Sunday, September 11, 2005

Flaws in Polls

by Sec. Rigoberto Tiglao

"Polls: Either the 'echo-chamber' for the media, or the views of political leaders and media without sufficient information and examination to represent any meaningful popular control." -J.S. Fishkin, US political scientist

Other than their sanctimonious invocations of "morality" and "truth," the anti-Gloria personalities proclaim as a mantra their claim-based on polls-that most Filipinos want her out. These anti-Gloria forces no longer even refer to the actual surveys; they merely state cavalierly-or casually write in innumerable opinion columns-that "Filipinos" want President Macapagal-Arroyo to step down from power.

For starters, the names of the two notable polling surveys should be instructive enough: Social Weather Stations; and Pulse Asia Inc. Polls capture a period-or even just a moment-of the public's perception. Boracay at the height of a typhoon is a hellish place. That doesn't mean though that Boracay isn't a great resort. A person's pulse quickens during moments of agitation and danger; that doesn't necessarily mean that his heart is breaking down.

And the people's pulse can change quickly, and not necessarily as a result of reasonable deliberation.

In early March (1995), despite the dramatic resolution of the country's power crisis, then President Fidel Ramos' popularity steeply fell, with his satisfaction ratings dropping to zero by mid-1995. (Should FVR have resigned then?) The reason for this was the Flor Contemplacion controversy. The decline in Ramos' popularity wasn't even caused by an action of his. Public outrage broke out because of a statement made by his foreign affairs secretary, a statement which, the people felt, was callous.

In November 1999, Estrada's satisfaction ratings, which were zooming up to the 70-point level, suddenly crashed to net 5 percentage points. Why? In that month there was a barrage of media exposes against him: front-page photos of the mansions allegedly owned by Estrada's mistresses, and reports of his drinking sprees in MalacaƱang. However, Estrada recovered his popularity to the 19-point level by September the following year.

The polls were invented by media, when the American newspapers Harrisburg Pennsylvanian and the Raleigh Star organized crude straw votes in 1874. The dialectic-maybe even the conspiracy-of media and polls continues. In the United States, other than political parties during election periods, US media outfits also commission poll surveys.

This prompted a noted American US political scientist J.S. Fishkin (in his book "The Voice of the People") to remark: "Polls are no more than the superficial means by which newspapers have come to speak for the people." Indeed, if newspapers do not publish polls, the latter will be as insignificant to politics as academic dissertations.

Other than capturing only a moment of public sentiment, the basic flaw of opinion polls is this: They report opinions even if the opinions are uninformed or misinformed. "Modern polling can give us back only what citizens know the moment the phone rings," another US political scientist, Robert Weissberg wrote.

And with the power of television and newspaper in the modern world, media images are enough to suddenly shift public perception. The sudden change in public opinion on the Vietnam War wasn't really due to the Pentagon Papers or student revolt or extensive debates. It was the images on television, such as the ones showing wounded US soldiers screaming in pain, and a young Vietnamese girl burnt in a napalm bombing.

With Congress turned into a venue for deliberating the impeachment issues, opposition demagogues were able to speak day in, day out-with full media coverage-with all venom and vitriol against the President. Rep. Alan Cayetano was even devious enough to keep repeating in every interview: "All we want to find out is whether the President is a thief and a liar." After people kept seeing and hearing congressmen saying for nearly three months that the President is a thief, what opinion would an ordinary Filipino express if he were suddenly interviewed by a pollster?

One pernicious impact of opinion polls is the bandwagon effect, and this is the reason why, in many countries, the publication of pollsters' findings on voting preferences are banned several weeks before elections.

A poll survey shows that a big percentage of respondents want Gloria out. "Middle forces" interpret this as the voice of the people. Therefore, they want Gloria out, too. The next month the percentage of Filipinos wanting Gloria out increases.

That is exactly what happened in recent months. "I look at the figures," former secretary Juan Santos said in a TV interview, referring to Pulse Asia's findings on the popularity of the President, "and they keep falling." Therefore, Santos joined the Hyatt 4-to increase the cabal's numbers.

A more insidious use of polls in the guise of objectivity is when introductory statements to a series of questions are posed to the respondent. Consider the following approach:

Introductory statements by pollster: President Arroyo admitted she called a Comelec official. There are some who claim that Ms Arroyo cheated in the elections.

Q 3: Do you think President Gloria cheated in the elections?

Q4: Do you think she should resign?

The questions might appear objective, but note how the responses would have totally changed if the introductory statements were different.

For example: Introductory statements-President Arroyo won in 2004 by a margin of more than 1.5 million votes over FPJ. Most of the Catholic Church, the Iglesia ni Cristo, El Shaddai, and business groups clearly supported Arroyo in that elections. It was Congress which determined that President Gloria won.

Question 1: Do you think President Gloria won only because she cheated?

Question 2: Do you think she should resign?

SWS and Pulse Asia would never have used such an approach.

The introductory statements were actually worse in their recent polls. The statements and questions were in Filipino. Their translation of the sentence "President Gloria admitted calling a Comelec official" was: "Inamin ni Pangulong Gloria na kinausap niya ang isang Comelec official."

The common usage of "inamin" is "confessed," as to a crime. But the President never confessed to a crime. And if you were an ordinary Filipino being told by a pollster that your President admitted to a crime, wouldn't you agree that he or she should be ousted?

Polls, with their flaws, can never be the basis for political action. They can never substitute for representative, republican democracy.

Some 236 representatives of the people studied in detail the impeachment charges and debated over them for weeks. The majority, 158 of them, decided to junk the complaints, and only 51, less than a fourth of them, wanted her impeached. Contrast that to Estrada's case, when 115 members of Congress voted for his impeachment.