On reading political numbers

General news - Thursday December 13, 2007

OPINION / POLITICS AND OPINION POLLS

On reading political numbers

Political opinion polls will be banned next week prior to the election, but they are certain to come around next time. It is better that the public learn the nature of these polls and judge their credibility accordingly

By BOONRAK BOONYAKETMALA

Does public opinion really exist? If it's tricky answering this, the next question would then be: aren't the findings of the so-called public opinion polls manipulative creations rather than genuine truths? Clearly, the basic assumptions behind public opinion polls _ such as (1) the production of an opinion is within the range of possibility for everyone; (2) all opinions have the same value, regardless of the sources; and (3) there is a consensus about the legitimate questions to be asked _ are at best questionable.

On top of such obvious artificialities, it is difficult to prove whether the findings of opinion polls are representational of the prejudices, both conscious and unconscious, of the pollsters, not to mention those of their financial sponsors, or of the random samples.

Despite such limitations, over time in Thailand what are casually referred to as public opinion polls have become more significant as a factor of political outcomes, particularly at this stage when confrontation between the polarised groups in this month's general elections is on the horizon. Dozens of bureaucratic organs, university outfits, political parties, even the media, have turned themselves into pollsters, whose numerical forecasts of the next general elections have dominated the political landscape as if they were designed to take over the public space so as to guide the decision-making of the populace, who are out there but whose opinions have not been formed as yet.

No doubt, many political actors and institutions thrive on propagating their public opinion polls, where heavily politicised numbers are disguised behind the veil of science. What are the parameters for determining whether they are designed to reflect or actually guide the herd mentality of the public at large? Once the real outcomes have actually turned out to be consistent with such figures, how can one trace the real source of such decision-making?

To have a better share of the impact on such decisions, many political parties have invested in the making of what are sometimes referred to as ''push polls'', which might be considered a form of corporate public relations.

In competition with other more or less independent polls _ each of which has its own reasons for attempting to exploit the opportunity for having a say in the future of politics more than the others _ the ''push polls'' attempt to draw the media to distribute their findings for some purpose, usually hidden from audiences.

In the name of science, such figures have been calculated to produce an effect on the public through the media, whose repercussions on the real world of politics are beyond any doubt. The reported results of the polls have in one way or another found their way into the minds of the public, many of whom might not have even begun to think about the issues in question.

In this way the polls can become tools for self-fulfilled prophecies of the pollsters. Though in practice for decades, public opinion polling as a science appears to be little understood in Thailand. Initially, a kind of public opinion study was introduced to the country in the 1960s by the National Statistical Office which began conducting some large-scale surveys such as census, demographic characteristics and other information to accumulate certain social data otherwise unavailable.

By the mid-1970s, some other actors such as the National Institute of Development Administration also conducted some public opinion polls as well.

By the 1980s, Thirayuth Boonmi, then associated with the Institute of Social Research at Chulalongkorn University, launched some surveys on a variety of political issues. Because of their fresh angles, the results of such studies were popular with the press hungry for news.

In later decades, the same man has switched his skills to giving periodical interviews to the press about his own reflections on Thai politics, which curiously are reported as if they were a form of public opinion _ a phenomenon which calls for a thorough study if we are to understand Thai political culture better.

By the 1990s, the intensification of political confrontation between the military bureaucracy and the middle class in Bangkok institutionalised public opinion polling, which had been transformed into a tool to gauge the nature of such conflict.

During that decade, certain frameworks such as the Rajabhat University at Suan Dusit and Assumption University started conducting some surveys on issues pertinent to immediate public interest, whose results were often widely distributed in the press. In time, such information has become part and parcel of the media's content, particularly during the administration of Thaksin Shinawatra when political marketing reigned supreme.

Even in the 2000s, such data were spread in the media in the format of scientific findings, often with limited attention to the processes of issue identification, data collection and interpretation. By and large, the findings of such studies have been reproduced without adequate reference to the key procedures, even sample sites and sizes, not to mention the built-in prejudices and technical error probabilities.

In short, reports on the polls have been taken for granted by the media without much interest in learning about the scientific integrity, technical capability, and political motivation of the pollsters, including the sources and amounts of funding, which could reveal the possible flaws and biases hidden in the studies.

In a very real sense, then, the media have complacently relayed much of the information provided by many public opinion polls, without a critical examination of the inner makings of the polls.

For instance, the media recently reported forecasts of a police unit and the Interior Ministry which speculated on the numbers of winners and losers in the next general elections, without questioning how those somewhat precise figures had come into being in the first place.

Unsurprisingly, the reports of such heavily politicised figures in the press have often aroused the anger of some political groups which have been belittled in the process. At a minimum, some public opinion pollsters who like to think of themselves as the true professionals in the field have understandably lined up to express a deep-seated wrath at the ''pseudo'' poll-makers who have been denounced as vulgarising the science of public opinion polling.

If anything, the recent controversies emerging from public opinion polling indicate that Thai democracy could benefit from poll literacy. As polling is an inevitable component of democracy, learning to read it is certainly better than the other way around. Without the ability to read the polls, the public could be duped into believing and even acting in accord with the messages conveyed by the polls.

The development of poll literacy can of course begin with the media caring enough to ask the right questions when covering the polls. A simple paradigm may try to answer questions such as: why a given poll is being conducted, by how much and whose money, to what extent is such a poll scientific, who are the likely beneficiaries of a given poll, in what manner should the results of a particular poll be reported so that they will be treated by media audiences as isolate facts rather than absolute truths?

If the curiosity of the media regarding poll reporting and analysis matches the trickeries of the polls, both intended and unintended, the future of Thai democracy will be better served.

The writer teaches at Thammasat University in Bangkok. Comments are welcome at:

Bangkok Post