Religion

In contrast to the many national probability-sample studies of patterns of religious participation, research on the determinants or predictors of religiousness in older adults has drawn mostly on small, nonrandom samples of patients, community-center attendees, church members, or students. Since the advent of research funding by the NIA in the 1990s, this has begun to change. Reliable national findings pointing to differences in religious participation by age, gender, race or ethnicity, social class, and other sociodemographic variables are starting to accumulate.

Taylor and Chatters have presented quite a bit of evidence for significant sociodemographic differences in religious participation among older adults, especially older African-Americans. Older age, more education, greater income, being married, female gender, and living in the southern United States each has been found in multiple studies to predict greater levels of organizational, nonorganizational, and subjective religiousness. These important findings firmly contradict commonly held assumptions that religious people, especially religious older people and older African-Americans, tend to be poorer and less educated.

In one NIA-funded study, Levin, Taylor, and Chatters analyzed data from four separate national probability-sample surveys of older adults conducted from the early 1970s to the late 1980s. Collectively these surveys examined twenty religious indicators of all three types (organizational, nonorganizational, and subjective) of religiousness in a total of over six thousand respondents. Significant racial differences were found for sixteen of these variables; significant gender differences were found for twelve variables. In every instance greater levels of religiousness were found among African-Americans and females. Gerontological research among older Hispanics, Jews, and Asian-Americans has focused less on religion, but sociodemographic correlates of religious participation have been identified in these groups.