Venue: The Fuqua School of Business, Duke University, 1 Towerview Drive, Durham, NC 27708-0120
Presentation
Does the Promotion of Community-Level Social Capital Reduce Obesity Risk?
Obesity has increased dramatically throughout the past several decades. Numerous studies have linked individual-level behavioral factors and familial characteristics to obesity. Recently, studies have begun to examine the influence of macro-level factors, suggesting a link between an increased risk of obesity and structural community characteristics such as urban sprawl and income inequality. Yet, a link between community-level social capital (CSC) and obesity has not been determined. This study explores whether community-level (structural) social capital is associated with obesity. CSC may affect information flows among individuals as to the risks of obesity as well as weight control techniques which may result in behavioral changes which reduce the likelihood of a person becoming obese. We also examine whether the relationship between CSC and obesity risk differs by levels of education as education may moderate the value of information to a given person. We posit that CSC has a larger impact for more educated individuals.
460,478 Individual-level observations were drawn from the Behavioral Risk Factor Surveillance System (BRFSS) for the years 2001, 2003, and 2005, which provides data on the behaviors and health status of the general population in all U.S. states and the District of Columbia. The BRFSS is a statewide telephone survey of civilian, non-institutionalized adults aged 18 and older, using a complex multi-stage cluster sample design. The individual-level BRFSS data were linked to our measure of CSC, the Petris Social Capital Index (PSCI). The PSCI is designed to capture the structural aspect of CSC. The county-level PSCI is derived from employment data on voluntary membership organizations in the County Business Patterns. We restrict our analyses to counties with at least 100,000 residents to ensure adequate sample sizes in each county. We estimate multivariate logit models first for the whole sample and then separately by levels of education (college, some college, high school education, and less than high school education). All models account for the complex survey design. Our covariates include individual demographic and socioeconomic factors (such as sex, age, race/ethnicity, marital status, income, and education), and county and year fixed effects.
We find a significant, negative association between the county-level PSCI and the likelihood of being obese. As predicted, the negative relationship between the PSCI and the likelihood of being obese becomes larger as education levels increase. However, a statistically significant relationship is found only for college graduates.
Our findings suggest that CSC reduces the risk of being obese, but only for those with more educated persons. Thus, promoting CSC may not be an effective tool for reducing the obesity risk for less educated individuals.