Master of Arts (MA)
Using a strategic politician model, I argue that we should be able to predict when legislators are going to switch parties by weighing the costs and benefits of party switching under different circumstances. Using variables that measure electoral risk and opportunities for advancement to higher office, an event history model is used to predict when individual legislators will switch. Although this is a rare occurrence, and electoral risk seems to play little in the decision to switch parties, we do find that opportunities for advancement and the relative power of the two parties (measured by offices held) has a noticeable effect on the phenomenon.
Document Availability at the Time of Submission
Release the entire work immediately for access worldwide.
Rickards, Dylan Scott, "Predicting party switching in U.S. state legislatures" (2004). LSU Master's Theses. 3553.