University of North Texas
Race and Ethnicity, Cognitive Reserve, and Measurement Bias in Cognitive Measures: A Multiple Indicators, Multiple Causes Approach
Abstract
dc:descriptionThe global community is aging. Older adults in the United States are expected to comprise 23% of the population by 2025 (US Census Bureau, 2023a). With people of color also anticipated to increase from 38.4% of the US population in 2020 to 56.4% in 2060 (Colby & Ortman, 2014), demographic trends demand understanding the contribution of cognitive reserve and measurement bias towards cognitive health disparities in later life. Cognitive reserve represents an indicator of the brain’s resilience against neuropathology and is typically operationalized as educational attainment, occupational history, and engagement with leisure activities (Stern, 2002). The current study used the Health and Aging Brain Study: Health Disparities dataset to explore the impact of race/ethnicity and cognitive reserve on cognitive testing among older Black, Latino, and White Americans. I evaluated performance-based measures (i.e., American National Adult Reading Test, Spanish English Verbal Learning Test) for uniform and non-uniform differential item functioning (DIF) across 1) race/ethnicity and 2) levels of cognitive reserve using multiple indicators, multiple causes (MIMIC) models. Results demonstrated pervasive DIF across both race/ethnicity and cognitive reserve strata for both measures that far exceed the conventional threshold for acceptable partial invariance. Using biased instruments to support clinical decision-making contributes to misdiagnoses and incorrect estimations of impairment, with downstream effects on older adult autonomy, access to services, and legal decisions. In research settings, psychometric bias limits the validity of between-group comparisons and distorts observed differences. Taken together, these findings highlight the importance of understanding the psychometric properties of commonly used instruments and incorporating modern test theories into test development and validation.
Degree
thesis:*- Grantor dc:publisher
- University of North Texas
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Su, Charlie Chung-Yih 1986-
- Contributors dc:contributor
-
- Cicero, David C.
- Neumann, Craig S.
- Barnett, Michael D.
Subjects
dc:subject × 3Rights
dc:rights- Statement dc:rights
-
- Public
- Su, Charlie Chung-Yih 1986-
- Copyright
- Copyright is held by the author, unless otherwise noted. All rights Reserved.
- Language dc:language
- English
Identifiers
dc:identifier.*- Identifier
-
local-cont-no: submission_4921
ark: ark:/67531/metadc2652432