Insight. Antics.

An Inconvenient Poll.

In Media on April 18, 2010 at 8:23 pm

One of the past week’s big political attention-getters was the release of a New York Times/CBS News Poll on the Tea Party. Numerous pundits jumped on it as proof of the narrow-minded, predictably stereotypical folks who comprise the Tea Party. For their parts, Bill Maher, Jon Stewart, Slate, and MSNBC all made hay of its results, often entertainingly, but mainly for fulfilling their preconceived notions.

The poll finds what those who were paying attention to the news expected to find: that the Teabaggers are made up of Obama-disliking white people who list heavily to the starboard side of the ship, if you catch my drift. (Starboard = Right = Republican.)

Easy as it is to view this poll as unassailable proof that the Tea Party conforms to all the norms of the Republican base, it is inconclusive.

For while it seems plausible, I have to take issue with the lack of reporting of another recent poll. Three weeks ago, Quinnipiac University came out with a detailed poll of its own on the Tea Party. It was re-posted by Mark Halperin, but beyond that didn’t prove to have much in the way of legs. There have been other polls in recent months on this subject, as well.

The Times/CBS and QU polls each take the temperature of the Tea Party, but come up with different readings on the thermometer. I was surprised to see some of the QU poll findings, as I mentioned earlier this month. Nonetheless, they don’t appear to be any less sound than the Times/CBS ones.

There are some clear similarities among the two polls. Most notably, that nearly 90% of those identifying as Tea Partiers are white. However, there are some marked differences that should have received more attention. Some of these measures have statistical significance. The chart I created below shows some key related findings from both polls. (Click to blow it up.)

The Times found a more educated Tea Partier; Quinnipiac found that to be less accurate. The Times found more men in the group; Quinnipiac found more women. The Times found a wealthier member; Quinnipiac found a less affluent one. The Times also found a much higher share of Independents and a much lower share of Democrats than Quinnipiac.

Why did the Times/CBS poll get so much more attention and airplay? Is it because it fit the stereotype everyone thought it would? Or wanted it to?

Maybe I am missing something, but so many prominent left-slanting outlets jumping on the Times/CBS poll’s bandwagon without so much as an acknowledgment of the QU data leaves a bad taste in my mouth.

If you work in the media, you hear about these polls. You are sent the releases. You very likely already visit a website each day that would report the QU poll in addition to the Times/CBS one.

So, is it a question of brand? The New York Times and CBS News are trusted cornerstones of the news business (if you can ignore Jayson Blair’s writings or Dan Rather’s reports on George W. Bush’s whereabouts during Vietnam). However, for its part, Quinnipiac University Polling Institute is a well-respected organization that’s been around for over 20 years and is frequently cited by top newspapers (including the Times) and cable news channels (including MSNBC). From a sampling methodology and survey construction standpoint, both polling organizations are widely considered credible and reputable. So, it strikes me as suspect that only one of these polls caught fire with the left.

The two polls’ accompanying articles/releases cut right to the point, as well. Some of those reporters and comedy writers must have seen both.

The Times’ starts:

“Tea Party supporters are wealthier and more well-educated than the general public, and are no more or less afraid of falling into a lower socioeconomic class, according to the latest New York Times/CBS News poll. The 18 percent of Americans who identify themselves as Tea Party supporters tend to be Republican, white, male, married and older than 45.”

Quinnipiac’s starts:

“Only 13 percent of American voters say they are part of the Tea Party movement, a group that has more women than men; is mainly white and Republican and voted for John McCain, and strongly supports Sarah Palin, according to a Quinnipiac University national poll released today.”

The only statistical case working in favor of the Times/CBS poll is that they oversampled, trolling for more Tea Partiers to get a larger group of them to ask questions. (Afterwards, they weighted the Tea Party and non-Tea Party averages to their relative proportions.) That can be useful, but the size of a sample alone does not indicate the quality of a sample. And yes, I learned that in my Survey Industry Research Methods course sophomore year. And no, I didn’t go to Quinnipiac. Wiseass.

The major event that occurred between the dates the two polls were conducted was the signing of the health care bill. That could account for some differences, but with that issue hogging news cycles and framing the political debate for a year, it’s rather unlikely that the composition of the Tea Party changed greatly as a result of it in a mere two or three weeks.

I’m the first to admit that you can find a set of numbers to make a case for most anything. And I’m not sure what I believe about the Tea Party. Whatever these two polls show, the people who are actually out and about at these rallies constitute the Tea Party in my head right now. And most of those people are old and white and I have no idea how much money they make. But if three more polls echoing the QU poll came out in the next six weeks, and I heard about them, I bet my perception would be altered.

The point is, as this group gets more exposure and tries to be taken seriously, its makeup and supporter involvement are not fully understood , so explaining them away with one convenient poll doesn’t cut it.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your 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

%d bloggers like this: