Peering Advocacy and Advancement Centre in Africa (PAACA) has raised concerns over transparency in the electronic transmission of results from the August 15 Osun State governorship election.
The organisation observed that same result sheets were uploaded multiple times on the Independent National Electoral Commission‘s (INEC) Result Viewing Portal (IReV), reports The Guardian.
The organisation, through EveryVotes, an independent election-observation initiative with collaboration with Legis360 said its polling-unit-level review of the election identified structural weaknesses in the IReV system that could undermine public confidence in the result verification process.
The organisation said its review found that the same result sheet was uploaded more than once for six polling units, with no indication on the portal of which version should be regarded as the current or authoritative copy.
It said the discovery was one of several transparency gaps identified during its independent review of the election.
EveryVotes independently collated Form EC 8A result sheets published on IReV for 3,749 of the state’s 3,763 polling units and found that the overall result announced by INEC was consistent with the polling-unit records.
According to INEC’s official figures, the Accord Party candidate, Ademola Adeleke, won the election with 511,067 votes, defeating the All Progressives Congress (APC) candidate, Bola Oyebamiji, who scored 444,814 votes.
Adeleke won by a margin of 66,253 votes.
However, while the independent review confirmed that the result was supported by the available polling-unit result sheets, PAACA said the weaknesses discovered in the electronic results system warranted urgent reforms.
The organisation also reported frequent handwriting and clerical errors on the physical result sheets uploaded to IReV.
More significantly, it said accredited-voter figures recorded on the physical result sheets differed from INEC’s official accredited-voter total by about four per cent.
According to the review, the discrepancy suggested that the paper records and the electronic figures generated through the Bimodal Voter Accreditation System (BVAS) did not fully reconcile.
The organisation said the findings did not alter the outcome of the election but highlighted the need for stronger safeguards around the publication and verification of election results.
Ezenwa said: “IReV made this independent verification possible, and that is its central value. The review also found structural weaknesses that warrant reform: on six polling units the same result sheet was uploaded to the portal more than once with no indication of which version is current; handwriting and clerical errors on the sheets were common; and the accredited-voter figures on the paper sheets diverge from INEC’s official accredited total by about four percent, indicating that the paper record and the electronic BVAS record do not fully reconcile.
“In full disclosure, EveryVotes notes the limits of its own review. Fourteen polling units are not yet in its collated set, which is why its party totals run about two percent below INEC’s official figures, and its turnout estimate is about two points higher than the Commission’s, a difference attributable to residual over-reads on a small number of sheets.
“None of these gaps changes the winner or the margin. EveryVotes publishes them in the interest of the same transparency it asks of others.
The organisation recommended that INEC move towards capturing results once and electronically at source, rather than relying on processes that could result in duplicate uploads or inconsistencies.
It further called for IReV to be published as a machine-readable, time-stamped and version-controlled public dataset, which would allow citizens, journalists, researchers and election observers to easily identify when a result was uploaded, whether it had been subsequently replaced and which version was authoritative.
The group also called for stronger measures to ensure that information contained in physical result sheets is reconciled with electronic accreditation data.
Despite the concerns, EveryVotes said its independent collation of the available IReV result sheets reproduced the same winner and runner-up as INEC’s official declaration, with a virtually identical margin.
It said its party totals differed from INEC’s official figures by less than two per cent because 14 polling units were not included in its collated set.
The organisation said none of the identified gaps changed the winner or the margin of victory.
The review also raised concerns about the broader conduct of the election, including vote-buying, low voter turnout, violence and inadequate provisions for voters with disabilities.
It said observers recorded voter inducement or interference by party agents at about two out of every five observed polling units, describing vote-buying as the major threat to the credibility of the election.
On voter participation, the organisation noted that INEC recorded 2,339,544 registered voters but only 1,010,684 accredited voters, representing a turnout of about 43.2 per cent.
It said the low turnout raised questions about voter confidence and participation in the electoral process.
The organisation nevertheless commended Osun voters for what they described as a largely peaceful election and INEC personnel for the near-universal availability of essential election materials.
The organisation urged INEC and other stakeholders to address the identified weaknesses ahead of future elections, stressing that the credibility of an election depends not only on the final count but also on the transparency and reliability of the process through which the public verifies the result.
