Who Benefits From Political Polling?


It seems self-evident (at least to me) that voters are ill-served by the process of political polling. Yet, oddly, few political commentators seem interested in questioning the rationale behind this process, or, its fallout.

Perhaps no aspect of America's political landscape stands before us more starkly than the almost absurdly ubiquitous urge to "poll." On the surface, the process may seem harmless, and even, ultimately, meaningless. But, a number of underlying objectives are achieved. And a number of powerful (or, power-hungry) constituencies are fed.

The mainstream media, for one, feast on superficial conflict, controversy, and drama. Polls, particularly political ones, help satisfy this craving. Polling results, always expressed in easy-to-understand percentages, provide the "voting-class" (that is, the powerless) with a sports-like sideshow, in which members of society's "leader-class" vie for supremacy.

Big-media's obsession with political polling is also a primary means through which the "common man" is conditioned to believe that he lives in a representative democracy. After all, if well-known politicians, newspapers, television networks, and websites are interested in my opinion, it must mean that I have a voice (however small) in determining my country's political and social direction.

It thus stands to reason that politicians, and their handlers, are especially enthralled with polling, which offers them the potential to provide "scientific" evidence of their "leadership" qualities to the voting public.

But, for me, the reality is that virtually all traditional mainstream polling remains, and will always remain, in service to a perpetually re-elected incumbent: the status-quo. The proof? We only need take note of the embarrassing paucity of controversial mainstream polling data, not to mention controversial (that is, truly representative) mainstream politicians.

The typical voter may be nodding at the wheel. But, I maintain, he is not yet totally asleep. But, based on the results of most national polls, you'd never know it.

Which raises the questions: How truly "mainstream" is your typical voter? And, how truly believable are polling data?

It is both amusing and distressing to persistently hear the polling process referred to as "scientific," with the implication being that it is therefore immune to purposeful distortion. To which I would counter: Yes, and Bill Clinton was a Rhodes Scholar. Could such a person lie?

But, some might additionally argue, national polling of the general public reveals them to hold, largely, the same mainstream views as their leaders, and is, therefore, validated as a meaningful process.

To which I would respond: In fact, the general public is extremely hesitant to express its true feelings to total strangers over the telephone; and, that the majority of polling processes are rife with outright fraud.

From my perspective, there is only a slender thread linking what we call "polling" with what we call "science." Simply, there are dozens of well-worn ways to deliberately skew polling results, all of them accepted as "scientific." And all of them brazenly deceptive.

For example, few realize how easy it is to word-craft political survey questions to achieve a pre-determined outcome. Every sophisticated pollster knows how to tweak a question to provide maximum "value" for his client.

And random sampling? Anyone who says that a random sample of less than 10,000 can statistically "represent" the entire population of American voters is either a liar or an idiot. What is a typical polling sample size? Almost always less than 1,000. Ridiculous.

Virtually every supposedly random sample we find in traditional survey research is fatally flawed, most often in ways that utterly strip the results of both validity and reliability.

And it is well to keep in mind that polls are designed to provide maximum convenience (and maximum profit) for polling organizations, with a minimum of effort. The result? A huge "scientific" compromise.

And, really, can you imagine your typical robotic, sub-minimum-wage (often volunteer) telephone pollster getting a reliable answer from a construction worker who has to get up from the dinner table to answer the telephone? Come on.

The bottom line? Traditional telephone surveys produce, at best, half-baked approximations of the truth, and, all-too-often, serve an agenda with only a fleeting connection to objectivity.

And who is the typical client for such pseudo-scientific endeavors? Someone with money and power, or someone who wants it. And what "value" is provided? The "right" answer, through which to "train" a target "audience."

Whatever this process is, it is not my idea of science. And it's not my idea of politics.

So, what does polling really accomplish for a voter? It can sometimes provide him with (insignificant) psychological confirmation of a choice he has already made. And it can make him feel part of a potentially "winning team."

Most often, however, it is a mechanism in his disenfranchisement, the core message of which is: Don't waste your vote on a losing candidate. Better to vote for a "winner," even though he (or she) may be the proverbial "lesser evil."

And don't bore me with the oft-heard claim that, by continually taking the public's "pulse," polling is somehow serving the public interest. This might sound impressive. But the reality tells a different tale.

So, does political polling have any intrinsic value at all? It depends on which side of the poll you're on. Better to ask: Who benefits from political polling?

My answer? It's usually everyone but the voter.

September 6, 2007