May 10, 2007
Checking the Poll
Yesterday I linked to a post at MLB Fanhouse suggesting the ESPN poll on Barry Bonds was conducted poorly. I wrote Mark Blumenthal at Pollster.com, and he responded in a post today:
Pollsters sometimes "oversample" a survey sub-population in order to increase the reliability of the results for that group. More interviews means less potential random sampling error. Before tabulating the data for the full sample, however, they "weight" back the oversample its correct proportion with the larger sample.
I checked with Gary Langer, the director of polling at ABC News, and he provided a few additional details. The ABC Polling Unit started with a nationally representative sample of 1,803 randomly selected adults interviewed between March 29 and April 4. Of these, 660 described themselves as baseball fans (on the survey's first question). Of these, 64 were African-American.
The pollsters wanted a bigger and more reliable sampling of African-Americans. So they continued calling from April 5 to April 22 and interviewed another 476 randomly sampled African Americans, of whom 139 were self-described baseball fans.
Thus (adding everything up), the ESPN/ABC survey interviewed 799 baseball fans, including 203 among African Americans. Before tabulating the data, however, they weighted the combined sample of 2,279 (the original 1,803 plus the oversample of 476 blacks) in a way that reduced the proportion of African-Americans to its correct value as determined by the U.S. Census.**
This practice is not at all unusual. The intent is to generate more statistically reliable results by race, not -- as Brown puts it -- to "generate racially charged results."
It was indeed a good poll. Thanks, Mark!
Posted by David Pinto at
10:11 PM
|
Cheating
|
TrackBack (0)
Great explanation by that guy. He should be teaching a college intro course on statistics!
It's still a jerky poll created to stir up needless controversy.
Thanks for doing extra work to clear up the oversampling issue. I'm still wondering why no data was included about what Hispanic and Asian fans think. They still kept it as just a "here's what white fans/here's what black fans think," which to me means they were trying to make it a black/white issue. Additionally, I still believe the question about Bonds "knowingly" taking steroids was poorly framed. To me the issue is more of whether or not he did it, not whether or not he knowingly did it.
Its called a push poll to support the propaganda of the mouth piece that the media is. Its a waste of time to even read. Much like 90% of the scientific community agrees on global warming but the press represents it at 50/50. Or the lovely job the press does making it look like Bush may have lied to get us in Iraqi and kill over 3000 us soldiers when he frakkin did lie. But no, Bonds is evil, no one else is taking steroids, its all Bonds fault and your kids are going to die because of Bonds and if they dont die be so morally corrupt by that bastard that they mind as well be dead.
