Connor M Sheehan1, Elliot M Tucker-Drob2. 1. Population Research Center and Department of Sociology, University of Texas at Austin. 2. Population Research Center and Department of Psychology, University of Texas at Austin.
Abstract
OBJECTIVES: The ability of older adults to live independently is often assessed with a battery of questions known as Instrumental Activities of Daily Living (IADLs). Many of these questions query the difficulty conducting household activities that have been predominantly conducted by women (e.g., the ability to prepare a meal), especially for cohorts now in old age. Although previous research has documented gender differences in IADL limitations, it has not been documented whether IADLs equivalently measure the same latent construct for men and women. METHODS: We apply psychometric tests of measurement invariance to data from the 1998 Health and Retirement Study. We then estimate corrected models that account for violations of measurement invariance across genders. RESULTS: We find that IADLs do not equivalently measure same latent construct for men and women. We find that men are more likely not to do the IADL activities for reasons unrelated to health limitations, which may reflect gendered expectations regarding household activities. Accounting for this we still find that women report greater health-related IADL limitations than men. DISCUSSION: Researchers should be cautious making gender comparisons for IADLs without attending to the gender-specific measurement properties of many of the items of which the IADL is comprised.
OBJECTIVES: The ability of older adults to live independently is often assessed with a battery of questions known as Instrumental Activities of Daily Living (IADLs). Many of these questions query the difficulty conducting household activities that have been predominantly conducted by women (e.g., the ability to prepare a meal), especially for cohorts now in old age. Although previous research has documented gender differences in IADL limitations, it has not been documented whether IADLs equivalently measure the same latent construct for men and women. METHODS: We apply psychometric tests of measurement invariance to data from the 1998 Health and Retirement Study. We then estimate corrected models that account for violations of measurement invariance across genders. RESULTS: We find that IADLs do not equivalently measure same latent construct for men and women. We find that men are more likely not to do the IADL activities for reasons unrelated to health limitations, which may reflect gendered expectations regarding household activities. Accounting for this we still find that women report greater health-related IADL limitations than men. DISCUSSION: Researchers should be cautious making gender comparisons for IADLs without attending to the gender-specific measurement properties of many of the items of which the IADL is comprised.
Authors: Sandra Bell-McGinty; Kenneth Podell; Michael Franzen; Anne D Baird; Michael J Williams Journal: Int J Geriatr Psychiatry Date: 2002-09 Impact factor: 3.485
Authors: Tina E Brinkley; Shana D Stites; Holly C Hunsberger; Carrie A Karvonen-Gutierrez; Mengting Li; C Elizabeth Shaaban; Roland J Thorpe; Stephen B Kritchevsky Journal: Innov Aging Date: 2022-08-21