Sunday, March 04, 2001

Subtle Bias to Polling Groups

A fortnight ago, nearly 350 people, supposedly chosen at random to reflect the nation as a whole, spent a weekend at Old Parliament House in Canberra to consider the future of reconciliation.  These "representative Australians", as they were called, discussed the issues among themselves in small working groups, and attended plenary sessions where panels of politicians and experts responded to prepared questions.

Before the weekend, the participants were asked whether governments should apologise to Aborigines.  45 per cent agreed and 48 per cent disagreed.  After their deliberations, they were polled again.  This time the support for an apology was overwhelming, with 69 per cent in favour, 19 per cent against, and the rest undecided.

Backers of the exercise, which is known as "deliberative polling", claim that this kind of change represents "the conclusions the public would reach, if people had a good opportunity to become more informed".  The ABC, which televised some of the proceedings, was delighted with the outcome, and The Australian's Mike Steketee wrote that it was pity Pauline Hanson hadn't been there to see how ordinary Australians respond when they are properly informed.

But sceptics argue that deliberative polls are just a subtle form of indoctrination, intended to strengthen political positions favoured by the cultural elites.  Having participated in one of the panels, and spoken to a few of the "representative Australians" after the weekend, I support the sceptics.

Deliberative polls are the idea of University of Texas Professor James Fishkin, and were devised largely in response to a major weakness with conventional opinion polling.

Over the years researchers have found that because many people don't like admitting they are ignorant about the issues to which they are responding, they offer random answers, or answers which they think will impress the interviewer.  In a number of classic studies, individuals offered seemingly decisive opinions about legislation and ethnic groups that did not even exist.

Professor Fishkin worked to develop techniques to provide citizens with the opportunity to come to an educated opinion on particular topics.  To date, nearly twenty deliberative polls have been held in various countries, including one in Australia dealing with the issue of a republic.

This was staged shortly before the 1999 referendum, and it produced an outcome strikingly different from the referendum itself, with nearly three-quarters of the participants supporting the "yes" position.  Results from other countries also show a marked tendency for participants in deliberative polls to move towards positions favoured by the posturing classes.

Of course, if such polls are to have any credibility, impartiality is crucial.  The "representative Australians" attending the reconciliation deliberations were certainly led to believe that the briefing paper they were given beforehand was carefully balanced, and that all relevant points of view would be propounded over the weekend.

But while Issues Deliberation Australia, the organisers of the weekend, were no doubt sincere in their desire to offer an even-handed presentation, I don't think they were successful.  Perhaps the problem lay in their inability to understand the complexities of Aboriginal issues, and the extent to which years of sanctimonious browbeating have truncated proper debate on these matters.

When I was first asked to become involved, both as a panellist and to review the written material, I was hesitant.  I suspected that I was being invited merely to give the illusion of balance to an event whose main purpose was to show that no decent human being could hold views on reconciliation contrary to those of ATSIC and the Australian Democrats.

My suspicions increased when I saw a list of the others being invited to address the gathering, and a draft of the briefing paper.  Neither showed evidence of balance, nor held any promise that the "representative Australians" would be taken much beyond feel-good moralising to a consideration of the tough questions and trade-offs involved in the key issues of reconciliation.

On the other hand, Dr Pamela Ryan, managing director of Issues Deliberation Australia, responded gracefully to my criticisms.  She said that she welcomed suggestions for other panellists, and for ways of improving the briefing document.

I gave her the names of a number of articulate Aborigines who question the value or direction of the reconciliation process, or the conventional stance on an apology.  But there were no dissidents in the final line-up of Aboriginal panellists in the plenary sessions.  I have since discovered that two of the Aborigines I suggested were not even contacted.  Another two heard nothing further after preliminary discussions, although one of them made it very clear that she wished to become involved.

And although the discussion of some issues in the final version of the briefing paper was less tendentious than in the draft, it was still a document that could be embraced by most "black armband" historians.  There was no indication, for example, that Bringing Them Home, the report on the "stolen generations" that first triggered the clamour for a national apology, is a document whose credibility is increasingly under serious challenge.

In this regard, it is unfortunate that the Lowitja O'Donoghue story did not break a week earlier.  At the very least it would have led to questions that were simply not raised during the deliberative poll.

But so long as no-one takes the results of the poll seriously, there may be some benefits from the weekend's proceedings.  They gave Aboriginal and non-Aboriginal Australians from all over the country an opportunity to meet with each other and form new friendships.  And that is what reconciliation should be all about.


ADVERTISEMENT

No comments: