
• Utilized CTT (variability, correlation, factor analysis, Cronbach's alpha) and IRT (Samejima graded response model) for item analysis.
• Selected 20 items based on combined CTT and IRT evaluation.

Main Results:

• Exploratory factor analysis showed high factor loadings (>0.6) and cumulative variance (>66%) for PSS-20 items.
• Confirmatory factor analysis indicated good model fit (RMESA=0.057, RMR=0.028).
• PSS-20 demonstrated excellent reliability (Cronbach's alpha=0.939) and criterion validity (0.558).

Conclusions:

• The PSS-20 scale is a reliable and valid instrument.
• PSS-20 is suitable for evaluating psychological strain in college students.