Methodology

The New York Times/Siena College poll of 980 registered voters nationwide, including 823 who completed the full survey, was conducted in English and Spanish on cellular and landline telephones from Feb. 25 to 28, 2024. The margin of sampling error is plus or minus 3.5 percentage points for registered voters and plus or minus 3.8 percentage points for the likely electorate. Among those who completed the full survey, the margin of sampling error is plus or minus 4 percentage points for registered voters and plus or minus 4.2 percentage points for the likely electorate.

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.

First, records were selected by state. 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 were calculated for each stratum. The mean expected response rate was based on a model of unit nonresponse in prior Times/Siena surveys. 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.

Second, state records were selected for the national sample. The number of records selected by state was based on a model of unit nonresponse in prior Times/Siena national surveys as a function of state (as a random effect), telephone number quality and other demographic and political characteristics. The state’s share of records was equal to the reciprocal of the mean response rate of the state’s records, divided by the national sum of the weights. In 12 states, the models of response were inadvertently fit using data that did not suppose that “dropoff” respondents – those who completed the questions used for weighting, but subsequently dropped out of the interview before answering the last question – would be counted as completed interviews for this survey. This was corrected with a post-hoc selection weight, described below in the weighting section.

Fielding

The sample was stratified according to political 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, 94 percent of respondents were reached on a cellular telephone.

The instrument was translated into Spanish by ReconMR. 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, 3 percent of interviews among self-reported Hispanics were conducted in Spanish, including 4 percent of weighted interviews.

An interview was determined to be complete for the purposes of inclusion in the ballot test question if the respondent did not drop out of the survey by the end of the two self-reported variables used in weighting — age and education — and answered at least one of the age, education or presidential election ballot test questions.

Weighting — registered voters

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, interacted by whether a respondent was from one of twelve states in which the sample was inadvertently drawn using response models that had been fit supposing that respondents who dropped off would not be counted as completed interviews.

Second, the sample was weighted to match voter file-based parameters for the characteristics of registered voters.

The following targets were used:

• Party (party registration if available, or else classification based on a model of vote choice in prior Times/Siena polls) by whether the respondent’s race is modeled as white or nonwhite (L2 model)

• Age (Self-reported age, or voter file age if the respondent refuses) by gender (L2)

• Race or ethnicity (L2 model)

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

• White/non-white race by college or non-college educational attainment (L2 model of race weighted to match NYT-based targets for self-reported education)

• Marital status (L2 model)

• Home ownership (L2 model)

• National region (NYT classifications by state) by whether a state was among the 12 states drawn inadvertently supposing that only completed interviews would be classified as complete (See unequal selection weight for additional details).

• Turnout history (NYT classifications based on L2 data)

• Method of voting in the 2020 elections (NYT classifications based on L2 data)

• Metropolitan status (2013 NCHS Urban–Rural Classification Scheme for Counties)

• Census tract educational attainment (NYT classifications based on American Community Survey data)

Finally, the sample of respondents who completed all questions in the survey was weighted identically, as well as to the result for the general election horse race question (including leaners) on the full sample.

Weighting — likely electorate

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

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 sample was weighted 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 intention to 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 intentions, 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 sample of respondents who completed all questions in the survey was weighted identically, as well as to the result for the general election horse race question (including leaners) on the full sample.

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 is 1.28 for registered voters and 1.45 for the likely electorate. The design effect for the sample of completed interviews is 1.35 for registered voters and 1.53 for the likely electorate.

Historically, The Times/Siena Poll’s error at the 95th percentile has been plus or minus 5.1 percentage points. Real-world error includes sources of error beyond sampling error, such as nonresponse bias and coverage error.