Abstract Search

ISEF | Projects Database | Finalist Abstract

Back to Search Results | Print PDF

Using R to Analyze Election Results and Accurately Predict the Results in Correlation with Social Media Use and Presence

Booth Id:



Finalist Names:
Crary, Casey (School: Greencastle High School)

In recent years social media has evolved from something that helps you keep in touch with relatives or friends to a driving influencer of all parts of life, politics included. Since its first inception, social media has been used to advertise social media campaigns, but in the past decade it has become a place for politicians to speak to the followers much like a rally, although everyone can now hear what they are saying, as a result, social media has obviously become a major factor in who takes an election. This study focused mainly on the effects that Twitter posts had on election results in the 2016 Senate Race. Using a data miner application and websites that archive photos of Twitter accounts from 2016, the number of followers and tweets were recorded. After comparing the percentage of votes with social media presence, it was shown that tweeting had little to no effect on an election. However a positive correlation could be seen between the number of followers and chances at winning, but that could be a result of the incumbent often winning.