Election misinformation spreads faster than facts can be checked: 3 insights from Kenya’s experience

- Source:

Kenya went into its general election in August 2022 with one of the continent's most developed networks for countering false information.

After the disputed 2007 vote – when rumours helped drive violence in which more than 1,200 people were killed – a dense network of fact-checkers emerged in the country. This includes organisations such as Africa Check and PesaCheck, backed by newsrooms, civil society groups and big tech platforms.

Fact-checking is the careful work of verifying a claim and publishing a correction if the claim is false. Fact-checkers set as their core mission identifying misinformation (false information shared by people who do not mean to deceive) and disinformation (false information spread on purpose, usually to gain an advantage).

Yet, false claims still saturated the information environment around the vote in 2022. In a survey taken shortly before polling day, 60% of Kenyans said they had come across election information they suspected was deliberately false or inaccurate. Separately, Reuters Institute data found that at least 75% of Kenyan news consumers struggle to tell real from fake news online.

I am a comparative politics scholar who studies how African governments seek to control the flow of information – and how courts, journalists and civil society respond. In a recent paper, I examined the limits of fact-checking during Kenya's 2022 elections, drawing on practitioner interviews and an analysis of 150 fact-checks.

I argue that fact-checkers were caught in what I call a “timing trap”: even in a well-resourced verification environment, they struggled to disrupt the spread of mis/disinformation during elections.

Three forces create this trap:

• false claims travel faster than anyone can check them

• careful verification is necessarily slow

• in a polarised race, people judge a correction by whose side it appears to help rather than by whether it is true.

The obstacles

A fact-check only helps if it reaches people before they have made up their minds. During an election, that rarely happens.

My study identified three reasons for this.

1. Speed: false claims travel faster than corrections. A 2018 study at the Massachusetts Institute of Technology tracked millions of social media posts. It found that false stories spread further and faster than true ones, with the gap being widest for political news. A rumour can reach millions of phones before anyone has finished checking it. Kenya's information environment makes that speed particularly hard to outrun. By 2022 the country had 11.75 million active social media users. False claims typically surfaced first in closed WhatsApp and Telegram groups before moving to Twitter, Facebook and vernacular radio, circulating through private networks that formal verification organisations cannot monitor.

2. The nature of good fact-checking: done properly, it is slow. It means pinning down the exact claim, gathering evidence, contacting the institutions involved and passing the result through editorial review. This makes a correction trustworthy, but it costs time an election doesn't give. One case from my Kenya study shows this clearly. A forged notice, dressed up with the electoral commission's letterhead, claimed that 6,000 new polling stations would open in areas that favoured one candidate. It began spreading on WhatsApp and Twitter on 14 July 2022. Africa Check confirmed it was fake on 18 July and published the correction, by which time it had already been shared thousands of times.

3. It's political: whether people accept a correction depends on the mood around them. Kenya's 2022 race was tight and tense, and was settled only when the supreme court confirmed William Ruto's narrow win over Raila Odinga. In that atmosphere, information becomes a matter of loyalty. A correction that seems to help one side over another feels like an attack, so people reject it. Fact-checkers told me they were accused of bias for whichever claim they checked. Confusion even came from official quarters. Four election commissioners publicly disowned the presidential results moments before they were declared, leaving fact-checkers to untangle a dispute that had begun inside the very body meant to safeguard the vote.

Put together, these three forces spring the timing trap. False claims peak at the exact moment when checking is slowest and minds are most closed.

Why this matters

Governments, donors and platforms increasingly treat fact-checking as the frontline against election lies and fund it as if it can hold that line on its own.

My findings suggest we are asking one tool to do a job it cannot do alone. When it is overwhelmed, the cost is not just a few false stories. It is eroded trust in results and, as Kenya's own history shows, the danger that disputes turn violent.

The timing trap does not press with equal force everywhere, however. It bites hardest where disinformation is well organised, politics is deeply polarised and trust in institutions is low. Kenya in 2022 sat close to the worst case on all three counts: structured campaign operations producing false claims at speed, a presidential result decided by fewer than 250,000 votes out of more than 14 million, and an electoral commission that split in public days later.

Where claims spread more slowly, or feelings run cooler, honest corrections have more room to land. Ghana's 2024 election illustrates this. A coalition of fact-checking organisations was in place before polling day, and the losing candidate conceded quickly rather than contesting the count. Where the result is not in dispute, my framework predicts that verification delays carry far less cost.

Knowing when the trap is tightest tells us where to concentrate effort.

What should be done?

1. Explainers to get ahead of falsehoods. Fact-checking organisations did this in 2022, publishing explainers on how tallying works and how forged documents are made before voting began. This work belongs with electoral commissions and public broadcasters, too, since they hold the procedural detail and national reach.

2. Keep the coalitions going. Kenya's 2022 fact-checking coalition brought together Africa Check, PesaCheck, Debunk Media, mainstream newsrooms, community radio networks broadcasting in local languages and platform partnerships with Meta and Twitter. Sustaining this coalition between elections, rather than assembling it during one, falls to the fact-checking organisations that convene it and the donors that fund it.

3. Strengthen local knowledge. Kenyan fact-checkers relied on networks of contributors who could monitor claims in local languages. Sustaining this capacity needs investment and training from organisations and content platforms.

As Kenya returns to the polls in 2027, the deeper lesson is to change what we expect fact-checking to do. Its real value is not the instant rebuttal but the slow building of informational infrastructure – the archives, habits and public scepticism that a healthy democracy leans on over time.

The point is not that fact-checking failed in 2022. It is that it was placed at the wrong moment in the fight. Move it earlier, and surround it with other defences, and the timing trap starts to loosen its grip.

Michael Asiedu does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.

By Michael Asiedu, Researcher, University of St.Gallen