How These Polls Were Conducted

Here are the key things to know about this set of polls from The New York Times, The Philadelphia Inquirer and Siena College:

• Interviewers spoke with 808 voters in Arizona from Oct. 7 to 10, 656 voters in Montana from Oct. Oct. 5 to 8, and 857 voters in Pennsylvania from Oct. 7 to 10.

• Times/Siena polls are conducted by telephone, using live interviewers, in both English and Spanish. Overall, more than 95 percent of respondents were contacted on a cellphone for these polls.

• Voters are selected for the survey from a list of registered voters. The list contains information on the demographic characteristics of every registered voter, allowing us to make sure we reach the right number of voters of each party, race and region. For these polls, interviewers placed about 235,000 calls to nearly 90,000 voters.

• To further ensure that the results reflect the entire voting population, not just those willing to take a poll, we give more weight to respondents from demographic groups that are underrepresented among survey respondents, like people without a college degree. You can see more information about the characteristics of our respondents and the weighted sample at the bottom of the page, under “Composition of the Sample.”

• The margin of sampling error among likely voters is about plus or minus four percentage points. In theory, this means that the results should reflect the views of the overall population most of the time, though many other challenges create additional sources of error. When the difference between two values is computed — such as a candidate’s lead in a race — the margin of error is twice as large.

If you want to read more about how and why the Times/Siena Poll is conducted, you can see answers to frequently asked questions and submit your own questions here.

Full Methodology

The New York Times/Siena College polls of 656 voters in Montana and 808 voters in Arizona and the New York Times/Philadelphia Inquirer/Siena College poll of 857 voters in Pennsylvania were conducted in English and Spanish on cellular and landline telephones. The Arizona ran from Oct. 7 to 10, the Pennsylvania poll ran from Oct. 7 to 10, 2024, and the Montana poll ran from Oct. 5 to 8.

For each poll, the margin of sampling error among the likely electorate is plus or minus 4.3 percentage points in Montana, plus or minus 3.9 percentage points in Arizona and plus or minus 3.8 percentage points in Pennsylvania.

The Times/Siena polls of Pennsylvania in 2024 were conducted in partnership with the Philadelphia Inquirer and were funded in part by a grant from the Lenfest Institute for Journalism. The poll was designed and conducted independently from the institute.

Sample

The survey is a response-rate-adjusted stratified sample of registered voters taken from the voter file maintained by L2, a nonpartisan voter-file vendor, and supplemented with additional voter-file-matched cellular telephone numbers from Marketing Systems Group. 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.

To adjust for noncoverage bias, the L2 voter file for each state was stratified by statehouse district, party, race, gender, marital status, household size, turnout history, age and homeownership. 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, or with differing numbers from L2 and Marketing Systems Group, the number with the highest modeled response rate was selected.

Fielding

The sample was stratified according to political party, race and region. Marketing Systems Group screened the sample to ensure that the cellular telephone numbers were active, and the Siena College Research Institute fielded the poll, with additional fieldwork by ReconMR, the Public Opinion Research Laboratory at the University of North Florida, the Institute for Policy and Opinion Research at Roanoke College, the Center for Public Opinion and Policy Research at Winthrop University in South Carolina and the Survey Center at University of New Hampshire. Interviewers asked for the person named on the voter file and ended the interview if the intended respondent was not available. Overall, more than 95 percent of respondents were reached on a cellular telephone.

In Arizona and Pennsylvania, the questions were 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, 6 percent of interviews (9 percent of the weighted sample) among self-reported Hispanics were conducted in Spanish, including 2 percent of the interviews (3 percent of the weighted sample) among self-reported Hispanics in Arizona and 26 percent of the interviews (34 percent of the weighted sample) among self-reported Hispanics in Pennsylvania.

An interview was determined to be complete for the purposes of inclusion in the questions about whom the respondent would vote for if the respondent did not drop out of the survey after being asked the two self-reported variables used in weighting — age and education — and answered at least one of the questions about age, education or presidential-election candidate preference.

Weighting (registered voters)

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

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

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

The following targets were used:

• Party registration (L2 data) by whether the respondent has requested an absentee ballot for the 2024 general election (L2 data), in Pennsylvania

• Party registration (L2 data) by race (L2 model), in Arizona

• Six categories of partisanship (Classification based on an NYT model of vote choice in prior Times/Siena polls), in Montana

• Partisanship (L2 model based on commercial data and partisan political contributions), in Montana

• Race or ethnicity (L2 model)

• Age (self-reported age, or voter-file age if the respondent refused) by gender (L2 data)

• 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/nonwhite race by college or noncollege educational attainment (L2 model of race weighted to match NYT-based targets for self-reported education in Pennsylvania; L2 model of race weighted to match NYT-based targets derived from census data in Arizona)

• Marital status (L2 model)

• Homeownership (L2 model)

• Turnout history (NYT classifications based on L2 data)

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

• State region (NYT classifications)

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

• History of voting in the 2020 presidential primary (L2 data), in Pennsylvania

• Census tract educational attainment, in Arizona

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 voters leaning a certain way) 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. Four-fifths of the final probability that a registrant would vote in the 2024 election was based on the registrant’s ex ante modeled turnout score, and one-fifth was based on 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 1.24 for the likely electorate in Montana, 1.29 for the likely electorate in Pennsylvania and 1.30 for the likely electorate in Arizona.

Among registered voters, the margin of sampling error is plus or minus 4.3 points in Montana, including a design effect of 1.26; plus or minus 3.8 points in Arizona, including a design effect of 1.20; and plus or minus 3.7 points in Pennsylvania, including a design effect of 1.23.

For the sample of completed interviews, among the likely electorate, the margin of sampling error is plus or minus 4.5 points in Montana, including a design effect of 1.29; plus or minus 4 points in Pennsylvania, including a design effect of 1.35; and plus or minus 4.1 points in Arizona, including a design effect of 1.30.

Historically, The Times/Siena Poll’s error at the 95th percentile has been plus or minus 5.1 percentage points in surveys taken over the final three weeks before an election. Real-world error includes sources of error beyond sampling error, such as nonresponse bias, coverage error, late shifts among undecided voters and error in estimating the composition of the electorate.

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

Israeli strikes have killed more than 150 Palestinians in an operation focusing on the Jabalia refugee camp this week, with thousands more trapped in the ruins of the settlement in northern Gaza after a year of war with Hamas.

The camp has been the scene of several pitched battles between Israeli forces and Hamas, as the militant group attempts to regroup in areas from which the Israel Defense Forces had retreated.

This week’s offensive started with the encirclement of Jabalia, leaving a single street for its civilians to exit from. The IDF said it had “eliminated” at least 50 Hamas fighters this week, including several it said had taken part in the October 7 cross-border raid that sparked the conflict. Local health officials said at least 150 people had been killed in and around Jabalia in the past week.

“It’s more than scary — the situation is critical,” said Mustafa, who managed to escape from Jabalia before the offensive began. “It seems that the Jabalia camp will be deleted from Gaza’s geography.”

Israeli Prime Minister Benjamin Netanyahu has refused to endorse a political solution for Gaza, leaving the military fighting an intermittent insurgency as international aid groups struggle to provide a population of nearly 2.3mn civilians with enough food, medicine and shelter to survive.

Some 300,000 civilians, the UN and others said, are living in the ruins of their neighbourhoods and homes in northern Gaza, separated from the rest of the population by an Israeli military corridor that divides the besieged enclave into two sectors.

In the past few days, the IDF has demanded that thousands of civilians evacuate the northern sector and run a gauntlet of checkpoints to reach al-Mawasi, a fetid and crowded “humanitarian zone” alongside the Mediterranean. Many are too scared by the violence outside their homes to flee.

Ibrahim al-Kharabishy, a lawyer and the father of four children including a baby, said they constantly hear explosions from artillery and warplanes. His family is safer indoors, rather than out on the streets, where Palestinian looters add to the risk from the military’s operation.

“[The army] called us this morning and ordered us to evacuate, but we are staying at home because it is the only refuge we have left,” he told the Financial Times over the phone. “I am not being obstinate with the army, but we are unable to go. We need a safe place to go to.”

All they have at home is flour, and he said Israel was using hunger to “empty out the north.” Israel has denied the accusation repeatedly.

Food has all but run out in northern Gaza, the World Food Programme said, since the main crossings closed on October 1.

“WFP distributed its last remaining food stocks in northern Gaza to partners and kitchens sheltering newly displaced families — but these are barely enough to last two weeks,” WFP said.

At least 42,000 Palestinians, local health officials estimate, have been killed since the war began on October 7 when Hamas killed 1,200 people within Israel, the Israeli government said, and took 250 or so hostage. More than 100 hostages are still being held.

The renewed fighting in the Jabalia camp has been overshadowed by Israel’s offensive against Hizbollah in Lebanon, where nearly a quarter of the country’s territory is under an evacuation order from the Israeli military, the UN’s High Commissioner for Human Rights estimates.

On Saturday, Israel warned 20 more Lebanese villages that they too could face harm as the IDF’s ground invasion spread.

According to an FT count, since it began its ground invasion to combat Hizbollah on October 1, Israel has warned about 140 communities in south Lebanon to flee their homes. The IDF has commanded residents to move north of the Awali river, which runs at least 80km north of the southern tip of Lebanon.

Mohammad Bagher Ghalibaf, Iran’s parliamentary speaker, arrived in Beirut on Saturday as fighting flared across the southern Lebanese-northern Israeli border. Hizbollah said it had targeted an explosives factory south of Haifa as well as an Israeli bulldozer in the south Lebanese village of Ramia.

Two drones from Lebanon made it as far south as Herzliya, a prosperous Tel Aviv suburb, before one was shot down and the other hit a nursing home. No casualties were reported.

A soldier in the UN international peacekeeping mission Unifil, which patrols the Lebanese-Israeli border, was shot and hospitalised on Friday night. Unifil said that the peacekeeper had been hit by gunfire that came from fighting near its headquarters in Naqoura, southern Lebanon. The UN force said it did not know which side fired the bullet. 

Hours earlier, two Unifil peacekeepers were hurt by unattributed explosions near an observation tower. Israel’s military said it was looking into the incidents, and accused Hizbollah of operating near Unifil positions. 

LOS ANGELES (AP) — With the presidency on the line in battlegrounds like Wisconsin and Pennsylvania, why would Donald Trump venture into California, one of the most solidly Democratic states, just weeks before Election Day?

Trump is almost certain to lose California, and that won’t change after his scheduled Saturday stop in Coachella, a desert city east of Los Angeles best known for the annual music festival bearing its name. Still, there are practical reasons for him to visit, despite the Republican nominee’s prospects Nov. 5 in the most populous state.

The former president lost California in a landslide in 2020. He did get 6 million-plus votes, more than any GOP presidential candidate before, and his margins topped 70% in some rural counties that typically favor conservatives on the ballot.

That’s an enormous pool of potential volunteers to work on state races and participate in phone banks into the most contested states. And Trump is likely to draw extensive media coverage in the Los Angeles market, the second-largest in the country.

Trump is visiting Coachella in between stops in Nevada, at a roundtable outside Las Vegas for Latinos earlier Saturday, and Arizona, for a rally Sunday in Prescott Valley. He narrowly lost those two swing states to Democrat Joe Biden in 2020.

Going to California gives Trump the “ability to swoop in and leverage this big population of Trump supporters,” said Tim Lineberger, who was communications director for Trump’s 2016 campaign in Michigan and also worked in the former president’s administration. He’s “coming here and activating that.”

Lineberger recalled Californians making calls to Michigan voters in 2016 on Trump’s behalf and said the campaign’s decision to go into safe, Democratic turf at this point was “an aggressive, offensive play.”

California is also a fountain of campaign cash for both parties, and Trump will be fundraising. Photos with the former president in Coachella were priced at $25,000, which comes with special seating for two. A “VIP Experience” was priced at $5,000.

With congressional races in California in play that could determine which party controls the House, the Coachella rally “is a get-out-the-vote type of thing that motivates and energizes Republicans in California, when they are not as close to what is going on in the national campaign,” Republican consultant Tim Rosales said.

Rosales also said to look for Trump to continue his long-running spat with Democratic Gov. Gavin Newsom.

For Republicans, “It’s motivating when you can pick at California a little bit and the governor … will take the bait,” Rosales said.

Newsom on Wednesday predicted Trump would be denigrating the state at the rally, overlooking its strengths as the world’s fifth-largest economy. The governor said that for the first time in a decade, California has more Fortune 500 companies than any other state.

“You know, that’s not what Trump is going to say,” he predicted.

Jim Brulte, a former chairman of the California Republican Party, said he thinks Trump is angling for something that has eluded him in previous campaigns: winning more total votes than his Democratic opponent.

“I believe Donald Trump is coming to California because he wants to win not only in the Electoral College, but he wants to win the popular vote. There are more registered voters in California than there are residents in 46 of the other 49 states,” Brulte said.

The Trump National Golf Club Los Angeles sits on the Pacific Coast, south of the city. But Trump has long had a conflicted relationship with California, where a Republican has not carried the state since 1988 and Democrats outnumber registered Republicans by about 2-to-1.

California was home to the so-called Trump resistance during his time in office, and Trump often depicts California as representing all he sees wrong in America. As president, he called the homeless crises in Los Angeles and San Francisco disgraceful and threatened to intercede.

He is likely to spend time on Saturday linking California’s problems to Vice President Kamala Harris, the Democratic nominee and a San Francisco Bay Area native who was California’s attorney general and represented the state in the Senate.

His campaign issued a statement alleging that under Harris, “the notorious ‘California Dream’ has turned into a nightmare for everyday Americans.”

Jessica Millan Patterson, chairwoman of the state GOP, said she looked forward to hearing Trump contrast his agenda with a Democratic White House that “has left Californians less safe and with less money in their pockets.”

Republicans, she promised, “will do our part to secure a House majority.”