Abstract

Purpose The purpose of this study was to apply Rasch analysis to examine the unidimensionality and reliability of the General Self-Efficacy Scale (GSE) in workers with traumatic limb injuries. Furthermore, if the items of the GSE fitted the Rasch model’s assumptions, we transformed the raw sum ordinal scores of the GSE into Rasch interval scores. Methods A total of 1076 participants completed the GSE at 1 month post injury. Rasch analysis was used to examine the unidimensionality and person reliability of the GSE. The unidimensionality of the GSE was verified by determining whether the items fit the Rasch model’s assumptions: (1) item fit indices: infit and outfit mean square (MNSQ) ranged from 0.6 to 1.4; and (2) the eigenvalue of the first factor extracted from principal component analysis (PCA) for residuals was <2. Person reliability was calculated. Results The unidimensionality of the 10-item GSE was supported in terms of good item fit statistics (infit and outfit MNSQ ranging from 0.92 to 1.32) and acceptable eigenvalues (1.6) of the first factor of the PCA, with person reliability = 0.89. Consequently, the raw sum scores of the GSE were transformed into Rasch scores. Conclusions The results indicated that the items of GSE are unidimensional and have acceptable person reliability in workers with traumatic limb injuries. Additionally, the raw sum scores of the GSE can be transformed into Rasch interval scores for prospective users to quantify workers’ levels of self-efficacy and to conduct further statistical analyses.

Original languageEnglish
Pages (from-to)332-339
Number of pages8
JournalJournal of Occupational Rehabilitation
Volume26
Issue number3
DOIs
Publication statusPublished - Sept 2016

Keywords

  • Rasch analysis
  • Self-efficacy
  • Traumatic limb injury
  • Unidimensionality

ASJC Scopus subject areas

  • Rehabilitation
  • Occupational Therapy

Fingerprint

Dive into the research topics of 'Rasch Analysis of the General Self-Efficacy Scale in Workers with Traumatic Limb Injuries'. Together they form a unique fingerprint.

Cite this