Mussa Dankwah Didn’t Get 2024 Election Research Right – Prof. Smart Sarpong

Prof. Smart Sarpong, a Senior Research Fellow at Kumasi Technical University, has come forward to defend his own 2024 election prediction while openly stating that fellow pollster Mussa K. Dankwah, Executive Director of Global InfoAnalytics, also did not get his projections entirely correct.

Speaking on the Ghana Se Sen Morning Show on Lawson TV/Radio, Prof. Sarpong acknowledged that while Dankwah came close, no researcher—including himself—accurately forecasted the final outcome of Ghana’s 2024 presidential election.

“Mussa Dankwah didn’t get his research right either. He was close, but no one—researcher or even pastor—got it spot on. We all made our projections based on the data we had at the time,” he said.

Dankwah’s Global InfoAnalytics predicted that President John Dramani Mahama would secure 52.2% of the vote in the first round, with Dr. Mahamudu Bawumia trailing at 41.4%.

He also forecasted that the National Democratic Congress (NDC) would win 150 parliamentary seats, while the New Patriotic Party (NPP) would secure 99 seats, with 29 constituencies too close to call. These projections sharply contrasted with Prof. Sarpong’s prediction of a runoff, with Bawumia leading at 49.1% and Mahama at 45.8%

Prof. Sarpong’s own survey, conducted 30 days before the election, was based on interviews with 100,000 Ghanaians across the country, selected specifically for being politically or socially active and concerned about community issues.

He clarified that the research was not based on guesswork, but on scientific sampling techniques and community-level engagement.

“We put what is supposed to be put in place and spoke to the people. We kept all details—where they went, how they voted. Evidence is there. This was serious work,” he emphasized.

He revealed that every individual spoken to was tracked and followed up, and that the results were shared openly with the public before the election.

“The results I published were bare to all—nothing was hidden. We documented everything and even called many people back later,” he added.

His projections had Vice President Bawumia leading with 49.1% and John Mahama trailing with 45.8%. However, the final certified results flipped those numbers, with Mahama winning by 56.55% and Bawumia managing 41.61%.

According to Prof. Sarpong, the primary reason for the deviation was low voter turnout, which was around 60.9%, far below the 81% turnout his model assumed.

“If the turnout had been around 81%, my findings would have been confirmed. The turnout affected everything,” he explained.
“My estimated turnout was not wrong based on the engagement we had. But people simply didn’t show up like we projected.”

He also cited undecided voters as a critical factor.

“At the time of our research, about 26% were undecided, meaning only 73% were sure they’d vote. That’s a major uncertainty any researcher has to navigate.”

Addressing criticism of polling errors, Prof. Sarpong urged Ghanaians to recognize that research is not prophecy.

“No researcher is God. We all estimate possible outcomes and share our findings. Whether they come true or not doesn’t mean the research is invalid.”

He pointed out that despite the criticisms, some political actors took his predictions seriously and adjusted their strategies accordingly.

“Some people worked on the research. They used it to inform their campaign. That’s the power of good data—even when the final numbers differ.”

Prof. Sarpong concluded by making a case for respecting data-driven political research in Ghana.

“If we take away emotions, we will encourage research. Let’s not attack researchers because the outcome didn’t match their predictions. Let’s learn and improve.”