When it comes to a neighborhood’s political leanings, look no further than the cars or pickups on the street.
Researchers at Stanford University used a computer algorithm to sift through 50 million Google Street View images from 200 cities across the U.S. — and what they found was that cars are a shockingly good predictor of whether a neighborhood votes Republican or Democratic.
In neighborhoods with more sedans than extended-cab pickup trucks, there’s an 88 percent chance voters picked a Democrat at the polls, researchers said. And the opposite was true as well, the study found: In neighborhoods where pickups outnumber sedans, there’s an 82 percent chance an individual precinct went Republican.
The election data researchers looked at was from the 2008 presidential race between Barack Obama and John McCain, researchers said. The research was published Nov. 28 in the Proceedings of the National Academy of Sciences.
Never miss a local story.
“Using easily obtainable visual data, we can learn so much about our communities,” Fei-Fei Li, director of the Stanford Artificial Intelligence Lab and the Stanford Vision Lab, where the research was done, said in a statement.
Li added that what can be gleaned from cheap or publicly available data is often “on par with some information that takes billions of dollars to obtain via census surveys.”
The 2016 presidential election was a high profile example of how big data is a growing part of our daily lives — and how it can be exploited in a number of ways. Each campaign dumped millions of dollars into data operations, hoping to find voters, target them and get them out to the polls.
“The more you know about someone, the better you can engage with them and the more relevant you can make the communications that you send to them,” Alexander Nix, the head of Cambridge Analytica — a firm the Trump campaign paid $5 million to target voters in September 2016 alone — told NBC News. “Our job is to use data to understand audiences.”
Cambridge Analytica told NBC last year that it had about 4,000 “data points” on each of the 230 million American adults it had in its system. That data had been ferreted out through just about every source imaginable — from voters’ gym memberships to their charity donations, their loyalty cards to their online profiles, NBC reports.
And while Li’s team may not have paid millions for their data, it did take a lot of work to train computers to comb through millions and millions of images, catalog which car was which and then associate the cars with demographic data about the area — and finally, to link that data to the area’s political leanings, researchers said.
Researchers spent two weeks training the algorithm to go through the roughly 22 million cars that were pictured in 50 million Google Street View images. Then, computers were able to file each into one of nearly 3,000 categories — broken down by make, model, and year, researchers said.
If a person were doing the same work, the study said, it would have taken about 15 years to complete (assuming it took 10 seconds to catalog each image.)
On the demographic side of the equation, the study found that Volkswagens and Aston Martins tend to be found in predominantly white areas. African-American neighborhoods, meanwhile, are more like to have Chryslers, Buicks and Oldsmobiles driving around or parked on the street. Asian neighborhoods were more likely to have Hondas or Toyotas, the study found.
And make and model weren’t the only useful data points researchers identified.
“If you walk around a neighborhood looking at cars, the density of traffic sometimes tells you things as valuable as the types of cars you see on the streets,” Timnit Gebru, a study author, said in a statement. “We can use all this information in our algorithms.”
Gebru hopes the algorithm used in the study could someday help monitor carbon dioxide levels, or even improve traffic on congested streets.