Surveys

E·N·Q·U·I·R·Y
DEMAREE J.B. RAVAL

Surveys
Sunday, 07 06, 2003

In 1948 Democrat US President Harry Truman made pollsters eat their forecasts. The pollsters had predicted an overwhelming Republican victory, with bookies giving odds of 15 and 20 to 1. The final count had Truman reelected by a plurality of over 2 million, polling 49.6 percent of the popular vote to his rival’s 45.1 percent, amassing 303 electoral votes to his rival’s 189. They called Truman’s triumph a miracle. The pollsters had been proved wrong.

In the Philippines, no such miracle albeit in reverse is going to happen in May 2004 – at least insofar as Sen. Noli de Castro is concerned. A first-term senator voted into office with the biggest number of votes ever, but with nothing yet to prove his worth as a senator, De Castro has sensibly counted himself out of the race for president. Despite his lead in recent surveys, he was candid enough to say he still has a lot to learn, that is, now is not the time for him to run. This early and happily too, the self-fulfilling result of surveys has been proved wrong, exposing the lack of a solid base of respondents who know how to choose intelligently or make their preference based on performance.

De Castro’s top lead in the surveys is a miracle in itself. His lackluster performance in the Senate has even prompted one paper to editorialize him as somebody who “has not produced an outstanding piece of legislation or done anything in the Senate that would make people sit up and notice his work.” Why, indeed, should De Castro be leading the pack of presidential wannabes? Which makes one wonder what methodology, scientific or not, goes into the conduct of surveys.

The standards for acceptable surveys are well defined: They are supposed to refer to the measurement of opinions and perceptions of the voters with regard to a candidate’s popularity, qualifications, platforms or a matter of public discussion in relation to the election, including the voter’s preferences for candidates or publicly discussed issues. But these standards seem to be woefully missing in the surveys of run-ups to the May 2004 elections.

Survey results these days give rise to many questions that require answers which defy the very numbers they present. For example, a negative 14 rating that is erased by a 28 percent gain in popularity begs the question: How did that happen, given that we have not witnessed any change for the better in our lives, or any improvement in the manner of governance?

A survey that preys on the respondent’s gullibility – or his indifferent propensity to make a choice without considering the candidates’ credentials or actual performance – is hardly the survey that educates and makes the public realize the importance of making good choices. One is led to suspect the agenda of the pollsters when the survey relies on the certainty that a predetermined set of respondents will unerringly pick one among a set of choices. Either the outfit has been brought to come up with figures that would confirm a predetermined conclusion, or that the outfit had started on the wrong preconceived premise that notwithstanding the choice made by the respondents to the survey, the conclusion must be supported at all costs.

The surveys that count are those surveys that never see print and are used exclusively to guide the drift of the campaign strategies. The survey results that see print and bandied about and flaunted by those favored by the results, should be viewed for what they really are: No better than tailor-made propaganda materials and, therefore, must never be the imprimatur on a candidate’s ability to win or predisposition to lose.

That the pollsters get paid in order to conduct surveys already militates against the validity of the survey results, or of a fair reading of the conclusions made out of the figures generated. In recent weeks we have witnessed results of surveys conducted on the same set of respondents answering the same set of questions pose by different surveys outfits. Why the different outfits should come up with different results, after they admitted they used the same methodology, is in itself confusing. It does not lend to an intelligent discussion on where our preferences are drifting, and to what extent the greater mass of the public who did not participate in the surveys is going to be swayed by the purported results.

It baffles the mind that the preferences of 2,500 respondents are allowed to speak for 38 million voters on a question like “If elections were held today, who would you vote for president?” Scientific (as the pollsters claim they are) or not, a conclusion derived out of this miniscule segment of population does not carry any weight to make the rest of those uninformed and uninvolved in the surveys go along with the results. It only tends to confuse and does not reflect intelligent choices at all.

Voters are intelligent, however varied their yardstick in making choices may be. But there is a thin line between intelligent voting and pack voting, the latter being the kind that seems to be promoted by most surveys and which causes a great disservice to the electorate. The voters should not be swayed one way or the other by the results of surveys (commissioned or not).

To regulate the use of surveys, the Fair Elections Act (Republic Act No. 9006) prohibits the publication of survey results 15 days before the election (for national candidates). This cut-off period, however, is anti-climatic; it comes too late after previous slanted surveys had already inflicted harm on the voting populace. Good candidates who do not figure in the surveys get waylaid, ambushed and routed; they drop out of the race, simply because the pollsters have said they cannot win, a proclamation that the media invariably gobble up and sensationalize. The poor good candidates never even get to the starting line of the election race; they lose early. The candidates who figure prominently in the surveys, for many millions of reasons, but who nonetheless could not hold a torch to those good candidates who have been slaughtered by the surveys, go on merrily, unhampered in the race and slug it out until election day. That is the disservice these surveys give us. The surveys decide for the voters, rightly and wrongly, but most often the kind that brings disaster to our governance.



For comments about this website:Webmaster@tribune.net.ph

The Daily Tribune © 2003