Throughout the 2020 presidential election, then-President Donald Trump continued his attacks on Vote-By-Mail (VBM). Although the extent of his influence remains unclear, only 30% of Republicans voted by mail despite nationwide polling places closures and the health risk of casting an in-person ballot during the COVID-19 pandemic. In comparison, nearly 60% of Democrats cast a mail-in ballot. Survey studies had found that Republicans were less likely to VBM than Democrats, but they did not consider other factors that might have influenced their preference of voting method.

In order to test whether VBM is indeed a partisan issue and less accepted by Republicans, I used a multiple logistic regression model to estimate the gap in the probability of VBM between Republicans and Democrats, controlling for other factors that influence one’s preference for VBM. I controlled for voter age, race, sex, and distance to a polling place. The data preparation processes involved joining various election datasets – voter registration, voter history, and polling places – geocoding residential and polling location addresses for each voter, and calculating their expected travel distance.

My results showed that: (1) being Republican decreases the probability of VBM, controlling for voter age, race, sex, and distance to a polling place; and (2) that being White has a differential impact on the probability of VBM across the political spectrum. Being White increases the probability to VBM among Democrats, and decreases among Republicans. After applying a defensible missing data imputation technique, this paper will be submitted as part of my PhD application.