Responding to the SWS survey

Standard

Some of its past surveys have elicited skeptical reactions before. But the most recent Social Weather Stations survey has provoked an outright accusation that its results had been rigged.

SWS released its findings on voter preferences for president of the Republic December 3, and in effect confirmed civil society and business community fears that an actor could once again be elected to that post.

Although the survey covered the period November 8 to 24, when he had not announced his candidacy, actor Fernando Poe Jr. was first in voter preferences together with Senator Noli de Castro, followed by former senator Raul Roco and President Gloria Macapagal-Arroyo. Senator Panfilo Lacson was last.

Poe was the choice of 25 percent of the 1,200 respondents, de Castro by 24 percent, Roco 18 percent, Arroyo 17 percent, and Lacson 10 percent.

What makes the results seem controversial is that Poe had been fourth among De Castro, Roco, and Arroyo during the months of June, July, August and September. Poe was also the preference of only 14 percent of those surveyed last September, which means that within less than two months he had gained 11 percentage points.

On the other hand, the results SWS released showed that voter preference for De Castro, Roco, Arroyo and Lacson had hardly changed, which makes Poe’s surge in the polls even more astounding.

At first blush, no event involving Poe can explain that surge. SWS President Mahar Mangahas suggested that something could have excited the electorate enough to push Poe’s numbers from 14 to 25 percent. Some commentators, reviewing the events during the survey period, say that what concerned the country at the time was the possibility of a constitutional crisis as a result of the impeachment complaint filed in the House of Representatives against Supreme Court Chief Justice Hilario Davide.

These commentators say that this issue had nothing to do with Poe, as indeed was the case. However, while the impeachment issue was the concern of the liberal media, the middle class, civil society and the business community—the sectors of Philippine society that had rallied to the support of Davide—it may not necessarily have been the primary concern of the Filipino majority during that period.

It may not even have been their concern at all. It was during those sixteen days that speculation had been rife about Poe’s political plans. It was during those sixteen days when his spokesman Vicente Sotto III had issued announcement after announcement that Poe was about to decide whether to run or not, and when Poe had deferred his decision at least thrice.

Please note too that the organizations that claimed to be speaking for the poor—former President Joseph Estrada’s People’s Movement Against Poverty, for example—had linked the Davide impeachment issue to the May 2004 elections, and had echoed Estrada’s support for Poe’s candidacy.

These factors together—and the public’s sense, despite Poe’s not having announced his candidacy yet, that Poe was definitely running as the grapevine had persistently suggested—could explain Poe’s surge in the SWS polls.

Former senator Raul Roco, however, has accused SWS’ Mangahas of rigging the polls. Although a lawyer, he has alleged that it is “statistically improbable” for Poe to leap from fourth to first place, and that “the only explanation” for Poe’s surge is that SWS rigged the polls because Mangahas “is a cousin of Poe.”

Social Weather Stations includes among its consultants and staff statisticians, including a former dean of the University of the Philippines Statistical Center, who should be better able than a lawyer to tell us what is statistically improbable. I suspect that a candidate’s surge by as much as 11 percentage points within six weeks is not in that category.

What is more improbable is that Mangahas, who is internationally recognized for his work in SWS and who has received an award for it, would risk both his personal reputation as well as that of SWS, which has been around for over a decade, by rigging the results of a survey for no clearly discernible reasons.

Assuming that survey results do create a bandwagon effect (that is, entice voters to vote for the sure winner), to influence the results of the elections Mangahas could do better than to rig survey results this early, and could reserve that for the last weeks before election day.

This argument is of course based on certain logical assumptions, not on factual evidence. It presumes that Mangahas would logically not risk his and SWS’ reputation in the Philippine academic community as well as the international survey community. It equally presumes that he would do so only if he were certain that rigging the results of his surveys would assure the election of Poe.

Though based solely on logic and not on empirical data, this argument is at least as good as Roco’s, which is similarly based on certain assumptions—“it’s statistically improbable” and Mangahas “is a cousin of Poe”—rather than hard evidence.

Without any evidence of malice on the part of the survey firm, accusations such as Roco’s don’t help the citizenry any. They do undermine the credibility not only of Mangahas and SWS, but also of all survey groups, among which there are admittedly shysters who’re basically mercenary and who will suit their findings to the demands of whoever commissions them. For all the complaints and skepticism that have been expressed about the results of some of their surveys, nothing in the track record of SWS nor of its closest rival Pulse Asia suggests that they belong in the latter category.

The basic issue is whether survey results are of any help to anyone, especially in the volatile and unpredictable Philippine political setting. Among other uses, surveys are supposed to provide decision makers a guide to the state of public opinion to enable them to respond more meaningfully to public needs.

It is not the responsibility of survey groups to suggest how the results of their polls may be positively used, but that of government and presidential candidates, at least in this instance. The results of the November SWS survey should thus help the Arroyo government govern more wisely, given the low ratings not only Mrs. Arroyo but her entire administration have been getting. The candidates for president themselves should be able to benefit from the results of the SWS voter preference survey, by making them reexamine their campaign strategies on the basis of their understanding of what the people are looking for in their next president. .

There is time enough for these rational responses. Those behind in the surveys should stop whining about how the masses are so benighted they worship only popularity, while those ahead should stop thinking that they have the May 2004 elections in the bag.

And yet these are exactly the responses we’re getting. What we’re getting is either smug self-satisfaction (from the Poe and De Castro camps, which have uniformly concluded that their candidates need not do anything between now and May 2004 except be their popular media selves); dismissal (from the Arroyo government, whose spokespersons dismiss the results as not yet deciding the outcome of the 2004 elections, and who thus miss what should be the main point, which is for the administration to try and do a better job between now and May); or an accusation of wrong-doing from former senator Roco.

None of these will do. The Arroyo government, Mrs. Arroyo herself, Raul Roco and even Panfilo Lacson have five months in which to respond to the results of the SWS survey, which I expect will eventually be confirmed by other reputable polling firms. They can do so by offering programs of government that can capture the imagination of the electorate, and fine-tuning their campaigns to offset Poe’s advantage. That advantage is not yet and not necessarily insurmountable. Without the appropriate responses, however, it will be—and popularity will once more triumph in 2004 as it did in 1998.

(Today/abs-cbnNEWS.com, December 6, 2003)

Leave a Reply