Method effects in Ryff’s scales: a study in an elderly population.

Abstract

Although widely used, Ryff’s scales have had some problems of factorial validity and recent investigations have also shown method effects. However, these method effects have not been deeply investigated in special populations such as the elderly, the population under study in this paper. The aim of this research is twofold: to study the factorial structure of Ryff’s scale in its different versions; and to evaluate the presence of method effects associated to negatively worded items. In order to accomplish these aims, a sample of 556 Valencian elderly people living in their homes has been surveyed. All confirmatory models used in the literature to test for factorial validity have been tested. Maximum likelihood with robust corrections has been used to estimate the models. Best fitting model posited a six factor structure, as originally proposed by the author. With respect to method bias, results shown a consistent and non-trivial method factor effect associated to negatively worded items.