The Executive Director of Global InfoAnalytics, Mussa Dankwah, has shed light on the methodology used to determine the sample size for their recent survey on the 2024 presidential election.
According to him, the sample size was calculated using a mathematical formula based on the voting population of the country, not on anyone’s imagination.
In an interview with Alfred Ocansey on TV3’s KeyPoints, Dankwah emphasized that the sample size was not arbitrarily chosen but was based on specific criteria such as the confidence level and margin of error required for the survey. He explained, “If you are saying we have 18.5 million people in Ghana who could vote, and that you want to have a 99% confidence level with a 1.6% margin of error, the result of the sample size we need is 6,128. It is Mathematics, you can prove it or disprove it.”
Dankwah challenged critics of his methodology to scrutinize it and confirm whether the approach was rigorous or not. He stressed that the credibility of research methodologies does not depend on someone’s opinion but on the transparency and accuracy of the approach.
The survey conducted by Global InfoAnalytics showed the ruling party’s candidate, Dr. Mahamudu Bawumia, trailing behind the main opposition candidate, John Dramani Mahama, in the race for the 2025 presidency. The poll results indicated that JDM led with 54.3%, followed by DMB with 34.9%. Other candidates, including Alan Kwadwo Kyeremanten and Nana Kwame Bediako, polled lower percentages.
In response to the survey findings, the Director of Communications for the NPP, Richard Ahiagbah, raised concerns about the survey instrument and its potential impact on the respondents’ understanding of the questions asked. He cautioned against placing too much emphasis on the survey results, citing the possibility of misinterpretation or confusion among participants.
Overall, the Executive Director of Global InfoAnalytics stood by the methodology used in the survey and invited further scrutiny from researchers and stakeholders to validate the credibility of their findings.
