MyFairElection.com's Results, Findings Mostly Positive

Editor’s note: A new Web site to track polling stations and conceived by Harvard Kennedy School Professor Archon Fung was put into use on Election Day. The Web site, MyFairElection.com, was updated in real time by reports filed by voters from throughout the country on the state of polling stations.

Fung reports on the findings and success of the project.

What We Found Out

Perhaps the most important finding from the MyFairElection.com project so far is that most people who reported on the site were fairly satisfied with their voting experience. We asked people to rate their voting experience on a star system, with one star being the worst and five stars being the best. Eighty seven percent of reports were four or five stars. About four percent were one or two star reports.

Now, is four percent a lot or a little? I think that it’s a lot. There were many people who faced very substantial obstacles to voting, who did not have the opportunity to vote or whose votes were not counted properly. So, many people reported on our site that they were unable to vote because their names were not on the lists of registered voters. Others reported that electronic voting machines were broken, or that the machines that count paper ballots were broken. A couple of people reported that their polling place had the wrong voter registration list, and so they had to check identities over the phone.

We asked people whether they encountered specific problems such as long lines, broken machines, and the like. Perhaps due to the high turnouts this year, the most common problem people reported was long lines. Eleven percent of people reported waiting between 30 minutes and one hour; eight percent of people reported waiting between one and two hours; and three percent reported waiting more than two hours. Two percent - which seems like a lot to me - reported that voting machines in their polling places were broken.

Importantly, nine percent of those who reported on our site thought that poll workers were not properly checking voters’ identities to make sure that they were eligible.

What Was Surprising

One very surprising thing to me about the reports is that people were so favorable about their voting experience despite the difficulties they faced. Many of those who reported waiting one or two hours or more still rated their voting experience very positively. It seems that the well of civic spirit is deep enough that many people cheerfully suffer inconvenience to exercise their franchise. So much of the popular reporting on this issue concerns the length of the lines to the polls, but many people seem to think that long lines are not so bad, or that waiting is well worth it.

A second surprise is that some places have systems that seem to work very well indeed. The average rating in Oregon, for example, is a perfect 5.0. People in Oregon vote by mail, and the ones who reported on our site are very satisfied, indeed proud, of this arrangement. One Oregon resident wrote on the site: “Picture this... A nice cup of peppermint tea, a comfortable recliner, a No. 2 pencil and my ballot. Voting by mail, an Oregon phenomenon is the best!”

Another surprising dimension of this project is the richness of the comments that people offered. In addition to the star rating and list of specific obstacles, we also allowed people to narrate their experience in free prose and to upload pictures. These comments and pictures convey a rich sense of the diverse voting experiences across the country.

Was the Project a Success?

As a proof-of-concept, I believe that the project is an unequivocal success. The platform itself was quite robust. It collected valuable information and showed how this approach and use of communication technology can, in principle, gather rich and deep data about the electoral process from every part of the country very quickly.

Two Things to Consider

There are two very important caveats to these results.

First, unlike public opinion surveys, the participants on MyFairElection.com were not randomly selected. They chose themselves, and so are likely to be unrepresentative of the general population. For example, those who participate are likely to have greater access to computers and communication technologies than the general population. So, those who live in disadvantaged communities where there may well be greater voting problems are likely to be under-represented in MyFairElection.

Second, about 1,700 people had reported their voting experience on the site as of Wednesday morning. During the week that the site was up, we had just over 14,000 visitors, and slightly more than 10,000 of those used the site on Election Day. That’s not bad, given that the site launched one week ago with very minimal outreach. But the number of reports are simply too small to gain a sharp picture of voting conditions throughout the country. We will try to address these problems if the project continues.