Religion, health, and psychological well-being in older adults: findings from three national surveys.

This study examines the impact of religious involvement on health status and psychological well-being using data on older adults from three national probability surveys: the Myth and Reality of Aging (N = 2,797), the Quality of American Life (N = 1,209), and Americans' Changing Lives (N = 1,669) studies. Constructs are measured by single items and indices that vary across data sets. A proposed theoretical model specifies direct effects of religiosity on health and well-being and indirect effects on well-being through health. Analyses consist of structural-equation modeling of confirmed measurement models using weighted least squares estimation in LISREL 8.03. The model is analyzed first as specified and is then rerun controlling for the effects of six exogenous constructs: age, gender, race, marital status, education, and geographical region. Findings reveal excellent overall fit in all three samples and the presence of statistically significant religious effects, notably positive net effects of organizational religiosity, in all three samples. These results build on those of prior studies based mostly on samples limited regionally or methodologically or to particular racial or ethnic groups. This study also underscores the value of replicated secondary data analysis as a strategy for gerontologists seeking to confirm or examine a given structural model. Finally, an agenda is proposed for future research in this area.