Home > Research > Publications & Outputs > Measurement models for understanding the social...

Electronic data

  • 2018PattinsonPhD

    Final published version, 3.96 MB, PDF document

    Available under license: CC BY-ND: Creative Commons Attribution-NoDerivatives 4.0 International License

Text available via DOI:

View graph of relations

Measurement models for understanding the social challenges of caring for the elderly in Brazil and England

Research output: ThesisDoctoral Thesis

Published
Publication date2018
Number of pages301
QualificationPhD
Awarding Institution
Supervisors/Advisors
  • Titman, Andrew, Supervisor
  • Lancaster, Gillian, Supervisor
  • Moreira Dos Santos, Dirley, Supervisor, External person
  • de Moraes, José Rodrigo, Supervisor, External person
Publisher
  • Lancaster University
<mark>Original language</mark>English

Abstract

An ageing population is a challenge faced by many developed and developing countries. By understanding the needs in health and well-being of the elderly governments can ensure policies and services are as efficient as possible, in order to provide the best care amidst growing demand. The thesis analyses categorical data from the English Longitudinal Study of Ageing (ELSA) and Brazil’s National Household Sample Survey (PNAD) to form measurement models for aspects of health, economic well-being and subjective well-being in the elderly, in England and Brazil.
Applying structural equation modelling (SEM), a measurement model for health and economic well-being in both countries was constructed using categorical variables from the survey data, as well as for subjective well-being in England. For health the variables represented self-reported morbidity and functional status, while for economic well-being variables represented housing quality and durables owned within the household. Psychometric measures of satisfaction with life, quality of life, loneliness and depression constituted subjective well-being. Based on empirical evidence from exploratory factor analysis (EFA), the latent structure of each aspect was defined. Multilevel structural equation modelling was applied to the PNAD to capture the hierarchical structure, whereby individuals were clustered by household then the households were clustered by sector and the sectors clustered by municipality. Meanwhile, the longitudinal dynamic of the ELSA allowed for a multivariate latent growth modelling, using multiple indicators to model the trajectories of subjective well-being in multiple aspects.
Health and economic well-being were positively associated in both countries. Metabolic health was a factor that was identified in both England and Brazil, while musculoskeletal health was identified only in Brazil. In England, economic well-being had two separate factors relating to the inside and outside of the household. Health and household quality significantly influenced subjective well-being, but not neighbourhood quality, with better health and economic wellbeing associated with greater subjective well-being. In the multilevel SEM a different structure of latent constructs was identified for the health of the sectors of elderly individuals, and the variables had different priorities.
In both countries, health was significantly better for men, but worse for those of non-white race and with older age. Regional disparities were also present in health and economic well-being. In Brazil, economic well-being was worse with older age, in rural areas, male gender and non-white race. In England, economic well-being was better for older age with no difference between the genders and races.
Subjective well-being was persistently better for men and those that were married, while being persistently worse for those of non-white race or divorced/separated marital status. Those that were widowed recovered from initially lower subjective well-being. Marital status was highly influential to the subjective well-being of the elderly with significant impact from becoming divorced or widowed and benefits from getting remarried.