Google Trends and the Last Minute Voter

Guest post on joeam.com:

http://joeam.com/2015/12/01/google-trends-and-the-last-minute-voter/

 

What is Google Trends?

From wikipedia: “Google Trends is a public web facility of Google Inc., based on Google Search, that shows how often a particular search-term is entered relative to the total search-volume across various regions of the world, and in various languages. The horizontal axis of the main graph represents time (starting from 2004), and the vertical is how often a term is searched for relative to the total number of searches, globally.”

Why is this interesting?

A survey is scientific but it takes time to do and it takes time to process. It costs a lot and, without releasing all the questions and the data, it is most valuable to the politician who has all the variables. It misleads the common person who is ill equipped to question and dissect the data.

Google Trends is free and it is a public facility so anyone can verify the results. It is almost real time which allows crafty people to use it as a leading indicator.

2004 Elections

First let’s take a look at the weeks prior to the 2004 elections through the lens of Google Trends:

GMAvsFPJ

What this shows is that, except for 1 week, FPJ had more searches than GMA until the first week of April and the last week of April when they were tied.  FPJ was ahead until May 2-8 with GMA overtaking FPJ for searches dated May 9-15.

We have to look at the internet population during 2004. The penetration then of the internet were mostly through dial-up and internet shops [Source: “Internet Live Stats“].

Capture01

 

We only had 4.5 Million internet users then but, if we remove non voters, I suspect this would be close to 2-3 Million voters. What we can see is that the Internet-Enabled Filipinos were really torn between the two candidates.

2010 Elections: Pnoy vs Erap vs Villar

Fast forward 6 years:

VillarVsPnoyVsErap

Some observations:

The searches escalated with three distinct parts: October to December | January to April | April to May

October started with Pnoy and Erap neck and neck, and with Manny Villar a distant 3rd. Manny Villar’s campaign was unique because it intelligently used the campaign components of Party, Senatorial Slate and Partylist to promote the Presidential Candidate of the Party. However, Pnoy, riding in the glow of the Aquino name, coasted to high google searches while, curiously, Villar’s and Erap’s search volumes mirrored each other up to the start of January.

From January to April, Manny Villar became the front runner in terms of google searches. I attribute this to two things: (1) Villar’s snappy campaign song was such an ear worm that it drove people to search for him online, and (2) Villar had a decent website and good fluff articles about his rise from poverty. When you see someone with such a compelling narrative, you want to know more about him. This drives more google searches.

January to April was not all roses for Villar because Google Trends are not always positive. The “truth campaign”, or in real terms, the demolition squads of the other campaigns, went full blast. This may also have driven the google searches up.

What we do know from the data is that, by the end of April, Erap and Pnoy were rising and Villar had flat-lined.

May search trends show us that Pnoy had overtaken Erap while Erap had overtaken Villar. This makes me believe that, prior to Facebook, we waited until the last minute to decide who to vote for. When all the ads and all the TV guestings and all the newspaper articles have been read, we hunker down and search for compelling reasons to vote for someone. Yes, there are people like members of the Society of Honor who read everything online and watch every interview, but the majority wait and decide during the last two weeks before the election.

Mar vs Binay

MarVsBinay

The graph showing Mar Roxas vs Jojo Binay made me think of that “Last Two Weeks conjecture.” That conjecture tells me that the campaign is long and, if I am right, this race is far from over.

Last Two Weeks Conjecture and the Political Narrative

We are biased by how we were trained, our religion, our experiences, our education and status. If people really avoid making the decision till the last two weeks, what does this warn us about how the irresponsible tabloid media will affect the last minute deciders? When you are wanting for time, there is less time for reflection and less time to figure out if people are honest about their intentions. There is less time to figure out if the points in a 20 point agenda contradict each other and there is less time to reflect on the impossibility of a statement like “Ginawa ko sa Makati gagawin ko sa buong Pilipinas” or “Nagawa ko na sa Davao gagawin ko sa Pilipinas”.

People who are politically active have to make a stand. They have to make people care, whoever their candidate is. They have to help people understand why the candidate’s statements are true, that his plans are not drawings in the sand, not motherhood statements, fluff and nothing else.

Read, Watch, Meditate and Reflect. If I have one wish for our voters, that is my wish.

__________

Note from author: “Inspired by Nate Silver of FiveThirtyEight