Abstract
OBJECTIVE: The Cognitive Abilities Screening Instrument (CASI) is a screening test of global cognitive function used in research and clinical settings. However, the CASI was developed using face validity and has not been investigated via empirical tests such as factor analyses. Thus, we aimed to develop and test a parsimonious conceptualization of the CASI rooted in cognitive aging literature reflective of crystallized and fluid abilities. DESIGN: Secondary data analysis implementing confirmatory factor analyses where we tested the proposed two-factor solution, an alternate one-factor solution, and conducted a χ(2) difference test to determine which model had a significantly better fit. SETTING: N/A. PARTICIPANTS: Data came from 3,491 men from the Kuakini Honolulu-Asia Aging Study. MEASUREMENTS: The Cognitive Abilities Screening Instrument. RESULTS: Findings demonstrated that both models fit the data; however, the two-factor model had a significantly better fit than the one-factor model. Criterion validity tests indicated that participant age was negatively associated with both factors and that education was positively associated with both factors. Further tests demonstrated that fluid abilities were significantly and negatively associated with a later-life dementia diagnosis. CONCLUSIONS: We encourage investigators to use the two-factor model of the CASI as it could shed light on underlying cognitive processes, which may be more informative than using a global measure of cognition.