THE EFFECTS OF PRISON GERRYMANDERING ON ELECTION RESULTS

Abstract

This study looks at whether or not there is a correlation between the voting patterns of a county and the presence of prisons within that county. There is a history of prisons being used to inflate population numbers within states to change where voting power lies. Previous research exists indicating a speculative relationship between Republican politicians and increased prison gerrymandering, so this research was developed to reveal if there is a genuine relationship between these two factors. Some of the variables included to better isolate the influence of prison location are per county income, biographical information, total population, voter turnout, if the winner was an incumbent, along with presence and sizes of prisons within counties. Implementing regression models like multinomial logit and logit, I was able to encapsulate all my variables and their effects on voting patterns to measure the different relationships between them. I found that there was a significant relationship between US Representative results and prison placement, and a null to negligible relationship in statewide elections.