Election Bias: Comparing Polls and Twitter in the 2016 U.S. Election

TL;DR

This study compares traditional polls and Twitter sentiment analysis to evaluate their accuracy in predicting the 2016 U.S. presidential election outcome, highlighting the potential of social media as an alternative prediction tool.

Contribution

It introduces a comparative analysis of polls and Twitter sentiment data for election prediction, emphasizing the need for improved methods in forecasting electoral results.

Findings

Polls were less accurate and biased in 2016

Twitter sentiment showed potential as an alternative predictor

The study suggests combining methods for better accuracy

Abstract

While the polls have been the most trusted source for election predictions for decades, in the recent presidential election they were called inaccurate and biased. How inaccurate were the polls in this election and can social media beat the polls as an accurate election predictor? Polls from several news outlets and sentiment analysis on Twitter data were used, in conjunction with the results of the election, to answer this question and outline further research on the best method for predicting the outcome of future elections.