Thursday, July 15, 2010

Another Lousy Teabagger Poll

Alas, we have yet another poll purporting to survey the Tea Party "movement" that, in reality, does no such thing.

Gallup, its source, is a repeat offender on this matter. Its pollsters went down this same road back in March. It's a much-traveled road that leads only to a dead end, yet Gallup and every other major polling organization that has purported to survey the teabagger "movement" has insisted on this same trip to nowhere.

The critical flaw in all of this polling is that all of it is based on samples that don't reflect the actual "movement." This time around, Gallup's pollsters determined their sample by asking respondents if they were "Tea Party supporters." Back in March, they'd asked people if they were "supporters of the Tea Party movement." Pollsters have used variations on this wording to build their samples every time they've set out to survey the teabaggers.

Back in April, I outlined the many reasons why this is a problem, one that renders nearly all of the polling on this matter worthless. If you want to survey the opinions of the Tampa Bay Buccaneers, you would build your sample by asking respondents if they, in fact, played for the Tampa Bay Buccaneers. If, instead, you asked respondents if they were "supporters of the Tampa Bay Buccaneers," it's obvious that the resulting sample is going to be much larger, and the results of the questions asked of that sample will not be representative of the views of the actual Bucs. Why this logic has so doggedly escaped the allegedly professional pollsters who have made such a show of surveying the teabaggers is becoming an enduring mystery.

When Gallup asked for "Tea Party supporters," they got yet another ludicrously high number: 30% of the population. The biggest sample yet. If those at Gallup hadn't given any thought to their methodology before, that result alone should have given them serious pause, as the teabagger "movement" has never shown itself to have anything even remotely approximating those kind of numbers. Obviously, a huge chunk of that sample is made up of people identifying themselves with the "movement" who, in fact, aren't a part of it in any meaningful way, yet they're the ones being surveyed, the data they provide that which is being presented as representative of the "movement."

As I noted at the time, CBS News--seemingly inadvertently--identified this problem in one of the teabagger polls conducted in April, but utterly failed to understand its significance:

"More than three in four Tea Party supporters (78 percent) have never attended a rally or donated to a group; most have also not visited a Tea Party Web site."

In other words, they aren't a part of the Tea Party "movement" at all. Of those who identified themselves as "Tea Party supporters" in that CBS survey, only 20% said they'd actually given money to a Tea Party org and/or attended a Tea Party event. That equals 4% of the general public. That's a number that's also wildly inflated, but it's a lot closer to reality than 30%. But it's the demographics and views of that larger sample that is being persistently surveyed by pollsters and presented as representative of the "movement."

In reality, the "movement" is exactly as I described it back in April; "an astroturf project, a tiny group of more-angry-than-thoughtful conservatives whipped into a persistent lather by a well-financed campaign of misinformation and sent into the street to provide the appearance of a mass movement." If it really commanded the allegiance of 30% of the population--or of even half that--it would be able (depending on dispersion) to dictate, at will, the outcome of the ongoing Republican primaries across the country. In reality, this year's teabagger candidates have been noteworthy primarily for their inability to unseat Republican incumbents in open elections. Even in a teabagger stronghold like Texas, incumbents managed a complete shut-out against them. In contested primaries without a party incumbent in the mix, the teabagger candidates who have succeeded--Rand Paul in Kentucky, Sharon Angle in Nevada, etc.--have quickly become national embarrassments, as the spotlight falls on their nutty, fringe views.

One could make the argument that the teabaggers are of so little consequence that it doesn't really matter that we have so little real polling, and I wouldn't necessarily disagree with that reasoning, but this polling that so radically inflates their numbers plays their astroturf game of making them look like a great deal more than what they are, and to the extent that it's believed, that can only have a negative effect on our politics.