What good are pre-election public opinion polls ?

General News : Monday December 10, 2007

FOCUS / COMMUNICATIONS

What good are pre-election public opinion polls?

By ANCHALEE KONGRUT

Universities do it. Politicians do it. Even the Council of National Security and Royal Thai Police Office conduct their own pre-election polls. As the Dec 23 election draws near, institutes release the varying results of their opinion polls, informing voters which party and politicians are forecast to win the election.

Previously, local media usually ran astrologers' predictions for the political outcome. But this year pre-election polls have stolen the show.

Here, there and everywhere, people are told which party is believed to be ahead in voter preference, what percentage of votes they are likely to receive and who is expected to win the premiership.

The details on how many people are sampled or how they are screened to represent which segment of population _ or even if they have voting rights or not _ are not always available, however.

This trend is worrying, said Chirmsak Pinthong, former senator and drafter of the 2007 constitution.

''A public opinion poll is a scientific, research-based method of data collection. Every procedure, starting from sampling selection to question formulation, must be designed to garner opinions that are as credible as possible,'' he said. ''But in Thailand, we never know who funds the poll, which group of people was sampled or what questions they were asked.

''It seems that as polls have become popular, organisations simply hire some pollsters to do them and the media buys into the results while hardly questioning their credibility,'' he said.

He tried, without success, to convince the Election Commission (EC) to ban polls a full month before the election.

Mr Chirmsak, also a member of the senate committee on public participation, had received complaints from both the public and politicians regarding the credibility of public opinion polls and then inspected the procedures that some pollsters use.

The senate committee later recommended the EC forbid the media from reporting poll results one month before the election. The EC partially followed the advice, but shortened the ban to one week before the election, starting on Dec 17.

Thailand is not alone in forbidding the publication of opinion polls before the election.

Luxembourg and Portugal impose a one-month ban.

Mr Chirmsak said voters would not miss much with a poll ban.

''Can anyone tell me what benefit voters get from reading these pre-election polls almost every week?'' he asked.

What voters may be deprived of is some fun _ and fodder for speculation or to gamble on the election result.

''For me, this type of information has become more and more like astrologers' predictions [on who will be the next prime minister],'' Mr Chirmsak said.

Unregulated, pre-election polls can become a tool to advance vested interests. Political parties have used these polls to benefit their political campaigns believing the ''ratings'' can influence or change the mind of the voters.

In that respect, opinion polls are marketing strategy, Mr Chirmsak said.

''It is similar to the way the condom industry conducts polls on people's sexual behaviour to make its products more familiar and popular,'' he said.

''This works the same way. Politicians hire pollsters to conduct surveys of opinion for the sole purpose of feeding them to the media and being broadcast nationwide.

''People watch these reports. The more the parties get exposed, the better they will be remembered,'' he said.

As a result, pre-election polls can do harm to small parties that are beyond their radar. Since they get no ''hits'' in these surveys, their names are not mentioned and voters can forget about them.

Mr Chirmsak cited the election campaign for Bangkok governor in 1995 when pre-election polls said that only Chamlong Srimuang and Bhichit Rattakul had a chance to win. Despite their accuracy _ Mr Bhichit did win the governor's seat that year _ other candidates suffered from the prediction because voters tended to cast their ballots only for those believed to have a chance.

''What happened was voters who supported another candidate, Mr Arkorn [Hoontrakul], were convinced it was useless to vote for him and instead cast their votes for Mr Bhichit,'' Mr Chirmsak said.

The only parties that benefit from large-scale opinion polls are front-runners in a tight race, in this case the People Power party and Democrat party.

However, Noppadol Kannikar, director of the well-regarded Abac poll, argued that the information derived from pre-election polls can help voters make their decision.

Without the polls, voters have no choice but to obtain the necessary information _ some of it biased _ from columnists in media, said Mr Noppadol, who gradated from Michigan University's programme on public opinion surveys.

''Believe it or not, the Abac poll found that 65-70% of our sampled respondents believed in the media and only 35-40% believed in the result of opinion polls. This is rather strange because results of professionally done surveys are scientifically tested,'' he said.

By : Bangkok Post