Race/ethnicity and the measurement of cognition in NSHAP: Recommendations for robustness.

Abstract

In this study, we examine the measurement of cognition in different racial/ethnic groups to move towards a less biased and more inclusive set of measures for capturing cognitive change and decline in older adulthood. We use data from Round 2 (N=3377) and Round 3 (N=4777) of the National Social Life, Health, and Aging Project (NSHAP) and examine the study's Survey Adjusted version of the Montreal Cognitive Assessment (MoCA-SA). We employ exploratory factor analyses to explore configural invariance by racial/ethnic group. Using modification indexes, two-parameter item response theory models, and split-sample testing, we identify items that seem robust to bias by race. We test the predictive validity of the full (18-item) and short (4-item) MoCA-SAs using self-reported dementia diagnosis, instrumental activities of daily living (IADLs), proxy reports of dementia, proxy reports of dementia-related death, and National Death Index reports of dementia-related death. We found that four measures out of the 18 used in NSHAP's MoCA-SA formed a scale that was more robust to racial bias. The shortened form predicted consequential outcomes as well as NSHAP's full MoCA-SA. The short form was also moderately correlated with the full form. Although sophisticated structural equation modeling techniques would be preferrable for assuaging measurement invariance by race in NSHAP, the shortened form of the MoCA-SA provides a quick way for researchers to carry out robustness checks and to see if the disparities and associations by race they document are "real" or the product of artifactual bias.