Liat Ayalon1. 1. Louis and Gabi Weisfeld School of Social Work, Bar Ilan University, Ramat Gan, Israel. liatayalon0@gmail.com.
Abstract
OBJECTIVES: The present study evaluated the measurement invariance of a 7-item scale designed to assess elder neglect across three groups of informants: Older adults, family members, and home care workers. In addition, differential item functioning (DIF) was evaluated in order to examine whether individuals of certain characteristics have a different probability of endorsing certain items even at equivalent levels of the overall construct of neglect. METHOD: A cross-sectional sample of 686 family members, 388 older adults, and 523 home care workers was drawn. A series of sequentially nested confirmatory factor models was examined to identify whether configural (the same items are associated with the same factor across groups), metric (factors have a similar meaning across groups), and scalar (group means can be meaningfully compared across groups) invariances can be established. Multi indicators multi causes analysis was conducted to identify DIF across age, gender, and education. RESULTS: Five items were adequate indicators of the overall construct. The findings provide support to configural, metric, and scalar invariances across the 3 groups of informants. None of the items resulted in DIF. DISCUSSION: The findings advocate for the use of the 5-item neglect scale across different groups of informants and call for the evaluation of elder neglect within the constellation of the caregiving unit.
OBJECTIVES: The present study evaluated the measurement invariance of a 7-item scale designed to assess elder neglect across three groups of informants: Older adults, family members, and home care workers. In addition, differential item functioning (DIF) was evaluated in order to examine whether individuals of certain characteristics have a different probability of endorsing certain items even at equivalent levels of the overall construct of neglect. METHOD: A cross-sectional sample of 686 family members, 388 older adults, and 523 home care workers was drawn. A series of sequentially nested confirmatory factor models was examined to identify whether configural (the same items are associated with the same factor across groups), metric (factors have a similar meaning across groups), and scalar (group means can be meaningfully compared across groups) invariances can be established. Multi indicators multi causes analysis was conducted to identify DIF across age, gender, and education. RESULTS: Five items were adequate indicators of the overall construct. The findings provide support to configural, metric, and scalar invariances across the 3 groups of informants. None of the items resulted in DIF. DISCUSSION: The findings advocate for the use of the 5-item neglect scale across different groups of informants and call for the evaluation of elder neglect within the constellation of the caregiving unit.