Why polls do not count anymore

Poor record keeping, indifference to data generation and disdain for public opinion could be described as the major social engineering shortcomings of Nigeria and its attitude to governance. Most commentators blame the long period of military interregnums in the nation’s politics to the dubious and belligerent nature of its politics.
 
In most democracies, polling is a common denominator that helps candidates for elections, as well as, political leaders to get conversant with the issues of interest to citizens. This quality also marks out the United States of America (USA), as the bastion of democracy and good governance. Nearly two decades of Nigeria’s democracy, the question that bogs the mind of lovers of good governance and issues-based politics is, why does polling not count in the country?
 
Polling, particularly opinion research, helps institutions and governments to gauge the pulse of the people or test the acceptability of a given policy or plan of action. It represents a social science approach to governance.
 
Moreover, opinion research helps in finding out the expectations or perception of citizens on their leaders. In Nigeria, most political parties have scant regard for research and documentation; a development which most people say could be traced to the lack of faith in the electoral process.
 
What other reasons could support the assertion that polls do not count in Nigeria? Professor Chidi Odinkalu, a former chairman of Nigeria Human Rights Commission, said part of the challenge is that polling has gone through some challenges of credibility.He said: “I should start with a general point. As you probably know, polling has been in some difficulty for a while, globally. The pollsters, for instance called the UK’s 2015 general elections wrong.
 
“Most of them predicted a hung parliament with a possible Labour victory. In the event, the Conservatives won a Parliamentary majority. They also called the BREXIT referendum wrong. Even the exit polls on the night of June 23, 2016, gave it to the Remain campaign. If you remember, there were articles like these in the immediate aftermath of the referendum querying why the Pollsters got it all so wrong.”
 
The six polls, included Populus (45-55 remain); YouGov (49-51 remain); Ipsos MORI (46-49remain); Opinium (45-44 leave); ComRes (46-54 remain) and TNS (43-41 leave). But while Opinium and TNS gave what came out as a likely true picture of the BREXIT poll, their statistics were way off the mark, because BREXIT succeeded by 51 to 49 percentage points.
  
Odinkalu stated that polls are dependent on a lot of things like methodology and contexts. He added: “Increasingly also, it seems that political factors have become a major factor in polls. Major parties in the major democracies of the world keep their own pollsters. If you compare Rasmussen and Gallup numbers for instance, they are invariably not the same. The divergence is explained in terms not just of methods, but also of political leanings.”
  
On the Nigeria situation, Odinkalu, who is also a lawyer by profession, contended that “the underlying context and attitude of Nigeria to scientific data makes polling here pretty unreliable.”

Indifference, Denunciation of Poll Results
PERHAPS, the greatest challenge to polling in Nigeria is the predilection of candidates and political parties to denounce results of surveys. For instance, the survey conducted by NOIPolls for ANAP Foundation, which elicited some disagreement from the Ekiti State government in 2014, was adjudged to be objective.
  
Result from the ANAP poll clearly showed that Ayo Fayose led other candidates with 31 percent, while the incumbent, Dr. Kayode Fayemi had 29 percent, followed by Mr. Bamidele Opeyemi, who came third with three percent. But the interesting aspect of that poll, which happened to be the exception, rather than the rule; is that it discovered that 37 percent of those polled were not only undecided as to their choice of candidate, but were the crucial voters that would decide who eventually wins the election.
 
Such data could be relied upon by the campaign organisations to strengthen their messages or vary their pitch. On the other hand, the incumbent who was trailing in the survey, would have reached out to the polling agency to gather other information that could have helped it to turn the table. It is in this way that polling helps to drive issues-based politics. Until the election was won and lost, Nigerians could not say what the issues were that shaped the minds of voters, except that the teachers who were being owed salaries decided to use the election to extract their own pound of flesh.
 
Run as a not-for-profit organisation, ANAP conducted polls also for Lagos and the presidential election in 2011. In 2015 the survey it commissioned revealed that the race between former President Goodluck Jonathan of PDP would have a tough challenge from Muhammadu Buhari of APC. From the results, it was established that Buhari was leading Jonathan by a margin of two percentage points (32-30). And the survey put perspective to the race, which statistically explained the variables that defined the 2015 election, namely, rate of collection of PVCs.

Institutional Collaboration
Former National Commissioner in the Independent National Electoral Commission and Lecturer of Sociology at the University of Lagos, Prof. Lai Olurode, told The Guardian that during their stint in the commission they relied on polls to gauge the level of preparations. He added that phone-in and other audience participation programmes in radio and television help government and public institutions to sample public opinion.

The sociology professor noted that though Nigeria is not a reading nation, letters to the editor of national newspapers are veritable avenues for getting public opinion on issues of concern.

On the ways of making opinion polls credible and popular, Prof. Olurode contended that extending the sample size for survey would help to generate representative data; adding sociologists could help in drawing up the test questions. He said the people’s opinion should matter, stressing that newspaper houses like The Guardian should use its reporters across the various geopolitical zones to sample how the people feel about government policies.

Prof. Olurode disclosed that the national Bureau of Statistics (NBS) has been doing its best to generate data, even as he called for institutional collaboration to mitigate the prohibitive cost involved in public opinion research. He noted that a lot of people were becoming indifferent and skeptical about answering questions, mainly because nothing seems to have come out from their previous observations.

US Howler
THE recent presidential election in the United States of America ended up as a general howler, as virtually all the opinion polls, including ABC News/Washington Post, Ipsos, YouGov and Fox News, CNN, The New York Times, The Huffington Post, among other newspapers, magazines and cable television networks, were out of the mark in predicting a close election tending towards an eventual Hillary Clinton win.

Applying different styles, including random samples of telephone calls and online, the pollsters pointed in the same direction in the days leading to the election, which ended up wide off the mark.

There is no doubt that the research agencies would use the coming days and weeks to examine how and why polling failed to guide the people properly. Whatever the system examination would throw may differ from the points already mentioned above, especially regarding sample size, representative population of actual voters and the ratio of undecided voters that were neither polled one way or another.

Polls Or Prophecies
In the place of polls, Nigerians are rather taken in with prophecies. During elections, self-acclaimed prophets churn out predictions that help to rev up anxieties about the contestants. The credibility the people place on prophecies, which flows from extreme religiosity, help to deflect the power of scientific exposition. Apart from the entertainment value, the prophecies do not offer useful knowledge.