Arizona Cross-Tabs: October 2023 Times/Siena Poll

Methodology

The New York Times/Siena College polls of 3,662 registered voters in Arizona, Georgia, Michigan, Nevada, Pennsylvania and Wisconsin were conducted in English and Spanish on cellular and landline telephones from Oct. 22 to Nov. 3, 2023. When all states are joined together, the margin of sampling error is plus or minus 1.8 percentage points for all registered voters and plus or minus 2 percentage points for the likely electorate. The margin of sampling error for each state poll is plus or minus 4.4 percentage points in Arizona, Michigan and Nevada, plus or minus 4.5 points in Georgia, plus or minus 4.6 points in Pennsylvania and plus or minus 4.8 points in Wisconsin.

Sample

The survey is a response rate-adjusted stratified sample of 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, nonresponse and significant variation in the productivity of telephone numbers by state.

The L2 voter file for each state 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 samples for each state were 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 for 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, 94 percent of respondents were reached on a cellular telephone.

The instrument was translated into Spanish by ReconMR, and Spanish-speaking interviewers were assigned to the modeled Hispanic sample. Bilingual interviewers began the interview in English and were instructed to follow the lead of the respondent in determining whether to conduct the survey in English or Spanish. Monolingual Spanish-speaking respondents who were initially contacted by English-speaking interviewers were recontacted by Spanish-speaking interviewers. Overall, 12 percent of interviews among self-reported Hispanics were conducted in Spanish, including 19 percent in Nevada and 13 percent in Arizona.

Weighting — registered voters

The survey was weighted by The Times using the R survey package in multiple steps to account for the oversample of Republican voters.

First, the samples were adjusted for unequal probability of selection by stratum.

Second, the six state samples were weighted separately to match voter file-based parameters for the characteristics of registered voters by state.

The following targets were used:

• Party (party registration if available, else classification based on a model of vote choice in prior Times/Siena polls)

• 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)

• State regions (NYT classifications by county or city)

• 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)

• City type (Nevada only, added based on a post-hoc analysis of the difference between the weighted sample and voter file parameters. The weight had no meaningful effect on the topline result.)

• Census tract educational attainment (Georgia only, added based on a post-hoc analysis of the difference between the weighted sample and voter file parameters. The weight had no meaningful effect on the topline result.)

Finally, the six state samples were balanced to each represent one-sixth of the sum of the weights.

Weighting — likely electorate

The survey was weighted by The Times using the R survey package in multiple steps to account for the oversample of Republican voters.

First, the samples were adjusted for unequal probability of selection by stratum.

Second, the first-stage weight was adjusted to account for the probability that a registrant would vote in the 2024 election, based on a model of turnout in the 2020 election.

Third, the six state samples were weighted separately to match targets for the composition of the likely electorate. The targets for the composition of the likely electorate were derived by aggregating the individual-level turnout estimates described in the previous step for registrants on the L2 voter file. The categories used in weighting were the same as those previously mentioned for registered voters.

Fourth, the initial likely electorate weight was adjusted to incorporate self-reported vote intention. The final probability that a registrant would vote in the 2024 election was four-fifths based on their ex ante modeled turnout score and one-fifth based on their self-reported intention, based on prior Times/Siena polls, including a penalty to account for the tendency of survey respondents to turn out at higher rates than nonrespondents. The final likely electorate weight was equal to the modeled electorate rake weight, multiplied by the final turnout probability and divided by the ex ante modeled turnout probability.

Finally, the six state samples were balanced to each represent one-sixth of the sum of the weights.

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 battleground sample of registered voters is 1.29. The design effect for the likely electorate is 1.46, which includes the added variance due to incorporating the probability that a respondent will participate in the 2024 election.