Formula used to rig votes at 2021 Ugandan election, data analysis suggests

By Matthew Kayanja

A FORMULA was used to edit results of the 2021 Ugandan presidential election, new data analysis suggests.

Data collected by the opposition National Unity Platform (NUP) party using a nationally distributed app, and official releases by the EC were compared to generate the findings.

International election observers described the 2021 election as fraudulent, with allegations of voter intimidation, and even the shutdown of the internet to ensure victory for 35-year incumbent Yoweri Museveni (pictured), of the ruling National Resistance Movement (NRM) party.

Museveni, 77, counts on the support of older voters who laud him for bringing peace after the catastrophic 1980-86 Ugandan Civil War. 

But the election marked a generational shift as younger voters became increasingly fed up with poor economic opportunities and corruption, and who supported Bobi Wine, 39, (real name Robert Kyagulanyi) the NUP’s presidential candidate.

The NUP created the ‘UVote’ App to allow supporters to send in copies of ‘Declaration of Results forms’, which detail vote tallies for each candidate certified by election observers from all parties.

They used these to form an alternative tally of votes for each district to the government’s and shared them online, before the government confiscated the documents and shut the website down.

Bobi Wine’s vote percentage seems decreased more in districts where he had low support initially (Click picture link for interactive graph showing districts)

Districts where Wine had relatively less support and Museveni more, saw a greater percentage drop between NUP tallies and EC tallies.

For example, the Kiruhara district in the Museveni stronghold of West Uganda saw Bobi Wine gain 31 per cent of the vote according to NUP figures, but this was dropped by 30 per cent in EC figures.

In comparison, the Masaka City district, in the NUP supporting central region, saw Bobi Wine receive 84 per cent of the vote according to NUP figures, and this was only dropped by 6 per cent in EC figures.

EC figures show no more than 5 per cent less support for Bobi Wine past a point of popularity, with the result being an S shaped graph where the drop in percentage vote for Wine is less the more support he had initially.

A similar pattern is also observed in reverse, where the higher support Museveni had initially, the higher his vote percentage is increased in EC results.

Museveni’s vote percentage seems increased more for districts where he had high support initially (Click picture link for interactive graph showing districts)

The only exceptions to this rule, where Bobi Wine saw the highest percentage drop despite receiving an average vote percentage, are districts concentrated in the small Teso sub-region of Uganda, posing further questions about vote manipulation there.

Outliers (highlighted in grey) are almost overwhelmingly in the Teso sub-region

A report compiled by the designers of the UVote App to explain how the election was rigged, argued results released by the EC may have been based on estimates of voter support, but still engineered to ensure Museveni’s victory

Cited evidence includes the fact the EC announced some district results before their vote count had finished.

Nicholas Cheeseman, a Birmingham University political scientist and author of the book ‘How to Rig an Election’, told Redaction Report: “The data makes total sense from an election rigging perspective – indeed, it is the same pattern that we set out in our book How to Rig an Election.

“There is less rigging where Bobi Wine is strong, because in these areas the ruling party has weaker control over the electoral process and there will be more opposition party agents in polling stations.

“Rigging is therefore harder and more likely to be exposed. Clever governments try to rig in their own strongholds because here they can control the process and cover it up”.

Rita Abrahamsen, a professor in the Graduate School of Public and International Affairs at the University of Ottawa, told Redaction Report: “If these statistics are correct, they are further proof of what we already know, namely that the January elections were a far cry from free and fair.

“While Museveni is still popular and retains legitimacy and support among sections of the population, this information shows that he knows that he is widely unpopular among others.

“In other words, his legitimate power is slipping, and he needs to resort to rigging to make sure that he doesn’t lose. The fact that NUP got 57 seats in Parliament shows this, as does the fact that Museveni share of the Presidential ballot was the lowest in his 35 years rule.

“The performance of Bobi Wine and the NUP at the polls exposed significant weaknesses in the regime’s grip on power. “

The Ugandan Electoral Commission was contacted for comment.

Featured Image: Foreign Commonwealth and Development Office @Flickr (CC BY 2.0)

Subscribe to stay updated, or follow us on FacebookTwitter, and Instagram. 

You can also keep up with our video content on YouTube.

Redaction cannot survive without your help. Support us for as little as $1 a month on Patreon:

4 thoughts on “Formula used to rig votes at 2021 Ugandan election, data analysis suggests

  1. There should only be 2 terms for a president, either 4 or 5 years at most. 35 years a president? Is he a king? It’s really outrageous!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s