Date Presented

Winter 12-1-2006

Document Type

Thesis

Access Type

1

Degree Name

Bachelor of Arts

Department

Business Administration

First Advisor

Joe Turek

Second Advisor

Dan Messerschmidt

Third Advisor

Joe Prinzinger

Abstract

Using Ordinary Least Squares (OLS) regression analysis, this study attempts to capture variation in dropout rates across Virginia counties and cities. With the respective dropout rates as the dependent variable, seven independent variables are used accordingly in order to provide as much explanatory power as possible. At the 10 percent significance level, four of seven variables are statistically significant with an adjusted R2 of .374. Important policy implications can be derived from the model and its statistically significant variables. The model finds that the percentage of blacks in the population, university access, the unemployment rate and single female-headed households to be statistically significant with coefficients that have a relatively large impact on dropout rates. Median household income, percentage of the population with advanced degrees, and student to teacher ratios were found to be insignificant. Using these regression results, local government can more effectively move funds to areas that will help to decrease dropout rates. Investigating into the black population and their increased propensity to drop out as well as focusing on mentoring programs to help relieve extra stress and decreased parental supervision found in single female-headed households will provide the most effective decrease in dropout rates the model.