Simulative and empirical comparison of health scales’ scores derived by confirmatory factor analysis and sum of items.

We focus on how combining questionnaire’s items to derive a score. Data simulation and real data were used to compare questionnaires’ scores derived by sum of items and scores derived from a confirmatory factor analysis. We investigated the equivalence of the sum of items and the latent factor scores from confirmatory factor analysis in simulative and empirical studies. We investigated to what extent the derivation of the score may have influenced the assumptions of linear model analysis. Deriving the questionnaire’s scores using the sum of items or the application of the confirmatory factor analysis is equivalent. This evidence is confirmed in the simulation settings and by empirical data.