Leader

NST Leader: Poll on polls

Public affairs — the art of public policy, to put it simply — is a complicated thing. Not everyone is interested or engaged in it, even during the heat of general elections.

Add to this the fact that humans, who get asked about public policy in polls, are themselves complicated beings. They are a bundle of opinions, some of them manufactured in the heat of the moment.

Shrouding this further with mysteries of its own is the time-bound nature of polls, themselves being snapshots of views at a given time. Change time and the findings change.

So how much weight should we place on polls? Some but not much, we say. Here is why.

First, who runs the polls? Are they free and fair? Or are they sponsored, like many commercial surveys are? If so, expect a skew in the views.

Some necessary adjustments must be made to align the polls with reality. Call it a fiction deficit.

Cigarette manufacturers are known to have had a hand in designing questions to shape a positive view of their products.

This may come as a surprise, but there is very little difference between commerce and politics.

It is not hard to turn them both into profiteering ventures. Malaysia has had its examples.

Second, pollsters are not adept enough to tame the elusive nature of public opinion, as an article by Brookings Institution on polling points out.

It was there like the Himalayas when the polls were conducted, but like the optical illusion of the mirage, it was gone on polling day.

Pollsters shouldn't be surprised if this happens on Nov 19. If it does, it will not be the first time. Remember the 14th General Election? No one thought that the voters would show the door to the ruling government of 60-odd years.

Not only the pollsters got it wrong, even the political parties got it wrong. Optical illusion? Perhaps.

Thirdly, it is about the tools used to obtain data and how it gets interpreted. Some tools work, some don't. Not all pollsters acknowledge this. Besides, not all of us are walking libraries on public affairs.

Knowledge is key. As more and more information on political parties' positioning and promises reach the public, their opinions necessarily change.

This is why opinion polls in October and November have different results.

Finally, the 15th General Election, just like the 14th General Election, is turning out to be the most difficult election to predict.

Consider a few complexities. Some six million voters have joined the electoral roll this election.

A good percentage of them are between 18 and 39 years old. How will they vote? And how will the rest of the 21 million vote? How many will turn up at polling stations? We just don't know. Pollsters, too, don't.

Analysts say that if there is a low turnout, it will favour the ruling coalition; but if the turnout is more than 80 per cent, it will favour the opposition parties.

Maybe. Will they turn out in full force? Or will they not? There are just too many "ifs" to have the confidence of certainty.

Most Popular
Related Article
Says Stories