The research design is not exactly bias free, since they are encouraging students to "oversample" fields that are likely to have more lefty profs and under sample the righty sciences. ❋ Unknown (2004)
In addition to a general sample of the British population, our Faith Matters survey in the U.K. also included an oversample of the British Muslim population. ❋ Robert D. Putnam (2010)
There's another (less reliable in my view because they oversample Republicans) that shows the race going in the opposite direction, so in this race the surge is more uncertain, but given that Gov. ❋ Mike Lux (2010)
The CBS News/New York Times poll is thought to skew a little to the left due to its tendency to oversample Democrats. ❋ John Fund (2011)
However, the unrepresentative nature of the samples in developing countries might oversample wealthy and urban respondents and therefore underestimate the degree to which parents need to rely on children for old age support. ❋ Unknown (2010)
Both groups were the subject of a poll conducted by The Pew Research Center June 18-29, which included an oversample of respondents interviewed on cellphones. ❋ Unknown (2008)
While critics will say that the Markey poll must be biased because it is an internal poll, it's hard for that argument to get much traction given the significant oversample of Republicans. ❋ Unknown (2010)
Hmmmm…well, each of these 3 polls have a oversample of women. ❋ Unknown (2010)
The survey included an oversample of 200 Latino Catholic likely voters for a total of 295 Latino interviews. ❋ Unknown (2010)
They oversample so that they have enough African American responses to break them down. ❋ Unknown (2009)
Any professional pollster will make sure a national poll does not oversample any particular area, so right there with your bit about Cali and NY you pretty much blow your own argument out of the water. ❋ Unknown (2009)
Without a breakdown by race for this question, and knowing that the poll is (appropriately, since it is a random national sample) dominated by Whites (there is no Black oversample or weighting of Black responses reported), this simply appears to reaffirm our understanding that Whites are more likely than ever to believe that we are in a post-racial society. ❋ Unknown (2009)
You read far too much into these sub samples, They aint balanced at all - indeed the UK pollsters almost always oversample in rural areas as it kills two birds with one stone - Scotland and rural, Hence the SNP always does better labour worse. ❋ Jeff (2009)
The Ohio survey included an oversample of 27 African-Americans for a total of 104 blacks (weighted back to their share of the total population). ❋ Unknown (2009)
So did the previous poll oversample Dem/Dem leaners, does this one undersample them, or did party ID really change that much in fairly boring month? ❋ Unknown (2009)
The oversampling was simply to improve the MOE among the black sub-sample (which is normally astronomical -- NEVER trust a poll that comments on what "black people think" unless they did an oversample!) ❋ Unknown (2009)
Man have I got to clean off my [Kindle] because I have [run out] of memory!
Dude, [you need to stop] oversampling! ❋ OlyMark (2010)
