Methods, not Madness: Forecasting U.S. Presidential Elections

Tony, Bryan

Publication

Methods, not Madness: Forecasting U.S. Presidential Elections

Tony, Bryan

Date

2016

Additional date(s)

2016-05-19

Abstract

Through my thesis, I discuss the current literature on numerous factors that could affect how voters approach presidential elections including the economy, midterm election outcomes, public perceptions of the candidate, foreign entanglements, and, most interestingly, the personalities of the candidates' spouses. I then lay out my hypotheses as to what and how different variables impact an election outcome in my own theory development. Next, I operationalize these factors to set the parameters and expectations for my study. In the results section, I assess how successful my model was in predicting the election outcome by comparing the results to the past election outcomes and looking for trends. Finally, the results indicate how effective or ineffective these variables are before surmising a conclusive answer to the factors that most affect election results. The concluding discussion further dissects the results of what to expect in 2016 before closing with meaningful implications, limitations, and areas for future research.