Game of Polls
Yesterday, Sahara Reporters published the results of a poll conducted by ‘an independent, third party service’ called POLL, which showed Muhammadu Buhari defeating Goodluck Jonathan by 78% to 22%. This poll ran for 24 hours, and according to them, featured nearly 15,000 people from 21 countries, some as far away as Australia.
To their credit, they did say that the poll was unscientific, but was “engineered to provide a point of discussion, one expected to intensify over the course of the coming months”.
So let us discuss it.
This ‘poll’ is rubbish for a number of reasons.
1. I was unable to find this ‘independent, third party service’ called POLL anywhere on the internet, so the temptation is to write this off as the latest hack job by SR. However, for the sake of argument, let us assume it is not.
2. I have no idea why votes were taken from 21 countries, when it is only the citizens resident and registered to vote in ONE country that will go to the polls in a few months. This is not the Eurovision Song Contest or Big Brother Africa. Serious things should be done seriously.
3. Only those with access to the internet would have been able to vote, and those who do not have access to the internet are thereby excluded. For the time being, the former group remain firmly in the minority.
4. The poll only ran for 24 hours. So, anyone not on the internet at that time would not have seen and would not have taken the poll, introducing a further element of bias into the results.
5. Sahara Reporters is a well-known, long standing, anti-government news outlet, which means that its readership will be mostly those who hold similar sentiments and viewpoints. People tend to gravitate toward news outlets that confirm our opinions. This is called confirmation bias. It also means that a significant number of those who think the government is doing well would have gone elsewhere looking for news.
Any of the issues above will be enough to cast doubt on the validity of the ‘exercise’, if indeed it can be that, but taken together, they are fatal. It also highlights a very real need in our democracy: the absence of credible organisations who poll the electorate on various issues. The result is that anyone or group supporting any politician can simply concoct a narrative on said politician’s electability, and thus push him to commit a lot of time and resources to a fool’s errand.
There are various examples of this, but I will name just a few: On what basis, for instance, is there a clamour for Musiliu Obanikoro to run for Lagos governor? He ran for office in 2007 and got only 300,000 votes. What is the basis for an improved showing this time? More to the point, does the PDP know who would be best placed to contest in Lagos next year? I am sure that scientific polling would really help to come to a decent conclusion.
Another example: What is the basis for both Atiku and Buhari to run against Goodluck Jonathan for President? Has anything changed for both men in the eyes of the electorate even after multiple attempts? Or are they both just flying blind? Have either of them conducted a benchmark poll before declaring for President? Do any of their handlers even know what a benchmark poll is?
All these questions are necessary because in the absence of good data, the space is left open to all sorts of half-baked opinions, prognostications, and conspiracy theories about why this or that candidate lost the nomination or the election, whether or not there is any basis for such in reality. It may not have crossed the minds of some that their preferred candidate, however progressive and likeable he or she may be, was not ‘rigged out’, but is simply not popular enough to win a national election.
Rather to get to the bottom of some of the trends that drive elections in Nigeria, we remain stuck with this anti-intellectual manner of campaigning that does no one any good. Or perhaps it does do some people good. Maybe this lack of information is in fact a boon for all sorts of charlatans to cart away lots of money from gullible candidates, who are only too willing to believe that if they try just one more time, decamp just one more time, throw their hat into the ring once more, that they will finally get the office they seek.
Credible data has a way of fighting against such delusions. It does not always manage to pierce through the fog, of course, because people believe what they want to believe. It does, however, help the rest of us to put the authorities on their toes. If a solid pre-election poll or exit poll says a candidate is only leading by a few percentage points, a sudden reversal will be harder to engineer, and there will be fewer allegations of rigging. Candidates will spend their money only after knowing where they stand with the electorate, and political parties will also be better informed about which candidates to support for the nomination.
The Sahara Reporters poll is likely to be the first of many ‘polls’ which will come out as we approach voting day. Most of them will be worthless, because they will be calculated to represent a specific kind of reality. Watch what you believe.