Development of a self-administered dementia checklist (SDC) (1): Examination of factorial validity and internal reliability

Abstract

Aim: The aim of this study was to develop a self-administered dementia checklist (SDC), in order to help community-residing older adults realize their declining functions and encourage them to begin using necessary services, and to examine its factorial validity and internal reliability. Methods: A panel of dementia clinical experts developed a questionnaire according to pre-selected items and conducted a selfadministered survey with community-residing people aged 65+(n = 2,483). The team developed a scale through an exploratory factor analysis and item response theory (IRT) analysis (Study 1). Using this scale, they conducted a self-administered survey with community-residing people aged 65+(n = 5,199), conducted another exploratory factor analysis, and developed a 10-item scale. A confirmatory factor analysis was subsequently conducted and reliability coefficients were computed. Results: The exploratory factor analysis of the proposed 37 items extracted 5 factors: Factor 1 was named “subjective decline in daily living functioning,” and Factor 2 was “subjective cognitive decline” in the early stage of dementia. The team developed a 20-item scale by selecting 10 items from each factor which had high factor loadings and high slope values in the IRT analysis (Study 1). After the exploratory factor analysis of the 20-item scale, they developed a 10-item scale by selecting 5 items from each factor which had strong associations. The confirmatory factor analysis verified the 2-factor model. The Cronbach α coefficients for the subscales of Factors 1 and 2 were 0.935 and 0.834, respectively, and 0.908 for the overall 10-item scale. Conclusion: The authors developed a 10-item SDC with 2 factors and confirmed its factorial validity and internal reliability.