Factor Structure, Reliability, and Sex Invariance of the General Health Questionnaire (GHQ-12) in Peruvian University Students
Jacksaint Saintila, Percy G. Ruiz-Mamani, Marilú Elena Barreto Espinoza, Yaquelin E. Calizaya-MillaBackground
The General Health Questionnaire (GHQ-12) is widely used to assess psychological distress across cultural contexts. However, psychometric evidence in Peruvian university students remains limited. This study examined the factorial structure, reliability, and sex invariance of the GHQ-12 in this population.
Methods
A cross-sectional study was conducted with 768 Peruvian university students who completed the GHQ-12. Item-level descriptive analyses, exploratory factor analysis (EFA), confirmatory factor analysis (CFA), reliability estimation, and multigroup CFA by sex were performed. Internal consistency was estimated using ordinal alpha.
Results
EFA supported by parallel analysis indicated a unidimensional structure as the optimal empirical solution. However, considering previous evidence for both unidimensional and bidimensional structures, competing CFA models were tested. Both models showed acceptable fit, but the two-factor model demonstrated superior fit indices (CFI = 0.991, TLI = 0.988, RMSEA = 0.050, SRMR = 0.053) compared with the one-factor model (CFI = 0.980, TLI = 0.975, RMSEA = 0.063, SRMR = 0.065). Multigroup CFA supported configural, metric, scalar, and strict invariance across sex. Reliability analyses showed adequate internal consistency for both factors and the total scale (ordinal α > 0.80).
Conclusion
The GHQ-12 showed adequate psychometric properties in Peruvian university students. Although EFA supported a unidimensional solution, CFA favored the bidimensional model, which also demonstrated sex invariance and satisfactory reliability. These findings support the GHQ-12 as a useful tool for assessing psychological distress in university settings.