HIV continues to disproportionately plague select communities across the United States. Individual sexual behavior serves as a dominant explanation for differences in HIV infection. In public health, little attention has been placed on community context and features of the community environment as important predictors of sexual health. The community environment and resources or lack of resources may influence sexual behavior. This may contribute to disparate rates of HIV transmission. I examine how socioenvironmental factors and sex ratios influence sexual behavior and HIV rates, and contribute to race/ethnic differences in risky sexual behavior. I used multilevel models and data from the Behavioral Risks Factor Surveillance System (BRFSS) to determine the cross-sectional association across 15 states between county-level factors and risky sexual behavior for different racial/ethnic groups. The measure of sexual behavior examined was the number of sexual partners. The county level factors examined included imbalanced sex ratio, residential segregation, and the percent of residents below poverty. Results: Increased residential segregation was associated with higher odds of risky sexual behavior in all groups although a clear dose response trend was only observed in Whites. The association between sexual behavior and county sex ratios partly followed the pattern predicted by the Alternative Sex Ratio Mate Preference Shifts Hypothesis. Racial/ethnic differences in risky sexual behavior were reduced after adjustment for marital status and age. I did not find a substantial effect of adjustment for racial residential segregation, percent below poverty, or the sex ratio at the county level. Conclusion This is one of the first studies to examine the hierarchical association of county-level variables with risky sexual behavior. Future multilevel work with different measures of sexual behavior and alternative contextual measures is needed to better understand the social processes affecting HIV risk and the factors contributing to persistent race/ethnic differences.Ph.D.Epidemiological ScienceUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/63624/1/zgant_1.pdf