Access Type

Open Access Dissertation

Date of Award

January 2014

Degree Type


Degree Name




First Advisor

Janet R. Hankin


The purpose of this study was to assess the relationship between measures of public and private religiosity and various measures of health behavior. Previous studies have shown that religious involvement is associated with health outcomes (Koenig & Vaillant, 2009; Chatters, 2000; Meisenhelder, 2003). However, many studies use religious attendance as the single measure of religiosity This research used multiple indicators of religiosity that reflected both public and private dimensions in order to predict various health behaviors, including preventive health behaviors, abstention from alcohol and smoking, and physical activity. In addition to taking a multi-dimensional approach to assessing the religion-health relationship this study described the effects of religiosity on health behaviors for racial/ethnic and gender groups.

Data were obtained from the 2004 Survey of Texas Adults (Musick, 2005). Logistic regression and OLS regression models were used to estimate the effects of 6 independent measures of religiosity on 8 independent measures of health behavior. Religiosity measures included two public dimensions: service attendance and religious participation; and 4 private dimensions: prayer, meditation, and religious scripture reading. Health behavior measures included: preventive care measures (physical exam, dental care, and vitamin use), abstention measures: (alcohol and smoking), and exercise measures: (walking, moderate exercise and strenuous exercise). Interactions describing the effect of religiosity on health behaviors for gender and racial/ethnic groups were also assessed.

Results of the analyses showed that public measures of religiosity, attendance and participation, were consistent predictors of positive health behaviors in logistic and OLS models. However, interaction models found that both public and private forms of religiosity have inconsistent effects on health behaviors for gender and racial/ethnic groups. In some types of health behaviors among Blacks, Hispanics, and males, religiosity had a negative effect on heath behavior.

The results of this study show that researchers cannot take a unidimensional approach in studying the effects of religiosity. While both public and private religious activity had beneficial effects on some measures of health behavior, these findings were inconsistent across gender and racial groups.

Included in

Sociology Commons