More Than Just Kissing Babies: The Strategy of Campaign Visits in United States Presidential Elections
Every four years, the Democratic and Republican parties each send out a nominee to tour the country, press the flesh and convince the American people to put them into the White House. In my Master thesis, I analyse how candidates decide where to hold their campaign events. To this end, I employ a poisson regression on data from the 2004, 2008 and 2012 presidential elections. I find that the competitiveness of a state as well as its number of votes in the electoral college serve as an excellent predictor of campaign visits. Beneath the state level, candidates from both parties focus their efforts on cities and regions which support them politically.
I am quite happy with my Master thesis, written during the spring semester 2013, under the supervision of Prof. Thomas Gschwendt, at the University of Mannheim. However, before submitting it for publication, I intend to improve it further by replacing the poisson estimator with a negative binomial model to account for the zero-inflation inherent to this data. Furthermore, I want to incorporate temporal variations in campaign tactics into my main model, as they are currently relegated to a separate part of the paper. This is what I am currently working on.