If they Google You, Do you Win?

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!

Up to +0.9 Correlation between Twitterwonk Scores and National Polls

455906_2bf3212686734defb5f3b366a29c3c86 I have found, on average, a +0.6 correlation between Twitterwonk Scores and national polls. Of the four leading candidates, Bernie Sanders (displayed) shows the highest correlation, +0.9. The national poll data used in the analysis is from fivethirtyeight. The findings suggest Twitterwonk can be used as a substitute to traditional national polling methods, which are resource intensive and delayed in reporting. While national polls are not always predictive of state primary results, they play a big role in media coverage. 

Republican Debate Nov 10, 2015: Google Search Correlation Matrix

Screen Shot 2015-11-10 at 1.37.37 PM (2)Where does the competition stand leading to tonight’s Republican Debate? Using Google Trends data from the past three months, I plotted a correlation matrix comparing search queries among the candidates.


The matrix reveals trends. Notably, Donald Trump searches are the least related to the other candidates. This supports Trump having created his own news cycle, apart from the Republican news cycle. As you may see in the time series plot, the Trump search artifact exhibits peaks and nuance not seen in the other artifacts. Furthermore, Trump searches are the least related to Ben Carson searches, with a 0.01 correlation. Other patterns to note, Ted Cruz and Marco Rubio searches are highly related, 0.94 correlation, and there is high inter-relatedness among searches for Jeb Bush, Carly Fiorina, and Rand Paul. Screen Shot 2015-11-10 at 12.55.19 PM (2)

Screen Shot 2015-11-10 at 12.56.07 PM (2)

It becomes clear that, aside from Trump and Carson, Google Searches for Republican Candidates hit inflection points around the debates. Furthermore, there is clustering among candidates, which can be considered as different news cycles:

  1. Donald Trump
  2. Ben Carson
  3. Ted Cruz, Marco Rubio
  4. Jeb Bush, Carly Fiorina, Rand Paul

We will see tonight if any of the Candidates makes a move to shake up the trends.