Howard Barrell
Over a Barrel
Could we be headed for an exceptionally high voter turnout in the elections in three weeks time? Until recently, there seems to have been a fairly general assumption that there would be a moderate to low turnout. Observers were suggesting that only about seven out of 10 of those of us whose names appear on the voters’ roll would bother to cast our ballots on June 2.
These low expectations were based on a notion that apathy had taken over among South African voters – that, yes, everyone had got excited in 1994 because it was the great liberation election, but now many of us were bored with politics.
This may, on balance, indeed be the case. As we showed in this newspaper last week, about one in four potential voters – that is resident South African adults – could not manage or be bothered to register as a voter. As a consequence, they are excluded from this election.
But according to a number of opposition politicians, apathy does not rule among those of us who are registered as voters. The politicians suggest that upwards of nine out of 10 of us may vote on the day. That would be recorded as a percentage poll of around 90 – even though it would mean that only 68% of potential voters, had cast their ballots.
Why are we registered voters apparently such a highly motivated species? The answer coming from a number of the political parties is much the same.
“It’s because many of our supporters who are now registered had to fight bloody hard to get through the bureaucratic bungles, get bar-coded IDs and get on the voters’ roll. They had to be motivated to do that. And I believe that motivation will carry through to June 2,” was the way one Democratic Party official put it to me this week.
He was evidently talking mainly about coloured and white registered voters from whom his party draws most of its support. Lower proportions of coloured and white people had bar-coded IDs when these became a requirement for registration, and the difficulties many applicants experienced in jumping through the bureaucratic hoops en route to registration tended to become a bigger issue within these two communities than it did among black people.
But evidence produced in polls conducted by, among others, the Institute for Democracy in South Africa (Idasa)/Markinor and the Human Sciences Research Council (HSRC) suggests black voters are relatively highly motivated as well. Hence predictions that we could have a high percentage poll in the 90s.
There is talk among opposition parties that millions of people tried to register more than once – not because they were trying to get fraudulent second or third votes, but because they wanted to be absolutely sure that their names made it on to the voters’ roll. Likewise, since registering, many millions are said to have since checked on the Internet and by other means to ensure that their names do indeed appear on the roll. Attempts to confirm this impression among opposition parties with the Independent Electoral Commission were unsuccessful.
If we do have a percentage poll as high as the 90s on June 2, how will it affect the share of the vote that each party achieves?
Since last year, the opinion pollsters have been arguing that, as a general trend, the larger the number of resident South African adults who turn out to vote on June 2 the smaller will be the ANC’s margin of victory. For example, the latest Idasa/ Markinor poll suggested that, if 90% of all resident South African adults actually voted on the day, the ANC would get about 59% of the vote. On the other hand, if only 63% voted, the ANC could expect to get a two-thirds majority with about 67% of the total vote.
But that Idasa/Markinor poll was measuring the voting intentions of all resident South African adults. It was not measuring the voting intentions of the relevant group on June 2, namely those people whose names appear on the voters’ roll and who will, thus, actually be able to vote on the day.
The most recent HSRC poll measured the intentions of only those people who said they had registered as voters. But the HSRC has not yet determined whether the low- poll/high-ANC-vote trend also applies in the case of registered voters only. Nonetheless, HSRC pollsters tend to the view that, among registered voters alone, the ANC will probably enjoy less of an advantage if the percentage poll is low. This is because the rigours of the registration process have weeded out many politically apathetic opposition types, leaving behind well-motivated opposition voters highly likely to get to the polls on June 2. But HSRC pollsters, along with Bob Mattes of Idasa, say they cannot be sure if this trend will apply.
That uncertainty should be partially cleared up in the next two weeks or so, when the results of more opinion polls designed, among other things, to measure the low-poll/high-vote thesis among registered voters only will be forthcoming.
So, if this election – with an ANC victory considered a foregone conclusion – is not providing you with the entertainment you had hoped for, all may not be lost. It may turn out to be a heaven-sent opportunity for you, too. For me, it’s a chance to try to correct the kind of statistical illiteracy from which I have always suffered.
07