A Study That You Can Safely Ignore

Via Todd Zywicki at the Volokh Conspiracy, it appears that "the further left you are, the less you know about economics". This is based on a paper in Econ Journal Watch by Zeljka Buturovic and Daniel B. Klein. In addition to the problems pointed out by Matthew Yglesias, the entire basis of the study is severely flawed. The study is based on a "survey" (for some reason, they do not call it a poll) by Zogby:

The survey was administered by Zogby International by usual procedure. It was a nationwide survey of American adults, randomly selected from the Zogby International online panel routinely used in political and commercial research. On December 5, 2008, Zogby sent by email 63,986 invitations to the members of the panel.

In conducting opinion polls, you must start with a random sample to prevent sampling error (the difference between the people you interview and the people in the overall population being studied). It is good that was Zogby randomly selecting adults from the Zogby International online panel. Unfortunately, the Zogby International online panel is not randomly chosen: you can sign up for it here. This is like having an online poll, receiving 10,000 unique votes, and then "randomly selecting" 1,000 of the votes. Even though you are randomly sampling, the pool from which you are sampling was not randomly chosen. Anything coming out of such a poll (or survey) is garbage. But you don't have to take my word for it. Listen to Nate Silver:

All told, between 48 contests that he's surveyed over the past two election cycles, Zogby's Internet polls have been off by an average of 7.6 points. This is an extreme outlier with respect to absolutely anyone else in the polling community.

These Internet polls, simply put, are not scientific and should not be published by any legitimate news organization, at least not without an asterisk the size of an Alex Rodriguez steroidal syringe.

Note: As to the actual content of the Econ Journal Watch paper, it would not surprise me at all if self-described conservatives have better economic knowledge than self-described liberals. This study, however, does not prove anything either way.