Talk Elections

How the Poll Was Conducted

By THE NEW YORK TIMES
Published: August 1, 2012

The latest Quinnipiac University/New York Times/CBS News polls involved a total of 4,392 telephone interviews conducted July 24 through July 30 with adults in Florida, Ohio and Pennsylvania.

Results are based on responses from 1,177 likely voters in Florida, 1,193 likely voters in Ohio and 1,168 likely voters in Pennsylvania. Likely voters are adults who say they are registered to vote and will “definitely” vote on Election Day.

All interviewing was conducted from the Quinnipiac University Polling Institute in Hamden, Conn. Interviews were in English or Spanish.

The telephone numbers called, both for land lines and for cellphones, were from random digit dialing samples provided by Survey Sampling International of Shelton, Conn. The regions within each state were represented in proportion to their share of all telephone numbers. Random digit dialing ensures access to listed and unlisted numbers alike. The completed samples were adjusted to ensure the proper ratio of land-line-only, cellphone-only and dual-phone users. Within each household, one adult was designated by a random procedure to be the respondent.

Interviewers made multiple attempts to reach every phone number in the survey, calling back unanswered numbers on different days at different times of both day and evening.

Each state’s results have been weighted to adjust for variation in the sample relating to region, sex, race, Hispanic origin, age and education.

In theory, in 19 cases out of 20, overall results based on such samples will differ by no more than 3 percentage points in either direction from what would have been obtained by seeking to interview all voters in each of the states. For smaller subgroups, the margin of sampling error is larger. Shifts in results between polls over time also have a larger sampling error.

In addition to sampling error, the practical difficulties of conducting any survey of public opinion may introduce other sources of error into the poll. Variation in the wording and order of questions, for example, may lead to somewhat different results.

Michael R. Kagay of Princeton, N.J., assisted The Times in its polling analysis. Complete questions and results are available at nytimes.com/polls.

http://www.nytimes.com/2012/08/01/us/politics/how-the-poll-was-conducted.html

Wednesday, August 01, 2012

Partisan Trends: GOP Edge in Partisan ID Slips to 2012 Low

The number of Americans who consider themselves Republicans fell for the second straight month while the number of Democrats held steady.

During July, 34.9% of Americans considered themselves Republicans. That’s down from 35.4% in June  and from 35.7% in May.

The number of Democrats held steady at 34.0%. That marks a partial recovery for the president’s party. In February, just 32.4% described themselves as Democrats, the lowest level ever measured by Rasmussen Reports. Still, while the 34.0% figure is the best for Democrats in 2012, it’s down from earlier years.

http://www.rasmussenreports.com/public_content/archive/mood_of_america_archive/partisan_trends/summary_of_party_affiliation

How can you sample Democrats at only +6% in a state with a huge Democratic registration advantage (Pennsylvania), but go with +9% in Florida and +8% in Ohio?

To wit: As of June, Democrats only had a 52%-48% registration advantage in Florida. So you're going to sample Democrats at twice that rate in an election where enthusiasm favors Republicans?

Meanwhile, Democrats have a 57%-43% registration advantage in Pennsylvania, and you're going to sample Dems at a lower rate than Florida?

As for Ohio, the state doesn't release partisan registration totals, so we're out of luck there, but again, it's hard to imagine why you'd sample Pennsylvania at +6% in favor of Democrats and +8% in Ohio, which is a much redder state.

Here were polls as of 11/1/2000 in FL: +7 Gore, +5 Gore, +2 Bush, +4 Bush, +5 Gore, +5 Bush. Ranges happen folks.