Baseball statistician-turned political fortune teller Nate Silver at FiveThirtyEight.com posts a devastating critique of public opinion pollster Zogby International — replete with enough charts and margin of error figures to make a math geek weep.
Silver’s long-simmering agitation about Zogby’s questionable polling methodology came to a head with the news that the pollster would release a poll that President Barack Obama’s approval rating would sink to 50-50 — a statistical outlier of 10 points or more compared to other more accurate pollsters.
So what’s the big deal? Isn’t this just a stats nerd slap fight?
To an extent, until the news media begins to shape public opinion on this seriously erratic data. Or more directly — one man’s propaganda is another man’s lousy poll.
So, Silver digs into Zogby’s numbers stretching back through the firm’s 2008 presidential, 2006 Senate and 2006 gubernatorial polling figures with comparisons to the actual vote tallies.
Zogby’s credibility whithers under Silver’s scrutiny:
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These Internet polls, simply put, are not scientific and should not be published by any legitimate news organization, at least not without an asterisk the size of an Alex Rodriguez steroidal syringe. But I’ll bet you that Matt Drudge already has the siren cued up by now.
Yeowch.
In contrast, Silver accurately predicted the 2008 presidential race results in 49 of 50 states and the District of Columbia.
h/t @mkapor on Twitter
