Two graduate students in sociology at NYU also did some model building based on racial bias, which they found to be about 16%. The students, Brian McCabe and Jennifer Heerwig, think they got a bigger bias number because the respondents had "greater privacy" (bias is more openly expressed in more private situations). They did their survey over the internet.
However, they should have read, or if they did, heeded, David W .Moore's new book THE OPINION MAKERS: AN INSIDER EXPOSES THE TRUTH BEHIND THE POLLS. Moore points out that the inner net polls draw on an unrepresentative sample of the American people, one that seems to be "disproportionately white, male, young, better educated, techno-oriented, and, apparently, conservative." It is impossible to tell from the SCIENCE article how representative the NYU student's group was and consequently how trustworthy their results.
Last, but not least, Couzin, reports on the findings of James Campbell, a political scientist at the University of Buffalo, who predicts that the popular vote, based on his studies, will go to McCain.
So we can't answer the question with which the article opened. But we should bear in mind this. There is already considerable evidence that the Republicans are engaging in voter suppression and intimidation, electronic vote rigging, and inflaming of racial passions and attitudes. One only has to look at the number of Troglodytes turning up at McCain-Palin rallies to be convinced of this. If this election, as have the last two, is stolen from the American people, you can be sure the media and the powers that be will be explaining that the illusion of an Obama victory was due to faulty polls and the "Bradley effect."
Thomas Riggins is the associate editor of Political Affairs magazine and can be reached at pabooks@politicalaffairs.net.
(Note: You can view every article as one long page if you sign up as an Advocate Member, or higher).
