COURTESY OF THE NY TIMES
COURTESY OF PAUL KRUGMAN
ALL RIGHTS RESERVED TO THE LEGAL OWNERS
BOOKOFLANNES HOLDS NO RIGHTS ON THIS POST OR ARTICLE
IF REQUESTED BY LEGAL OWNERS WE WILL REMOVE IT PROMPTLY
Margin of Error Error
OK, not actually an error, but maybe a failure to take the context into account.
This Times report on the Romney campaign’s “pay no attention to the polls” memo cites the CNN poll, but seems to minimize its (literal) significance:
Even as one of the first post-convention polls by a major news organization, from CNN and ORC International, showed Mr. Obama with a slight gain, 52 to 46, over Mr. Romney, within the poll’s margin of sampling error of plus or minus 4 percentage points, the two campaigns agreed that readings right after the conventions can be ephemeral and that the race was likely to remain competitive until the end.
One point is that the margin of error is a 95 percent confidence interval, which is a pretty strict test. But anyway, the key point missing here is that there have been multiple polls showing an Obama bounce; six if I have it right, the four trackers plus CNN and now ABC. This means that in effect we have a much larger sample than in any one poll, and hence a much smaller margin of error.
Of course the Obama bounce might prove ephemeral. But it’s real, and that reality is why Nate Silver’s model has moved substantially toward Obama.
Aucun commentaire:
Enregistrer un commentaire