The Relationship of Lottery Outlet Density to Income and Race/Ethnicity in Essex County, New Jersey

Background and Purpose: Excessive participation in lottery play is prevalent among low income and otherwise disadvantaged populations served by social workers (citations omitted). National studies in the U.S. and the U.K have reported that lottery expenditures are predictive of problem gambling, particularly in low-income households. In the U.S., 42 states and the District of Columbia and Virgin Islands all sponsor lotteries. In response to recent budget crises, several states have recently proposed expanding the number of outlets to “big box” department stores (i.e. Walmart), which are prevalent in low-income neighborhoods. The purpose of this study is to utilize geographic information systems (GIS) to analyze the relationship of socio-economic and other demographic variables to lottery outlet density in Essex County, NJ, the largest and most diverse county in the state. Specifically, the study aims to identify differences in outlet density by median household income and race/ethnicity.

Methods: The year 2000 TIGER/Line files were used to map year 2000 census data for the residential census tracts in Essex County. Addresses of all lottery-selling retail outlets in Essex County, New Jersey in 2008, obtained from the New Jersey Department of the Treasury, were then matched with ArcGIS,® resulting in successful geographical coding of more than 90% of the lottery outlets. The number of lottery outlets per 10 kilometers of roadway was used as the primary density measure, consistent with other research in other areas (e.g., tobacco control) investigating outlet density issues. Median household income and the percent of African American residents and Latino residents were based on year 2000 census data.

Results: ANOVA was performed to examine differences in lottery outlet density across quartile categories of each demographic predictor. Initial results indicated statistically significant effects for median household income (F = 5.46; p ≤ 0.01), percent African American (F = 4.62; p≤0.05), and percent Latino (F = 4.89; p ≤ 0.05). Outlet densities ranged from a high of 2.8 outlets to a low of 0.3 outlets per 10 km/roadway. Compared to tracts with higher median household income, tracts in the lower income quartiles had significantly more lottery outlets per 10 km./roadway (0.5 and 1.9 outlets, respectively). Relative differences among the quartiles were similar for the race and ethnicity variables. Census tracts with higher proportions of African American and/or Latino residents had significantly more lottery outlets per 10 km/roadway compared to tracts with lower proportions of African American or Latino residents.

Conclusions and Implications: Findings suggest that lottery outlet density was significantly related to demographics for the county. Census tracts with greater density of lottery outlets had lower median household incomes and higher percentages of African American or Hispanic residents. These findings have important implications for the development of land-use policies and prevention strategies to reduce the burden of problem gambling and limit future expansions in geographic areas with excessive lottery expenditure. Social workers in these areas should screen for excessive lottery play, obtain education on identifying gambling-related problems, and refer those clients to specialized treatment services.