Data vs. Intuition or How the Election is Like Climate Change

| No Comments
elections.001.jpgWe live in a data rich world.  With a few minutes on google, I can look up the current wave height in the Gulf of Maine, whether the waters in the equatorial Pacific suggest an El Nino is coming, or the percentage of time a pro football team converts a 4th and 5.  Ostensibly, the purpose of all of this data is to help us make smarter decisions--whether we're deciding if it's a good weekend to go surfing, how to manage water resources in California, or if it's better to go for it on fourth down or settle for a field goal. In reality, though, many of the decisions that we make are based on intuition.  How many times have you heard a football coach talk about how their gut or their heart informed their bold call that won the game?

The conflict between data and intuition is at the heart of much of the opposition to climate change and evolution.  Our intuition about how climate is changing is informed by our daily experience with weather.  Because weather conditions are so variable, it is hard for us to detect the trends that the climate data says are there.  Many people conclude that climate is always changing (true) and that these changes are unpredictable (not true). You can also get people concluding the right thing for the wrong reasons.  Many people interpreted Superstorm Sandy as evidence for climate change.  (BTW, I've decided that when I finally form my band, it will be named "Superstorm").  I think the jury is still out on the degree to which climate change (warmer oceans, more water vapor, less ice) helped contribute to Sandy.  This summer's drought and ocean heat wave are much better evidence for climate change.

In my opinion, the most exciting fight over data versus intuition is playing out in the media coverage of the presidential election.  The news media, even "high-information" sources like the New York Times and National Public Radio are reporting that the election is a toss-up.  While the polls are very tight, the data point to a very strong advantage for President Obama.  My favorite website is Nate Silver's FiveThirtyEightblog. This is the ultimate data-nerd view of the US elections.  He has built a rigorous statistical model, which he described in detail throughout the campaign, that blends information from state and national polls, demographic data, and economic indicators and provides a probabilistic forecast for the presidential election.  His take is that there is only a 15% chance that Mitt Romney will win on Tuesday. 
Many pundits have accused Nate of being biased.  His data-driven interpretation of the election runs counter to their intuition about the election and their preconceived ideas about the mood of the country. As with climate change, many of them have a vested interest in portraying the race as close--if your career is based on your ability to interpret the news and possibly sway voters, then it's in your best interest to make it seem like things are close. The criticism of his model has a similar flavor to many of the criticisms leveled at climate models, especially when the predictions run counter to what our gut tells us. At least election modelers will have a definitive test of their predictions tomorrow.

Election morning update: Nate Silver has the odds of an Obama victory at 91.6%.  He had a great discussion with Stephen Colbert about gut instincts vs. polls:

Leave a comment

About this Entry

This page contains a single entry by Andy Pershing published on November 5, 2012 1:00 PM.

The Phenomenal Weirdness of 2012 was the previous entry in this blog.

Celebrating 10 years of Québec-Océan is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.