Bernie Sanders and Donald Trump finally had their day.
After maintaining a sizeable advantage in the New Hampshire polls since January, both candidates are showing that they are forces to be reckoned with in the 2016 presidential election. Sanders enjoyed a 21-point lead over Hillary Clinton with 60% of the Democratic vote, while Trump blew the Republican field away with 35.1% of the vote. John Kasich was the Republican runner-up, with 15.9% of the vote.
As an experiment, I collected tweets that mentioned the candidates in real time starting at 9am PST, and used machine learning models to identify the sentiment (e.g. positive or negative) expressed in the tweet. I then plotted the average hourly sentiment for each candidate over a 12-hour period.
What are the results?
- Clinton and Rubio garnered support early in the primary, but petered out as rumors of a Sanders and Trump victory surfaced in the mainstream media.
- Bernie Sanders is enjoying a groundswell of support on Twitter, especially after the victory. Not only did Sanders maintain a large lead over Clinton in the primary — he also maintained a large lead in sentiment on Twitter.
- Donald Trump remains a polarizing figure in this election, yet sentiment toward him balances out at the higher end of the spectrum.
- Nobody likes Jeb Bush.
Interestingly, despite Donald Trump winning in the Republican field, folks prefer John Kasich on Twitter. In fact, they love him — almost as much as Sanders.
Why does sentiment matter?
Candidates like Sanders have been painted as black sheep in the mainstream media despite a massive following on social hubs like Twitter and Reddit. This is due in large part to the echo chamber effect — the perception that a small online minority is controlling the narrative for Sanders, and that these supporters are not representative of the electorate. Therefore, the fundamental question being asked by pundits has been: “Can Bernie win in 2016?“
Establishing a correlation between support in online communities and success in the primaries is key to developing predictive models of the 2016 election. This preliminary analysis provides a first step into establishing such a correlation.
Stay tuned for coverage of the Nevada and South Carolina primaries on February 20th, with a more in-depth analysis of the underlying content of the tweets.
Photo: Gage Skidmore/Flickr