As the inauguration for United States President Barack Obama takes place next week, we take a quick glance back at how big data and microtargeting ultimately won Obama a second term.
With a less-than-impressive performance by Obama at the first presidential debate in October, some pollsters predicted the outcome of the election to be a close, nail-biting, photo-finish result, as it appeared that Mitt Romney was gaining ground in the weeks leading up to the election. As it turns out, this wasn’t the case. On the eve of November 6th, Barack Obama won the election by, quite honestly, a landslide. While some polls made these predictions based on opinion polling results, current economic climate, complicated mathematical model combinations and even things like instinct and hunches, they might be better off basing their predictions on tarot card readings and weather patterns when it comes to the idea of accuracy. This is where the use of big data and predictive analytics can come in quite handy.
As the phrases big data, data-mining and predictive analytics continue to rise on the popularity scale of big buzzwords, we examine how this particular method proved to be spot-on for Drew Linzer, an assistant political science professor at Emory University, who posted on his website in June 2012 that the election would be won with a result of 332 votes for Obama and 206 votes for Romney. Turns out, this was the EXACT outcome of the presidential election! So, how, you ask, did he do this? Simple. He used tarot cards and weather patterns. Ok, he actually did just the opposite. Linzer runs Votamatic, a website that uses a unified statistical model to forecast election results. The forecasts update in real-time once information from new polls becomes available and provides extremely accurate predictions.
The model also uses a unique format that fills in any holes or gaps, so to speak, by looking for common trends and similarities in voter preferences from state to state. An example of this included a contest to dine at the home of Sarah Jessica Parker in New York City that targeted a small group of women aged 40-49 who enjoy small-scale dinner parties, competitions, and are attracted to celebrity-endorsed promotions. These specific characteristics mixed with predictive analytics, allowed Obama’s team to determine exactly which type of message would persuade these women to not only vote in Obama’s favour, but to donate money to the campaign, as well. This is the exact principle of microtargeting at its finest.
From the significance of social media in 2008, to big data in 2012, what do you think will be the next big trend for 2016?
Meg Goodrich
Digital Account Executive