About the AJC’s September poll

Credit: Natrice Miller / Natrice.Miller@ajc.com

Credit: Natrice Miller / Natrice.Miller@ajc.com

This is the second of three polls conducted for The Atlanta Journal-Constitution leading up to the general election in November.

Polling is just one component of how The Atlanta Journal-Constitution covers politics and elections. We also background candidates, write about issues, and report on what the candidates say and do. We also cover the issue of voting and voting rights, and we provide nuts-and-bolts details about voting and other election matters.

Polling is not perfect, but it remains the most effective tool available for measuring public opinion. For every poll story, AJC reporters follow up with dozens of respondents to get a clearer picture of the results. Some of those respondents are quoted in our stories.

We know you may have questions about our polls, so we provide detailed explanations of how our polls are conducted so you can evaluate them for yourself.

Here are some common questions about our polls and their answers:

Q: Who conducted the poll?

A: The poll was conducted for the AJC by the School of Public and International Affairs Survey Research Center at the University of Georgia.

Students made the calls Sept. 5-16 under the direction of M.V. “Trey” Hood III, the director of the center and a professor of political science.

UGA began polling for the AJC in 2018. An archive of our polls can be found at www.ajc.com/news/georgia-polls/.

Q: Who selected the questions?

A: AJC editors and reporters chose the questions based on current issues in the news and questions we have about what Georgia voters care about. Some questions are the same from poll to poll. Every poll also asks a series of standard demographic questions.

Q: Whom did we talk to?

A: The School of Public and International Affairs Survey Research Center contacted 861 Georgians from across the state who are registered to vote and who indicated they are likely to vote in the November election. The numbers were randomly drawn from a voter registration list obtained through the sampling vendor L2.

The company maintains a database constructed from the state voter registration lists. Through commercial sources, phone numbers have been added to the individual records (registrants) that make up these lists.

Ninety percent of the calls were made to cellphone numbers; 10% to landlines.

Q: What is weighting, and how do you do it?

A: Some adjustments are made to the total population of people surveyed to accurately reflect the demographics of the state. This poll was weighted for race, age, sex and educational level. The poll was not weighted for party identification, which tends to fluctuate.

Q: What is the margin of error for the poll, and what exactly does that mean?

A: No matter how carefully a poll is conducted, there will always be some measure of uncertainty when you survey a small portion of a larger population, such as the state of Georgia. The margin of error is the measure of the uncertainty in the sample. For this poll, the margin of error is 3.3 percentage points.

The margin of error that we report accounts for these sources of uncertainty. For example, with a margin of error of 4 percentage points, an issue polling at 50% could have the support of anywhere between 46% and 54%, with a 95% level of confidence.

That means that if we drew 100 different samples using the same methodology, then no more than five times out of 100 should chance variations in the sample cause the results to vary by more than 4 percentage points from the answers that would be obtained if all Georgians were polled.

Q. When do you say a candidate is leading in a head-to-head matchup?

A. Sounds like a trick question, right? It might seem obvious that when candidate A’s number is higher than candidate B’s, candidate A is ahead, but because of the uncertainty described above, that might not be the case. The AJC follows The Associated Press’ guidelines for reporting on polls, which say that to describe a candidate as having a clear lead, the differences between the two numbers must be more than twice the margin of error. If the difference between the two is at least equal to the margin of error, but no more than twice the margin of error, then one candidate can be said to be apparently leading or slightly ahead. And if the difference is less than the margin of error, we say the candidates are essentially tied.