Skip to Content
Humans and technology

Data mining shows black people waited longer than white people to vote in 2016

The result, obtained from mobile-phone records, raises an obvious question–why?
An image of voters in line
An image of voters in line

Free and fair elections are one of the hallmarks of democracy. And a key requirement is that everyone should have equal access to voting. So how do US presidential elections shape up under this criterion?

It turns out that there is room for improvement. In the 2012 presidential election, 3.5 million people waited more than an hour to vote, with some standing in line for more than five hours. What’s more, political scientists gathered convincing evidence that the delays were worse in precincts serving black people than in those serving white people.

An important question, then, is how this has changed since 2012 and whether there were still racial disparities in wait times during the 2016 presidential elections.

Today we get an answer thanks to the work of Kareem Haggag at Carnegie Mellon University in Pittsburgh and colleagues, who have found an entirely new way to measure waiting times using mobile-phone records. Their worrying conclusion is that during the 2016 election, residents of black neighborhoods waited significantly longer to vote than residents of white neighborhoods.

Until now, voting times could only be estimated using surveys of relatively small numbers of people in just a few locations. The new finding comes from the study of anonymized mobile-phone records of 10 million people across the US.

The data covers the days from November 1 to November 15, 2016; the election took place on the 8th. This spread allowed the researchers to gauge the background location activity and so more accurately study the different patterns on Election Day.

The team used the location of 117,000 polling stations around the country to find cell phones that spent more than one minute there on Election Day. However, they ignored phones that visited the polling station before or after this date to exclude individuals who live or work at the site.

After this filtering process, the team was left with data on 154,495 voters at 43,414 polling locations across the country. They then cut and diced this data to see how long individuals waited to vote and how this varied from one precinct to another.

The results make for interesting reading. The average wait to vote across the US was 19 minutes in 2016, with 18% of individuals waiting more than 30 minutes.

The team’s main result comes from an analysis of this group. To find out how wait times varied by racial group, they plotted wait times for polling stations by the fraction of black voters in that area. They found that in the areas where the percentage of black voters was as low as 0%, the wait was significantly shorter than in areas where the percentage of black voters was over 50%.

“Voters from areas [with the greatest percentage of black people] spent 19% more time at their polling locations than those in the bottom decile,” they say. “Further, [these] voters were 49% more likely to spend over 30 minutes at the polling location.”

That’s a depressing result suggesting that there is still significant improvement to be made.

An interesting question is how these differences come about and whether they are the result of racism. In 2017, Stephen Pettigrew at Harvard University found that in the 2012 election, racial differences in waiting times occurred even in areas under the same administration. “White precincts tend to get a larger allocation of voting machines and poll workers than nonwhite precincts,” he said.

This suggests that the same set of administrators provide worse service in predominantly black areas than in predominantly white areas.

Pettigrew spent some time exploring potential reasons for this. One is that the turnout in black areas has always been lower than in white areas. So it makes sense to provide more resources to white areas.

If the allocation is done correctly, waiting times should still be about the same. But the Obama effect resulted in a higher turnout of black voters, and this may have caught planners off guard (although the increased turnout of black voters in 2008 should have forewarned them).

However, Pettigrew says the racial disparities were the same in midterm elections, when President Obama was not on the ballot. So the Obama effect can’t be to blame.

Another factor is that voters with lower socioeconomic status are less likely to complain. So officials might have allocated resources in a way that minimizes complaints, and that would naturally favor wealthier precincts, which are dominated by white voters.

Another factor could be that voting machines are limited in number and indivisible. Pettigrew explains it this way: If one precinct has 75 voters and another has 100 voters, and there are three voting machines to allocate, the optimal solution is to give one machine to the smaller precinct and two machines to the larger precinct. “This will create longer lines in the smaller precinct,” he says.

But if these factors do not fully explain the difference, an inescapable conclusion is that longer waiting times for black voters are evidence of institutional discrimination.

Researchers have been careful not to point the finger in this respect. But whatever the reason, the new approach by Haggag and co provides an entirely new way to study the problem and therefore to fix it. And with the 2020 presidential election rapidly approaching, time is of the essence.

Ref: : Racial Disparities in Voting Wait Times: Evidence from Smartphone Data

Deep Dive

Humans and technology

Unlocking the power of sustainability

A comprehensive sustainability effort embraces technology, shifting from risk reduction to innovation opportunity.

Building a data-driven health-care ecosystem

Harnessing data to improve the equity, affordability, and quality of the health care system.

Drive innovation with a tech culture of ‘connect, learn, and apply’

Offer an ongoing curriculum that aligns to strategic priorities and the latest technology to drive innovation, productivity, and social-good efforts.

People are worried that AI will take everyone’s jobs. We’ve been here before.

In a 1938 article, MIT’s president argued that technical progress didn’t mean fewer jobs. He’s still right.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.