Abstract
OBJECTIVE: The California Verbal Memory Learning Test, second edition (CVLT-II) is used to assess learning and memory impairment. However, CVLT psychometric validity is not established for American Indian populations, who have high risk, as well as important sociocultural considerations in standardized testing. METHOD: The Strong Heart Study conducted repeated CVLT-II short form testing among American Indian Elders across 3 U.S. regions in 2010-2013 (n = 801) and 2017-2019 (N = 386), in addition to sociodemographics and other neuropsychological testing. CVLT-II short form provides four main indices of delayed memory: 30-s (short), 10-min (long), cued, and recognition-discrimination. Statistical analyses examined inferences of CVLT-II short form delayed memory score validity with confirmatory factor analysis, generalizability and test-retest reliability with Pearson correlations, and extrapolation with path analysis. RESULTS: Participants were generally older (age 65-95), majority female, and majority with education 12-14 years. Fewer than 5% had missing data. Unidimensional single factor model fits were adequate (RMSEA 0.07, 90% CI 0.03-0.11; comparative fit index 0.99). Within-group and within-visit raw score comparisons were statistically significant for age, sex, and education, suggesting important score differences. Primary path analysis model suggested that age, sex, education, bilingual, and depression accounted for 16% of variance in test performance. CONCLUSIONS: Overall, these findings suggest that CVLT-II short form has good test structure in American Indian elders, but with needed contextualization by age, sex, and education. These findings offer support for CVLT in clinical and research use but also need for further research on cultural and community factors affecting standardized cognitive testing. (PsycInfo Database Record (c) 2026 APA, all rights reserved).