Latent variable analyses of age trends of cognition in the Health and Retirement Study, 1992-2004.

TitleLatent variable analyses of age trends of cognition in the Health and Retirement Study, 1992-2004.
Publication TypeJournal Article
Year of Publication2007
AuthorsMcArdle, JJ, Fisher, GG, Kadlec, KM
JournalPsychol Aging
Volume22
Issue3
Pagination525-545
Date Published2007 Sep
ISSN Number0882-7974
Call Numbernewpubs20071002_CognitiveAgeTrends2007.pdf
KeywordsAged, Aged, 80 and over, Aging, Cognition, Female, Health Surveys, Humans, Individuality, Longitudinal Studies, Male, Mental Recall, Mental Status Schedule, Middle Aged, Models, Statistical, Neuropsychological tests, Psychometrics, Retention, Psychology, Sex Factors, United States
Abstract

The present study was conducted to better describe age trends in cognition among older adults in the longitudinal Health and Retirement Study (HRS) from 1992 to 2004 (N = 17,000). The authors used contemporary latent variable models to organize this information in terms of both cross-sectional and longitudinal inferences about age and cognition. Common factor analysis results yielded evidence for at least 2 common factors, labeled Episodic Memory and Mental Status, largely separable from vocabulary. Latent path models with these common factors were based on demographic characteristics. Multilevel models of factorial invariance over age indicated that at least 2 common factors were needed. Latent curve models of episodic memory were based on age at testing and showed substantial age differences and age changes, including impacts due to retesting as well as several time-invariant and time-varying predictors.

DOI10.1037/0882-7974.22.3.525
User Guide Notes

http://www.ncbi.nlm.nih.gov/pubmed/17874952?dopt=Abstract

Endnote Keywords

Cognition/Methodology

Endnote ID

17950

Alternate JournalPsychol Aging
Citation Key7153
PubMed ID17874952
Grant ListU01 AG009740 / AG / NIA NIH HHS / United States
AG 007137 / AG / NIA NIH HHS / United States
AG 027010 / AG / NIA NIH HHS / United States