Measurement Invariance of a Quality-of-life Measure, CASP-12, within the English Longitudinal Study of Ageing (ELSA)

Abstract

AbstractCASP-12 is a frequently used quality of life scale for older people, but limited efforts have been made to test the factor structure or to explore the measurement consistency of the scale across key characteristics. The aim of this study is to examine if the CASP-12 questionnaire has a well-defined factor structure with a second-order structure factor nested within four first-order domains: control, autonomy, pleasure, and self-realization. The study also aims to investigates if this factor structure is interpreted similarly by respondents of different genders, ages, educational levels, net wealth, and at two time periods, using a multi-group confirmatory factor analysis (MG-CFA). The results show that CASP-12 with lower four first-order domains (CASP-12-4D) is consistent across genders and two time periods and satisfies the second-order strong-invariance criteria. Furthermore, the instrument demonstrates consistency in weak levels across three age groups (50–59, 60–69, and 70–90), educational levels and net wealth, but not strong invariance. The sample was divided into three subsamples based on age groups to address this issue. The consistency of CASP-12-4D has met the second-order strong invariance level requirement across gender, age, education level and two-time periods. Net wealth however still fails to meet the strong invariance levels. The CASP-12-4D version will suit social and public health research which controls for age and wealth status.