Graduation Year


Date of Submission


Document Type

Campus Only Senior Thesis

Degree Name

Bachelor of Arts



Reader 1

Professor Mike Izbicki

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.


The traditional methods of polling are an expensive and time-consuming process. The amount of resources expended in performing these polls, paired with their recent unreliability and the vast amount of data available on Twitter, has led many researchers to turn to natural language processing to gauge a candidate’s popularity in the population. During Donald Trump’s presidency, Twitter became a hotspot for debate and political discourse as it was Trump’s primary way of communicating to the public. With the amount of data available regarding people’s political beliefs on Twitter, the sample size easily exceeds the size of any survey that can be performed and in a fraction of the time. In this work, over six million tweets were analyzed with sentiment analysis and other statistical methods. The results display the effectiveness of Twitter data’s ability to predict election results and explain the results that were seen in the past election.

This thesis is restricted to the Claremont Colleges current faculty, students, and staff.