AP VoteCast: Methodology
AP VoteCast is a survey of the American electorate conducted in all 50 states by NORC at the University of Chicago for The Associated Press and Fox News. The survey of an estimated 140,000 registered voters is conducted Oct. 26 to Nov. 3, concluding as polls close on Election Day. AP VoteCast will deliver data about the presidential election in all 50 states as well as all Senate and governors’ races in 2020. AP VoteCast will highlight the opinions of voters and nonvoters nationwide.
AP VoteCast combines interviews with a random sample of registered voters drawn from state voter files; with self-identified registered voters conducted using NORC’s probability-based AmeriSpeak® panel, which is designed to be representative of the U.S. population; and with self-identified registered voters selected using nonprobability approaches.
Interviews are conducted in English and Spanish. Respondents may receive a small monetary incentive for completing the survey. Participants selected as part of the random sample can be contacted by phone and mail and can take the survey by phone or online. Participants selected as part of the nonprobability sample complete the survey online.
VoteCast National Survey
The national AP VoteCast survey of voters and nonvoters in 2020 will be based on the results of the 50 state-based surveys and a nationally representative survey of about 3,000 registered voters conducted between Nov. 1 and Nov. 3 on the probability-based AmeriSpeak panel. It will include about 30,000 probability interviews completed online and via telephone, and 110,000 nonprobability interviews completed online. The margin of sampling error is expected to be about plus or minus 0.6 percentage points for voters and 0.9 percentage points for nonvoters.
VoteCast State Surveys
In 20 states in 2020, AP VoteCast will be based on roughly 1,000 probability-based interviews conducted online and by phone, and roughly 3,000 nonprobability interviews conducted online. In these states, the margin of sampling error is expected to be about plus or minus 2.3 percentage points for voters and 5.5 percentage points for nonvoters.
In an additional 20 states, AP VoteCast will be based on roughly 500 probability-based interviews conducted online and by phone, and roughly 2,000 nonprobability interviews conducted online. In these states, the margin of sampling error is expected to be about plus or minus 2.9 percentage points for voters and 6.9 percentage points for nonvoters.
In the remaining 10 states, AP VoteCast will be based on about 1,000 nonprobability interviews conducted online. In these states, the margin of sampling error is expected to be about plus or minus 4.5 percentage points for voters and 11.0 percentage points for nonvoters.
Although there is no statistically agreed upon approach for calculating margins of error for nonprobability samples, these margins of error were estimated using a measure of uncertainty that incorporates the variability associated with the poll estimates, as well as the variability associated with the survey weights as a result of calibration. After calibration, the nonprobability sample yields approximately unbiased estimates.
As with all surveys, AP VoteCast is subject to multiple sources of error, including from sampling, question wording and order, and nonresponse.
Probability-based Registered Voter Sample
In each of the 40 states in which AP VoteCast will include a probability-based sample in 2020, NORC will obtain a sample of registered voters from Catalist LLC ’s registered voter database. This database includes demographic information, as well as addresses and phone numbers for registered voters, allowing potential respondents to be contacted via mail and telephone. The sample will be stratified by state, partisanship, and a modeled likelihood to respond to the postcard based on factors such as age, race, gender, voting history, and census block group education. In addition, NORC will attempt to match sampled records to a registered voter database maintained by L2, which will provide additional phone numbers and demographic information.
Prior to dialing, all probability sample records will be mailed a postcard inviting them to complete the survey either online using a unique PIN or via telephone by calling a toll-free number. Postcards will be addressed by name to the sampled registered voter if that individual is under age 35; postcards will be addressed to “registered voter” in all other cases. Telephone interviews will be conducted with the adult that answers the phone following confirmation of registered voter status in the state.
Nonprobability participants will include panelists from Dynata or Lucid, including members of its third-party panels. In addition, some registered voters will be selected from the voter file, matched to email addresses by V12, and recruited via an email invitation to the survey. Digital fingerprint software and panel-level ID validation is used to prevent respondents from completing the AP VoteCast survey multiple times.
During the initial recruitment phase of the AmeriSpeak panel, randomly selected U.S. households were sampled with a known, non-zero probability of selection from the NORC National Sample Frame and then contacted by mail, email, telephone and field interviewers (face-to-face). The panel provides sample coverage of approximately 97% of the U.S. household population. Those excluded from the sample include people with P.O. Box-only addresses, some addresses not listed in the U.S. Postal Service Delivery Sequence File and some newly constructed dwellings. Registered voter status was confirmed in field for all sampled panelists.
AP VoteCast employs a four-step weighting approach that combines the probability sample with the nonprobability sample and refines estimates at a subregional level within each state. In a general election, the 50 state surveys and the AmeriSpeak survey are weighted separately and then combined into a survey representative of voters in all 50 states.
First, weights are constructed separately for the probability sample (when available) and the nonprobability sample for each state survey. These weights are adjusted to population totals to correct for demographic imbalances in age, gender, education and race/ethnicity of the responding sample compared to the population of registered voters in each state. In 2020, the adjustment targets are derived from a combination of data from the U.S. Census Bureau’s November 2018 Current Population Survey Voting and Registration Supplement, Catalist’s voter file and the Census Bureau’s 2018 American Community Survey. Prior to adjusting to population totals, the probability-based registered voter list sample weights are adjusted for differential non-response related to factors such as availability of phone numbers, age, race and partisanship.
Second, all respondents receive a calibration weight. The calibration weight is designed to ensure the nonprobability sample is similar to the probability sample in regard to variables that are predictive of vote choice, such as partisanship or direction of the country, which cannot be fully captured through the prior demographic adjustments. The calibration benchmarks are based on regional level estimates from regression models that incorporate all probability and nonprobability cases nationwide.
Third, all respondents in each state are weighted to improve estimates for substate geographic regions. This weight combines the weighted probability (if available) and nonprobability samples, and then uses a small area model to improve the estimate within subregions of a state.
Fourth, the survey results are weighted to the actual vote count following the completion of the election. This weighting is done in 10–30 subregions within each state.
In a general election, the national survey is weighted to combine the 50 state surveys with the nationwide AmeriSpeak survey. Each of the state surveys is weighted as described. The AmeriSpeak survey receives a nonresponse-adjusted weight that is then adjusted to national totals for registered voters that in 2020 were derived from the U.S. Census Bureau’s November 2018 Current Population Survey Voting and Registration Supplement, the Catalist voter file and the Census Bureau’s 2018 American Community Survey. The state surveys are further adjusted to represent their appropriate proportion of the registered voter population for the country and combined with the AmeriSpeak survey. After all votes are counted, the national data file is adjusted to match the national popular vote for president.