Examining Presidential Candidates Via Google Trends

Michael Giuffrida writes "Google Trends is a free application produced by Google that shows how often a given keyword is searched for, over time. After seeing how candidates in the 2008 primaries have done in Google Trends in different states, it's clear that this tool can be very useful for campaigns." Read on below for some of the specifics about how these candidates have fared, Google-wise. "For example, in New Hampshire, in the days leading up to the Jan. 8 primary, Clinton was searched for the most, followed by Obama, followed by Edwards — which was how the primary results turned out. In other words, the candidates most searched for on Google by users in New Hampshire were also the candidates with the most number of votes. This works for many other states as well.

For the first 37 Democratic primaries and caucuses, 32 states had enough data on Google to make a prediction. This method correctly predicted 27 of those 32 sates. Predictions aside, the tool is also useful simply in finding out how popular a candidate is in different states, assuming that the more popular candidates are entered more often as a search term in Google (an assumption that was verified, at least for the Democratic primaries, by the positive association found)."

Comments

Cherrypicking the data to reach false conclusions

[MSTCrow5429]

"For the Republican Primaries, last names could easily be used. Ron Paul was excluded. His last name is too common. Using his full name is not a good solution either, because he had massive popularity on the Internet, becoming a meme of sorts, which did not at all correspond with his actual successes (or lack thereof) in the primaries."

So, in other words, any data that contradicts the hypothesis will be thrown out.

I Know It's Not Statistically Valid.

[mgiuffrida]

Hey, guys, You're all right. There are serious statistical flaws here. I chose the keywords "hillary" and "barack" and "edwards" because that was the model that showed the best success. If one were actually predicting this live, he or she would soon see that "hillary" and "barack" do produce better results and would switch to those terms, simply because it fits the model better. Every election will have its outliers (Obama, Ron Paul). But hopefully this brief overview will provoke some campaign analysts to actually try to interpret Google Trends results for themselves, using more advanced models.

Re:In case anyone looks at the pretty graphs...

[amccaf1]

"For the Republican Primaries, last names could easily be used. Ron Paul was excluded. His last name is too common. Using his full name is not a good solution either, because he had massive popularity on the Internet, becoming a meme of sorts, which did not at all correspond with his actual successes (or lack thereof) in the primaries."

Translation: Including Ron Paul would have indicated our hypothesis was incorrect, so we excluded him.