The just ended successful Ghana elections have deepened the country’s democratic credentials as a beacon of hope in Africa. In this respect, it is necessary to commend all stakeholders including the people of Ghana, the Electoral Commission, the Political Parties, the International Observers, the Coalition of Domestic Election Observers, the Security Agencies, the Peace Council and others.
However, there are still some issues which need to be addressed in order to further strengthen the electoral process in future. One such issue is the voter turnout which is an important parameter for determining the credibility of election outcomes.
The voter turnout, being the percentage of eligible voters who cast a ballot in an election, can only be determined at the end of the election after all the results pertaining to the number of ballots cast are appropriately collated to establish the accuracy of the total number of eligible voters who actually voted.
The voter turnout is made up of two components expressed as a proportion which is then converted to a percentage. The numerator of this proportion, being the top part of the component, represents those who actually cast their votes at the election while the denominator, being the bottom component, represents the total registered and eligible voters.
For the voter turnout figure to have any statistical validity, the numerator and the denominator in the proportion must, as a rule, refer to the same population data set. From the above explanation, it is obvious that one cannot derive a voter turnout for the whole country until collated data on the voting from all constituencies in the country are available and accordingly factored into the calculation. In the same way, voter turnout for any region of the country cannot be obtained until collated data on the voting from all constituencies within the region are available.
It follows that voter turnout in any constituency also cannot be declared until all collated voting data from all polling stations in that constituency are available and factored into the calculation. The process of data collation described earlier is therefore extremely essential to establish the top component of the proportion from which the voter turnout is obtained.
It is against this background that the announcement by the Electoral Commission (EC) in the evening of Friday, 9th December that the voter turnout was 49 percent for the country while collation of data for many constituencies were still ongoing was received with a degree of skeptism.
The Commission further expressed “utter shock and disappointment” at the purported 49 percent voter turnout considered to be the lowest since Ghana returned to multi-party democracy in 1992. However, this lamentation quickly proved to be premature as the EC later declared a voter turnout of 68.62 percent for 271 out of the 275 constituencies in the country.
The EC’s announcement of 49 percent voter turnout for the country raises a number of issues: In the first place it is of public interest to know how a declared result of 49 percent voter turnout for the country suddenly changed to 68.62 percent. In other words, the public is entitled to know from the EC by what measuring instruments it used to arrive at the 49 percent voter turnout. It would equally be interesting for the EC to come out with the coefficient of reliability attached to its result.
Answers to the above will help to delineate where the fault line of the calculation, if any, lies. Since the EC’s result was based only on 210 out of the 275 constituencies, it means as many as 65 constituencies constituting nearly a quarter of the total constituencies in the country were left out of the calculation of the purported 49 percent voter turnout for the country.
Since all the 210 constituencies consistently had constituency voter turnouts much higher than that declared by the EC, it was difficult to see and justify how the combined total voter turnout for these same 210 constituencies should fall steeply to the purported 49 percent.
One possible way by which this could happen would be to divide the total ballot cast in the 210 constituencies by the total registered voters in the whole country rather than the registered voters in the 210 constituencies. Such a methodology, however, would abrogate the fundamental rule referred to earlier that the numerator and the denominator of the proportion for the voter turnout calculation should refer to the same population data set.
In this example, the numerator will refer to the population in the 210 constituencies who exercised their voting rights while the denominator will refer to the total registered voting population in the country thus rendering the two population data sets different. Since the procedure adopted here would not be in conformity with the methodology, the result emanating thereof will fail to have the essential statistical validity.
Another serious repercussion of using the spurious methodology of only210 constituencies as proxy for the entire 275 constituencies in the country is that, the votes of the population in the remaining 65 constituencies who actually cast their votes at the election are completely ignored thus breaching the cardinal electoral principle that every vote counts.
The EC’s announcement of 49 percent voter turnout was also followed by a comparison with the 2012 voter turnout of 79 percent which was incorrectly interpreted as representing 30 percent difference instead of the correct interpretation of representing a difference of 30 percentage points.
The final general observation which raises the question of methodology for the voter turnout computation was when the voter turnout suddenly jumped from 49 percent for the 210 constituencies to 68.62 percent for 271 constituencies. To address such problems in the future, it is highly recommended that the EC should wait until all collated data on the voting from all constituencies in the country are available before declaring the voter turnout result of the country.
This would obviate the need to use any subset of constituencies as proxy to obtain only provisional and unreliable voter turnout results for the whole country. To ensure a smooth and efficient implementation of the above recommendation, it is again recommended that all remote and difficult to reach areas throughout the country should be equipped with modern networking facilities that would permit the use of the internet and other social media in these areas during elections to facilitate quick turnaround time in the data collation process of those who cast their votes to further improve the organizational effectiveness of the EC.
As mentioned in the introduction, the purpose of this paper is to assist in strengthening our democracy by drawing lessons from this election to enhance future ones. In this regard, it is hoped that all stakeholders will jointly assist in addressing the issues raised so as to build a much stronger Ghana.
By: Daasebre Prof. (Emeritus) Oti Boateng, Omanhene of New Juaben and UN Commissioner on the International Civil Service Commission.