What it was like to fact-check Kenya's election — and last week's rerun
It was surreal. That’s the way Eric Mugendi describes fact-checking August’s fake news-laden election in Kenya.
And it got worse.
"There was a lot of fake news around that part of the election,” said Mugendi, managing editor of PesaCheck, a fact-checking project based in Nairobi. “But I think it became even more prevalent after the Aug. 8 election, because that’s when there wasn’t that much information.”
Nine out of 10 Kenyan voters reported seeing fake news in the lead-up to the Aug. 8 presidential election, which incumbent Uhuru Kenyatta won, but the country’s Supreme Court ruled unlawful weeks later. The ensuing chaos not only ignited widespread violence, a constitutional crisis and made a rerun necessary, but also intensified political divisions in the country.
That meant more fake news — and more work for fact-checkers.
“There was a lot of noise, and looking at some of the facts we were checking bordered on the comical, while others were quite serious and sobering,” Mugendi said.
Working with the Election Observation Group and Code for Kenya, PesaCheck — whose main focus is public finance — helped set up a verification team to fact-check claims made on social and mainstream media outlets on election day. But the fakery didn’t stop then.
The nearly two-month period between the Supreme Court’s nullification and Kenyatta’s victory in the election rerun last week was marred by misinformation at every step of the way. Dramatic events in particular, such as when opposition leader Raila Odinga dropped out of the race in early October, caused notable upticks. Mugendi said the verification team checked about 100 claims over the course of the political crisis, a quarter of which were completely false.
Poynter caught up with Mugendi to talk about how fake news played out during the election and its subsequent rerun, the importance of fact-checking and the lasting impact of misinformation. This Q-and-A has been shortened for clarity.
We’re now a few days past the election rerun. What’s the feeling there in Kenya?
It’s still quite tense because there were a lot of random instances of violence, but for the most part everything is calm now. I think now we’re just waiting to see what next step the opposition is going to take — are they going to go to court? Are we going to see a repeat of what happened last time with the Supreme Court nullifying the results?
So I think, for the most part, it’s tense, we’re not scared of anything — we’re just nervous. We just want to see what happens next. It’s the not knowing that’s the worst thing.
Leading up to and after the August election, fake news became a big problem for Kenyan voters. How did that develop over the past couple of months, and what’s the status of misinformation in the country now?
Initially, you were seeing a lot of stories around targeting the main candidates. But with time, it took a form of its own, because I think for every step of the election, there was a burst of fake news.
We’d see a lot of fake stuff circulating about the voter registration process; you’d see some people trying to question the legitimacy of the thing. Then when it came to the party primary, there was also some fake news around it as well. You’d see stuff around candidates withdrawing and they hadn’t actually withdrawn, and there was also stuff about just trying to confuse the people who would actually vote in the process. And then, for the main election, there was also a lot of mix-ups about what would happen when you enter the polling station to vote.
There was a lot of fake news around that part of the election, but I think it became even more prevalent after the Aug. 8 election. Because that’s when there wasn’t that much information about what was actually happening in terms of how the electoral commission (was working) and what actually was informing the process. For example, there was one about the money that was supposed to plan the election not actually being paid by the government. Then there was another one about the commissioners revolting. There were fake resignation stories popping up over and over again.
You mentioned a few examples of fake news stories. Which do you think had the biggest effect on voters?
There was one that we actually covered. It was about an opinion poll that had come up just before the repeat election that was showing this repeat election would actually lead to a runoff. And what had actually happened was a couple of (people) had done a big site where they had actually sent copies of a fake survey to newsrooms hoping the newsrooms would pick up the story. So what actually came out of it is that there was a couple of stories in the press about this opinion poll, and then when the company whose name was actually attached to the poll came out to disown the poll, everyone was confused because it was supposed to be credible.
I think that’s one story that had some impact. And then another one I would say would be a story about the violence that came up after the Aug. 8 election … They were saying that these criminals wearing police uniforms were going door to door and attacking people. So there was a logical tension as a result of that, but trying to get in touch with the people actually in those places that were supposed to be affected, we’d actually ask them what’s going on and they’d say it’s calm, nothing is happening and it’s all safe.
There was a lot of misinformation about the violence immediately after the Aug. 8 election, and one of the unfortunate consequences of this is every time there’s a legitimate story about someone who was killed by police or someone who was actually attacked, most people just dismiss it as a fake story or someone who’s trying to provoke action from the general public.
What do you know about the people who are writing these stories? Is it a coordinated effort, or more ad hoc?
A lot of it looks coordinated. Because for most of the content, you can tell the leaning of the person writing it based on what sort of story they’re writing. So a lot of the stuff tends to be pro-opposition — it tries to elicit a reaction that would be in favor of the people on the side of the opposition.
There was actually some research that was done about how fake news circulates in Kenya, and they found that a lot of the people doing the circulation of fake news tend to be people that are more inclined toward the opposition. There was some pro-government fake news content being published, but a lot of it actually looks like it’s been published by people with opposition leanings.
Did you find that fact-checking efforts became less relevant because the competition was essentially defined by Odinga dropping out, or did the uncertainty create rifer grounds for fake news?
I think the fact that Raila Odinga withdrew actually made it more necessary to do the fact-checking because … people were trying to make sense of it all. In that confusion, there was some stuff that came up that was trying to confuse the masses. And then the knock-on effect of Raila’s withdrawal is there was a really low voter turnout.
One of the things that we saw, because we were doing this fact-checking during election day, most polling stations opened around six in the morning so we set up at about eight because we expected to get stories related to the polling stations. But by the time we were setting up, we found that the stories were coming in where people were saying at one polling station, for example, no one showed up or you’d find another story where people were trying to get into a polling station but were chased away.
I think the shadow of Raila Odinga’s withdrawal was actually felt in a lot of the stories we had to deal with, a lot of the fact-checking we had to do. Because the first thing that came out as a result of his withdrawal is people started publishing all these things, all these opinion pieces, about what it actually means. They were trying to base it on what the Supreme Court decided in 2013, and the fact that this was judicial precedent that they were citing when they were withdrawing led to some people coming up with sorts of theories, like why he would do something like this.
So some of the fact-checking we had to do was in line with that — what he was looking to accomplish with his withdrawal.
Do you think the divisions that were created over the course of the election — notably online — will have a permanent effect on mainstream media and trust in news?
I think so, because for the most part, the mainstream media has avoided attacking the subjects directly. But what they’ve been doing is they’ve just been trying to get talking heads and other people who are familiar with the subject, but they aren’t actually asking the hard questions.
So we’re seeing a lot more of the hard questions being asked online and we’re seeing a lot of individual Twitter users and Facebook users coming up as almost thought leaders about the subjects, so people tend to gravitate toward them and away from mainstream media because they feel that mainstream media is obsessed with the process rather than the people actually involved and what it actually means. For example, they would be doing a running tally of what’s coming out of the polling stations, but then they weren’t really asking the hard questions, like what does the low voter turnout mean and why are we in this situation?
I feel like they could have done a lot more but, because they didn’t, a lot of online media — Facebook users, Twitter users — were actually coming to the fore in terms of leadership in this situation.
And that can provide much-needed context to a situation, but it can also open up the floodgates for misinformation and fake news, right?