Today, in the United States, the jury selection process consistently yields juries that fail to be sufficiently diverse to summarize the populations they represent. Research has focused on analyzing the portion of the selection process that takes place in court, however this is problem that permeates throughout selection processes that occur before the court date as well. Throughout my analysis I concentrated on the causes for the disparity in juror representation in jurors called to court and those actually empaneled, to study if the political affiliation of a county had a relationship to the proportion of jurors selected from each demographic group. Approaching the problem using fractional multinomial regression of both the selected jurors and all jurors called to appear, I was able to find some evidence for a negative relationship between the proportions of registered Democrats in a county and the proportions of white jurors, and a positive relationship between the proportions of registered Democrats and jurors of color.