Topline Results: Times/Siena Poll of Likely Iowa Republican Caucusgoers

Methodology

The New York Times/Siena College poll of 432 likely Iowa Republican caucusgoers was conducted in English on cellular and landline telephones from July 28 to Aug. 1, 2023. The margin of sampling error is plus or minus 5.9 percentage points.

Sample

The survey is based on a response-rate-adjusted stratified sample of non-Democratic registered voters on the L2 voter file. The sample was selected by The New York Times in multiple steps to account for differential telephone coverage and nonresponse.

First, registered Democrats were excluded from the Iowa voter file. Iowa Republican caucus rules allow registered Democrats to re-register as Republicans at a caucus site, so excluding Democrats does introduce noncoverage bias. However, less than 2 percent of the likely Democratic caucus electorate were registered Republicans in Times/Siena polling of the Iowa caucus in 2020.

Second, records were selected. To adjust for noncoverage bias, the L2 voter file was stratified by statehouse district, party, race, gender, marital status, household size, turnout history, age and home ownership. The proportion of registrants with a telephone number and the mean expected response rate, based on prior Times/Siena polls, were calculated for each stratum. The initial selection weight was equal to the reciprocal of a stratum’s mean telephone coverage and modeled response rate. For respondents with multiple telephone numbers on the L2 file, the number with the highest modeled response rate was selected.

Fielding

The sample was stratified by party, race and region and fielded by the Siena College Research Institute, with additional field work by ReconMR, the Public Opinion Research Laboratory at the University of North Florida and the Institute of Policy and Opinion Research at Roanoke College. Interviewers asked for the person named on the voter file and ended the interview if the intended respondent was not available. Overall, 73 percent of respondents were reached on a cellular telephone.

Weighting

The survey was weighted by The Times using the R survey package in multiple steps.

First, the sample was adjusted for unequal probability of selection by stratum.

Second, the sample was weighted to match voter file-based parameters for the characteristics of non-Democratic registered voters, including targets for the self-reported educational attainment based on previous Times/Siena polls.

The following targets were used to weight the sample:

• Party x primary participation (NYT classifications based on L2 voter data)

• Age (Self-reported age, or voter file age if the respondent refuses)

• Gender (L2 data)

• Race or ethnicity (L2 model)

• Education (four categories of self-reported education, weighted to match NYT-based targets derived from Times/Siena polls, census data and the L2 voter file)

• Marital status (L2 model)

• Home ownership (L2 model)

• Metropolitan area (2013 National Center for Health Statistics Urban-Rural Classification Scheme for Counties)

• State region (NYT classifications)

• Turnout history (NYT classifications based on L2 data)

• Vote method in the 2020 elections (NYT classifications based on L2 data)

• Census block group density (A.C.S. 5-Year Census Block Group data)

• Home value (L2 data)

Likely caucusgoers

The sample of registered voters was screened to the Republican caucus universe, based on a question asking whether respondents were likely to participate in the Republican caucus, the Democratic caucus or wouldn’t attend a caucus. Only respondents who said they were likelier to participate in the Republican caucus were included in the potential caucus universe. All other respondents were excluded.

Next, respondents in the Republican caucus universe received a probability of caucusing, based on the relationship between self-reported intention to vote and validated turnout in prior Times/Siena polls. This probability includes a penalty based on the tendency for survey respondents to overreport voting; in this survey, that penalty was approximately twice as large as in a typical Times/Siena survey, based on the disparity between the implied and actual level of turnout in the Times/Siena survey of the Iowa Democratic caucus in 2020, which was conducted using the same likely voter screening questions. The probability of caucusing does not include a voter file-based turnout score, as the Iowa voter file does not contain the vote history of previous Republican caucusgoers.

The final weight was equal to the registered voter weight, multiplied by the respondent’s probability of voting in the Iowa caucus.

The margin of error accounts for the survey’s design effect, a measure of the loss of statistical power due to survey design and weighting. The design effect for the full sample of non-Democratic registered voters is 1.15. The design effect for the likely Republican caucus electorate is 1.57, which includes the added variance due to incorporating the probability that a respondent will participate in the Republican caucus.