Is health locus of control a 3-factor or a 2-factor construct?

Although the Multidimensional Health Locus of Control (MHLC) scales are commonly used, uncertainty about their factorial validity still remains. In the present study, confirmatory factor analyses of a French-Canadian adaptation of the Multidimensional Health Locus of Control scales (MHLC) were conducted to compare in a non-clinical group of 224 adults two factor models. Multigroup analyses were also conducted using this group and a clinical group of 132 diabetics to assess the equivalence of the MHLC factor structure across these two groups. A 3-factor model postulating an internal, external, and chance dimension, and accounting for measurement errors provided the best fit to the data. Multigroup analyses failed to support the equivalence of the MHLC factor loadings across the non-clinical and clinical group. These findings suggest that comparisons of MHLC scores across such groups may be problematic.