I'm not a numbers man,
...as that little hiccup earlier tonight clearly shows. Can someone else, then, explain to me why anyone would trust a poll with a sample size such as this:
This poll was conducted by Princeton Survey Research Associates International from April 16-17, 2008. Telephone interviews were conducted with 1,209 registered voters. Registered voters were screened from a random-digit-dial (RDD) telephone sample of 1,356 national adults. Registration status is self-reported. Eighty-three percent of adults in the sample reported being registered. Thirty-nine percent of registered voters are Republicans or lean Republican and 53% of registered voters are Democrats or lean Democratic. Results are weighted so that the sample demographics match Census Current Population Survey parameters for gender, age, education, race, region, and population density. The overall margin of sampling error is plus or minus 3 percentage points for results based on 1,209 registered voters. (Emphasis mine. - ExKev).
A poll that sampled 39% Republican and 53% Democrat shows that both Hillary and Obama would beat McCain.
Go figure.
Obama bests McCain by 4 points (48 percent to 44 percent), and Clinton also wins by 4 points (47 to 43 percent). Neither lead is considered statistically significant.
Actually, I'd say the whole poll isn't that significant.
Comments