Credibility of an Opinion Poll: The Case of the 2021 General Elections in Zambia

Zambia went to the polls on 12th August, 2021 to elect the president of the republic of Zambia. The Presidential race seemed tight between the two main political rivals, Dr. Edgar Chagwa Lungu popularly known as ECL of the ruling Patriotic Front (PF) and Mr. Hakainde Hichilema popularly known as HH of the main opposition political party United Party for National Development (UPND). To this effect, wildly different predictions were made for Zambia’s elections. The Political Science Association of Zambia in collaboration with Faraline of the United Kingdom and Media Theory of USA showed that none of the candidates would get the outright 50+1 % of the vote but that ECL would get 40.4% and HH would get 30.33%. The research was done in five (5) provinces. A survey by Afrobarometre which was conducted in all the 10 provinces of Zambia, showed that the support for ECL had declined by 44.8% and that 25.2% said that they would vote for HH. 45.6% refused to answer and said they did not know. Another survey by Zambia Decides tipped ECL to win by 56% landslide over HH who would have 41% of the votes. However, the truth is that HH beat his rival ECL in this election amassing 59.02% of the electoral votes, while ECL managed 38.71% according to the results released by the Electoral Commission of Zambia (ECZ) on 16th August, 2021 from One hundred fifty-five (155) out of One hundred fifty-six (156) constituencies. In this article, it is argued that opinion polls add value to democracy when they are credible. Therefore, the aim of this paper is not to analyse these opinion polls but to give an in-depth understanding of what makes a credible opinion poll.