Author. Developer. Investor. Musician.
Finished reviewing the manuscript for my ‘Can Twitter predict the outcome of the U.S. Presidential Election?’ case study, which will be included in Dr. Kristen Sosulski’s forthcoming book Insights into Becoming Visual, out on Routledge Press this Spring 2018. Feeling the author vibe, and stoked! 💪💪💪
4 Days before the 2016 US Presidential Election, what does the Twitterwonk Model show?
Here are monthly, weekly, and daily views starting June 2015 through October 2016.
The model, which is a function of Twitter activity, shows Trump in the lead.
While the model correctly predicted the primaries, jury is still out whether the same dynamic applies to the general election. Regardless, 2016 will be remembered as an election in which social media, on both the left and the right, played a dominant role. I believe social media will play an even greater role in future elections, shaping political communication for decades to come, as well as becoming a primary source for the measurement of public opinion. In the meantime, let’s see how this election plays out in the remaining days…I predict many will be surprised at how close the results turn out to be.
In a way, this election is a referendum on “do actions speak louder than words”, is what people do in the privacy of their internet browsing more reflective of their future behavior than what they tell pollsters? And while I have focused on twitter as a barometer of public opinion, there are other data sources that could signal the private thoughts and future actions of voters. The linked NYT article, “If they Google you, Do you Win?”, mentions using the Google queries “Trump Clinton” vs. “Clinton Trump” as signals of voter interest, with the respective queries reflecting bias towards the candidate listed first, i.e. “Clinton Trump” would reflect bias towards Clinton. Using this methodology, I researched Google trends for Battleground states to see where public opinion may be. The data are displayed below.
For the month of October 2016, “Trump Clinton” leads “Clinton Trump” in every state with the exception of Nevada.
You might say Trump is a polarizing celebrity, and for that reason he may be top of mind even if the individual plans to vote for Clinton. Okay, well then let’s penalize Trump 10%. Even in that case, ‘Factored “Trump Clinton”‘ indicates that, with the exception of Nevada, the three states that are in play are Virginia, Iowa, and Florida.
So while it is unclear in which direction the election will result, I believe we may be surprised at how close the results turn out to be, and that one thing we may remember is the discrepancy between what was reported in the polls leading up to the election and what actually happened online. We only have 4 days left to see which source provides a clearer signal of truth, and until then….Good luck to both candidates!
https://clyp.it/u13dl5s3/widget
“Many people will be shocked at how close this election will be.”
Over the past months I have developed a process for understanding elections through the lens of Twitter data. Combining and comparing Twitter metrics, the process provides an ordinal representation of where the candidates stand amongst one another. I am excited to announce the process will be unveiled over the coming weeks as a website, named Twitterwonk. It is my hope Twitterwonk will serve as a tool for better understanding elections as we further immerse ourselves in the age of new media. To kick us off, here is a visualization displaying the Twitterwonk Score for the Republican and Democrat frontrunners. Good luck to all the candidates, and may Twitterwonk serve you.
As of January 27, 2016, Bernie Sanders leads Hillary Clinton in the Iowa and New Hampshire polls (Real Clear Politics). Here we present how Trump hit, Hillary fired back and left the goal open for The Bern.
Displaying a composite score, calculated with Twitter data in a process outlined in my previous post, the visualization presents the decline that allowed Bernie Sanders to rise above Hillary Clinton.
It started on December 8, 2015, when Donald Trump released his statement on preventing muslim immigration. Over the next two weeks, Hillary Clinton would respond to Trump in a fury of statements on foreign policy, ultimately leading to her decline. And while Bernie Sanders also responded to Trump, he did not waver from domestic policy, which better resonates with the compassionate tone of both Democrat candidates. Christmas brought a change in tone for Hillary, but by this point it was too late: Sanders had already gained the lead. Here are the kind of statements on domestic policy that reversed Hillary’s decline.
“There’s no reason for teachers and nurses to ever pay higher tax rates than top CEOS.”
“It shouldn’t be so hard to be a working parent.”
“We need to make it easier for women to get ahead at work while still being there for their families.”
NYMAG just published a profile on Trump supporters and the appeal to “Testicular Fortitude”, i.e. strength. While many pages could be spent on the topic, it rings clear that the Commander-in-Chief must be compassionate at home and ruthless abroad. A word no short of “destroy” is expected in reference to America’s enemies (used by President Obama, Trump, and Sanders), especially during times of uncertainty. Hillary’s gift for empathizing with the other has made her a skillful Secretary of State, and it may have cost her The White House.
Donald Trump: 6/16/15 – 1/16/16
Most Mentioned Moments on Twitter
Celebrity businessman Donald Trump has single handedly disrupted the right-wing political playbook with a winners and losers worldview delivered through a charismatic, entertaining social media feed, shaping what remains of mainstream media. Find below his Top 16 Tweets since 6/16, when Trump announced he would run for President.
Guns, Fences, ISIS, God, Christmas, Obama, Hillary, and George W — take note America!
Who will win the 2016 US Republican and Democrat Presidential Nominations?
For the time period 11/1/2015 – 11/30/2015, my team of researchers at NYU Stern collected Twitter data on the following candidates:
We designed a prediction algorithm using these five metrics:
The heat map represents the ordinal ranking produced by the algorithm.
Grouping the candidates by party, and comparing the results to actual poll rankings we see the algorithm matches the ordinal representation of the Democrat Candidates and has Ted Cruz as second among the Republicans, behind Donald Trump. 
While the Ted Cruz point is not consistent with current polls, we believe Twitter to be a directional leading indicator and predict Ted Cruz to break away from the other Republican candidates in the coming weeks. This visualization, based on the algorithm, shows Ted Cruz gradually pulling away from Ben Carson and Marco Rubio.
Interestingly, we find a relationship between Marco Rubio and Hillary Clinton, such that when the difference between their scores decreases, Rubio surpasses Clinton in direct comparison polls.
Taken together, we predict the 2016 Democrat Presidential Nominee to be:
We predict the 2016 Republican Presidential Nominee to be:
If not,
Based on his ranking according to the algorithm in addition to his standing in direct comparison polls with Clinton, we believe the 2016 Republican Vice Presidential Nominee to be:
For more details about the study, please email ash487@stern.nyu.edu
Special thanks to Troy Manos, Tarang Dawer, and Keita Shimizu.