Disco. spins some more tunes

I want to add a point to my response to the Disco. Inst.'s claim that TFN's survey of Texas biology teachers is a "push-poll" and "jackbooted thuggery."

That language is unbecoming and unprofessional, but we have all come to expect that from the Discovery Institute.

It is also hypocritical.

I know they read TfK, so they know why it is inaccurate and inappropriate to call the TFN survey "a push-poll," but here they go again, writing that "TFN is parading a push-poll survey of scientists they did recently." It still isn't a push poll. And the Disco. crew should know better than to throw stones when they've commissioned polls from firms accused of push polling, polls which employed nonstandard methodology and biased questions.

On the DI's website, they tout polls they commissioned from Zogby, purporting to show that Ohioans back the teaching of ID in public schools.

The problem is, Zogby's reputation as a pollster is pretty dubious:

Zogby has long been known for refusing to use sound methods in designing his samples. The use of only listed telephone numbers, and the self-selected samples of voters in his online surveys, are the two most salient problems.…

Regardless of how loopy are Zogby's results, or his sampling methods, his polls contribute to what Kathy Frankovic, in her AAPOR presidential address in 1993,[i] referred to as the "noise and clamor" of the polls. Thus, they're worth noting, if only in disbelief.

Indeed, their methodology in the Disco. poll was pretty awful. Good polls are conducted across three days (to reduce sampling error based on who is away from home on a given night) and do not sample on weekends (including Friday night), since that reduces response rates in nonstandard ways. Disco.'s poll was conducted on only two days, only one of them a weeknight. Furthermore, Zogby uses listed phone numbers rather than random dialing, which introduces bias into the subset of the population who are called. Finally, the questions Disco. asks in their survey are highly biased. Asking whether "Biology teachers should teach Darwin’s theory of evolution, but also the scientific evidence against it" assumes that there is, in fact, scientific evidence against evolution. As it stands, teaching no evidence against evolution actually does present all the existing evidence against evolution. Disco.'s poll obscures that fact. The small number of questions on Disco.'s survey could be a marker of a push poll, but the relatively small sample and the demographic questions asked at the end move it beyond that realm under standard criteria (as the extensive questions on the TFN questionnaire exonerate it of Disco.'s allegation of "push-polling").

Furthermore, Zogby just got hit with very credible accusations of push polling. Nate Silver knows from polls, and while he's using a definition slightly different than Mark Blumenthal's from yesterday, I think he makes a good case that Zogby's standards are lax, and that they are too willing to let their clients produce unscientific results, purely for the sake of propaganda. This accusation against Zogby has a long pedigree, and so long as Disco. is happy to avail themselves of such shoddy methods, they would be wise to avoid baseless and unjustified claims of push-polling against their opponents.