FAQs: Sampling

We are often asked how to we get our samples. So let me explain…

The vast majority of our surveys are dealing with campaigns, politics and public policy. So the opinions of people who vote, and actually influence the system, matter most. Therefore, our samples are usually based on the voter registration database of a given state.

We find this method to be more accurate for political purposes than random digit dialing.

Voter registration databases have lots of information on each individual voter including gender, race, party affiliation, age, address, precinct, municipality, county, districts, and voter history. But they do not have phone numbers.

A company called Aristotle International, Inc. takes the voter databases of each available state and then appends home phone numbers to match each voter. We purchase our samples from these lists.

In targeting our surveys we can select on any of these criteria. For example, we could conduct a survey of only black voters from Cumberland County who are registered Democrats.

Most of our surveys are only targeted to likely voters. For the most recent “Dole vs. Miller” poll we created a pool of likely 2008 general election voters. That pool consisted of any voter who voted in either 2004 or 2005 or 2006 in North Carolina. Then from that list of more than 2 million voters who also had home phone numbers, Aristotle removes duplicates and randomly selects numbers for us to poll.

In addition to the sample, at the beginning of each survey we ask everyone if they are a registered voter—to make sure we are talking to the right person, and weed out non-registered phone answerers. For some surveys we may get even more specific like “…do you intend to vote in the Republican primary in 2008…” It’s not perfect, but we try our best to talk with only the people who will actually be casting ballots.

College fan crosstabs

Here are the really important crosstabs. Is Carolina liberal? Is State conservative? Now we have the answers. According to the three measures below the answer is yes, sort of.

UNC fans are the most Democratic and anti-Bush, but Wake Forest is a close second. NC State is more Republican/conservative than UNC, but not nearly as Republican as Duke.

Of course this is a measure of fandom, not alumni. A poll of alumni could yield different results.

Straw Polls

The Brunswick and Richmond County Democratic Parties held straw polls over the weekend. I usually won’t discuss straw polls on this site because they are not scientific, but it’s a slow news day.

In Brunswick County 100+ people participated and they chose John Edwards as their top presidential choice. Clinton came in second and Obama third.

Results were much different in Richmond County. They chose Barack Obama first, with the support of 27% of the 75+ people participating. Clinton got 21% and Edwards got 20%.

Why was Richmond County different? Richmond County has a slightly higher than normal black population, which might have aided Obama’s victory. In Richmond County during the 2004 primary about 33% of the Democrats were black, that’s higher than the 26% state average.

Or this could be another sign of how popular Obama is among activist Democrats—the type of people willing to volunteer, have bumper stickers and make small campaign donations.

In the gubernatorial race, Richmond County chose Beverly Perdue first with 48% and Richard Moore second with 31%.

Who else is discussing the poll?

The Dole vs. Miller poll has proven to be our most popular poll to date. Here's a sampling of where others are discussing it...

Politics1
Blue South at Swing State Project
Capital Beat
DailyKos
Blue South at DailyKos
MyDD
BlueNC
The Fix
An Eye on Washington
US Elections Zone
Pollster.com
Leftword

More on the OLF

Tom Jensen at the NC Sierra Club blog had this to say about our latest poll, specifically about the OLF:

A poll released today by Raleigh's Public Policy Polling shows that North Carolinians are strongly opposed to the proposed OLF in Washington County with 47% opposed to the plans and only 32% supportive.

These numbers came without any explanation to respondents about the potential negative implications of the site, including disastrous impact on wildlife and the quality of life for local residents. It's a good bet that the 21% of folks who didn't have an opinion would be strongly opposed if they knew all the facts.

I've been bugging Justin Guillory over at PPP for a long time about doing a poll sometime and breaking it down by whether those who respond are Carolina, State, Duke, or Wake people and he finally did it for this one.

Apparently there's at least one thing that fans of all ACC schools can agree on- they think the proposed OLF location is wrong for North Carolina. Partisans of the ACC quartet, as well as North Carolinians who identify with a different school, are all opposed to the current plans by margins ranging from 13 to 17 points.

I have to agree with Tom that our OLF question was very limited. We did not give any arguments for or against the project. It was almost a test to see if people knew enough about it with little prompting to have opinion. If we had made the arguments or explained who has come out for and against the site, the results probably would have been even more anti-OLF.

Elizabeth Dole vs. Brad Miller

According to Public Policy Polling’s latest survey, Brad Miller is within striking distance of Elizabeth Dole if he were to challenge her in 2008. Complete results here (PDF).

THE RESULTS

Dole 44%
Miller 33%
Undecided 22%

Instant Analysis: Dole is below the 50% danger mark for incumbents. While Miller’s support still has plenty of room to grow amongst Democrats and African-Americans. It’s an uphill battle, but he would be competitive.

Other results from the poll:

Do you support or oppose the construction of an outlying landing field or OLF for U.S. Navy Jets in Washington County, North Carolina?

Support 32%
Oppose 47%

Bush Job Approval

Approve 40%
Disapprove 55%

Dole Job Approval

Approve 43%
Disapprove 40%

Easley Job Approval

Approve 46%
Disapprove 38%

And one final question we added just for fun (the crosstabs for this question are sure to interest many of you).

Who is your favorite college sports team?

Duke 18%
NC State 26%
UNC 35%
Wake Forest 5%
Other 16%

AARP NC Survey

AARP North Carolina released a survey last week that was conducted by our sister company OnPoint Polling and Research. The survey shows strong support among North Carolina voters for six key AARP legislative priorities. Here is a shorthand version of the results.

Do you favor creation of a rating system for adult care homes?
Favor – 78% Oppose - 12% No Opinion - 9%

Do you favor additional state taxpayer funding so that older adults and mentally ill persons can be housed separately?
Favor – 71% Oppose - 22% No Opinion - 7%

Do you favor a law requiring health insurance companies to cover treatment of mental illness?
Favor – 66% Oppose - 24% No Opinion - 10%

Do you favor requiring hospitals to publicly report information such as infection rates, medical errors and medication mistakes?
Favor – 82% Oppose - 12% No Opinion - 6%

Do you favor re-enacting a law providing a tax credit of up to $350 for purchase of long term care insurance?
Favor – 69% Oppose – 22% No Opinion – 9%

Do you favor additional state taxpayer funding for the Home and Community Care Block Grant, providing services such as home-delivered meals, in-home aide services and adult day care?
Favor – 75% Oppose – 18% No Opinion – 8%

Complete results including the exact question wording and crosstabs can be found here (PDF).

Polling Early and Often

Adam Nagourney of the NY Times writes about “Polling Early and Often” in the presidential race. Campaigns, media organizations and others are all doing their own polls. It seems that there is a new poll every day. Today’s poll comes from Gallup.

This is unprecedented, especially so far out from Primary or Election Day. The phenomenon is affecting the campaigns. How should a campaign handle the curse or blessing of instant judgment of daily polls? A drop or rise from one day to the next can make a campaign look like its faltering or surging. The polls also provide lots of free research.

I wonder how the increase in polling is affecting races in North Carolina. Between PPP and Civitas there are two new polls on the gubernatorial primary races each month—13 months before Primary Day. We are certainly providing a service to the public and interested parties by keeping tabs on how things are going. But are the campaigns worried by the latest upcoming polls? Do they have to strategize how to deal with the outcomes? Or do they appreciate the extra and free research?

Combining Polls

Charlie Cook from the the National Journal discusses the practice of combining polls. Essentially the idea is to combine the data of two or more surveys with the exact same question wording in order to increase the sample size. By doing this you increase the accuracy and decrease the margin of error for crosstab categories.

Maybe we ought to try it. For example, it could help explain the racial dynamics in the Democratic primary races.

An ogre disagrees with our latest poll...

I think he is a libertarian.