One thing that struck me is this...
Though he glosses over the issue I think reading between the lines he is obliquely saying the section on various possibilities that there is some jiggerypokery going on here.
He says there are three possible causes for the sample error in the Newsweek poll
1. There was a sudden, massive shift in party ID after the R convention.
2. Given that Newsweek's polling was done on Sept. 2-3 (partially overlapping the convention), one could argue that more R's than D's would have made it a point to be home to watch the convention, thus making themselves more accessible to telephone interviewers; even after the final day of the convention, R's may have been more politically energized, making them more likely to agree to participate in a survey.
3. It could have just been plain, "old fashioned" sampling error -- just as a coin, with probabilities of 50% heads and 50% tails, can yield 60% heads in a sequence of flips, random sampling of households could have yielded excessive R's just by chance.
Ok so three possibilities. He then rules out the first and third.
If Greenberg is correct about the stability of individuals' party ID, then the first of the three explanations (a sudden shift) seems unlikely. The fact that other recent polls besides Newsweek's have obtained samples with more R's than D's seems to go against the third explanation (chance).Leaving us with option 2. Which I guess is his favoured one...
And then he leaves the subject. Which begs a few questions.
http://www.pollingreport.com/wh2004.htmFirstly there wasn't two polls there were four - Newsweek, Time and CBS and ABC/Washpost - all which appeared to experience these sampling distortions. Secondly the CBS Poll was conducted well after the RNC, six days later, in the case of Time the energised republican polling base has now been struck twice by pollsters on 2 September and 9 September.
Those republican voters must have had a second wind. Either that or a fourth option needs consideration.
4. the sampling errors are deliberate.