Column: Breaking the survey mirror

Published on May 14, 2013.

I must disagree with the esteemed Randy David, when in his May 9 column he lumped election surveys together with “political dynasties, religious meddling in politics, [and] corporate financing of electoral campaigns” as obstacles to modernity.

By that measure, every single modern polity that the Philippines can possibly look to as template is premodern. In fact, given that mature democracies use election surveys even more heavily than the Philippines does, by Randy’s own criteria they must be even more backward than we are.

I must quarrel especially with his reduction of the purpose of election surveys to the general notion of trending, and thus of the bandwagon. To quote the passage in full: “Interestingly, theorists of modernity do not fret over the fact that premodern societies do not measure up to these standards. They believe that societal evolution eventually favors the emergence of autonomous political systems. In short, whether or not there’s an explicit law banning them, political dynasties, religious meddling in politics, corporate financing of electoral campaigns, and the use of surveys to sway voters are bound to become less important, or even obsolete, as society becomes modern.”

That is certainly interesting, especially the idea that surveys are used to sway voters. Mahar Mangahas’ polite but pointed response to Randy’s column on May 11 explodes the myth that surveys influence voting results to any significant degree. In fact, as the surveys themselves point out, only a minority of voting-age Filipinos are aware of survey results, and even fewer say they will be influenced by them. In the latest Social Weather Stations poll, only 11 percent of respondents said survey results will shape their vote—and essentially the so-called bandwagonners and the underdoggers (SWS’ own terms) cancel each other out.

(I had a chance to say almost the same thing on’s News To Go recently, using data from a 2007 SWS survey.)

This is the survey paradox, isn’t it? It is precisely our nature to vote according to our own views—in other words, we are not influenced by survey results—that serves as the guarantee of the surveys’ accuracy.

Where does this idea that surveys sway voters come from? Perhaps from the parallel action of political alliances. They use polls, Randy wrote, “to determine the suitability of candidates for public office.” That is certainly true, and explains the diverse political fates of, say, Quezon Rep. Erin Tañada and self-proclaimed 20-year-on-the-job trainee Nancy Binay. But, as in 2007 (Sonny Trillanes) or 2010 (Nancy’s own father), nothing prevents a candidate riding a national consensus or running a disciplined campaign from breaking into the winners’ circle.

Randy’s underlying argument about the use of big money to fund surveys is also problematic. When he writes, “And it is not right that paying clients should determine the agenda and content of scientific studies,” we feel ourselves almost agreeing—but is that in fact what happens when politicians hire pollsters to sketch the political landscape? Why would prospective candidates throw away good money if they want survey institutions to merely tell them what they want to hear? “Determining the agenda and content of scientific studies” seems to me to be a misleading label to apply to the survey work of SWS and Pulse Asia.

What worries me most, as a journalist, is the near-certainty that Randy’s views will add academic cachet to a strain of criticism against surveys that I find particularly insidious. A close reading of his columns will show that Randy himself does not hold this view, but I doubt if those who do will bother with that particular nuance.

The critique can be summarized simply enough: “The surveys don’t reflect what I know, therefore they must be wrong.” Of course no one says it quite that way, not even those politicians who are most aggrieved by survey results, but that, at bottom, is the fundamental argument. The surveys do not mirror our own perceptions.

When a candidate says he cannot believe he is not faring well in the latest survey, because his campaign rallies are always packed, or because his social media presence is solid, or because he has met so many people who had promised him their votes, he is essentially arguing that there is a disconnect between survey results and his own perception of reality.

This is the same kind of concern that an ordinary citizen might feel, when she says she does not believe in surveys because in all her life she had never been approached by a pollster, or knows of anyone ever included in a survey’s random sample.

So far, so familiar. But there is an extreme version of the critique that is, to use a label that I hope won’t prove misleading, anti-masa. It is the one that says that because surveys continue to show the undistinguished Nancy Binay thriving at the top, while the accomplished Risa Hontiveros has broken into the statistical Top 12 only once (in the mid-April Pulse Asia survey), then there must be something wrong with surveys as a whole. To appropriate what was a high-velocity meme, we can call this version the “Anyare?” critique.

But scientifically conducted social surveys, I have argued before, are like a mirror, which we hold up to nature. Or, to be more precise, to public opinion. If the preelection surveys from the last six months tell us that Nancy continued to enjoy high ratings, while the eminent Ramon Magsaysay Jr. continued to lag behind despite unstinting endorsements from Ser Chief himself, we should not smash the mirror, or hurl it to the ground. We should, instead, take a closer look.

A modern democracy, in short, should not hope that surveys become obsolete. It should run more of them, the better to see society’s changing face.


Leave a comment

Filed under Newsstand: Column, Readings in Media, Readings in Politics

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s