Midterm Election Poll: Michigan’s 11th District, Epstein vs. Stevens (original) (raw)
Haley Stevens, the Democratic candidate, has a modest lead in our poll.
Our poll is a good result for Democrats. But remember: It’s just one poll, and we talked to only 465 people. Each candidate’s total could easily be five points different if we polled everyone in the district. And having a small sample is only one possible source of error.
Where we called:
Each dot shows one of the 66770 calls we made.
Vote choice: Dem. Rep. Don’t know Didn’t answer
To preserve privacy, exact addresses have been concealed. The locations shown here are approximate.
Explore the 2016 election in detail with this interactive map.
About the race
- Haley Stevens is a former chief of staff for President Obama’s auto task force. 27% favorable rating; 13% unfavorable; 60% don’t know
Based on 465 interviews - Lena Epstein is a business owner who helped run Mr. Trump’s 2016 Michigan campaign. 22% favorable rating; 22% unfavorable; 56% don’t know
Based on 465 interviews - The retirement of Dave Trott, a Republican, leaves a seat open in this Republican-leaning, largely suburban district. The two survivors of their crowded primary contests are both women and both political newcomers.
- Ms. Stevens managed a Treasury Department task force in 2009 coordinating the federal bailout of two of Michigan’s automakers. She is campaigning on a $15 federal minimum wage and the need for a public option in Obamacare.
- Ms. Epstein’s professional life is also connected to cars: She works in her family's business, Vesco Oil, a distributor of automotive and industrial lubricants. She is emphasizing her business acumen and advocating lower taxes and less burdensome regulations, and she has tied her political fortunes to those of Mr. Trump.
Other organizations’ ratings:
Previous election results:
| 2016 President | +4 Trump |
|---|---|
| 2012 President | +5 Romney |
| 2016 House | +13 Rep. |
How our poll result changed
As we reach more people, our poll will become more stable and the margin of sampling error will shrink. The changes in the timeline below reflect that sampling error, not real changes in the race.
One reason we’re doing these surveys live is so you can see the uncertainty for yourself.
But sampling error is not the only type of error in a poll.
Our turnout model
There’s a big question on top of the standard margin of error in a poll: Who is going to vote? It’s a particularly challenging question this year, since special elections have shown Democrats voting in large numbers.
To estimate the likely electorate, we combine what people say about how likely they are to vote with information about how often they have voted in the past. In previous races, this approach has been more accurate than simply taking people at their word. But there are many other ways to do it.
Our poll under different turnout scenarios
| Who will vote? | Est. turnout | Our poll result |
|---|---|---|
| The types of people who voted in 2014 | 246k | Stevens +2 |
| People who say they are almost certain to vote, and no one else | 269k | Stevens +9 |
| Our estimate | 312k | Stevens +8 |
| People whose voting history suggests they will vote, regardless of what they say | 317k | Stevens +7 |
| People who say they will vote, adjusted for past levels of truthfulness | 332k | Stevens +10 |
| The types of people who voted in 2016 | 372k | Stevens +6 |
| Every active registered voter | 495k | Stevens +11 |
All estimates based on 465 interviews
In these scenarios, higher turnout tends to be better for Democrats.
Just because one candidate leads in all of these different turnout scenarios doesn’t mean much by itself. They don’t represent the full range of possible turnout scenarios, let alone the full range of possible election results.
The types of people we reached
Even if we got turnout exactly right, the margin of error wouldn’t capture all of the error in a poll. The simplest version assumes we have a perfect random sample of the voting population. We do not.
People who respond to surveys are almost always too old, too white, too educated and too politically engaged to accurately represent everyone.
How successful we were in reaching different kinds of voters
| Called | Inter-viewed | Successrate | Ourresponses | Goal | |
|---|---|---|---|---|---|
| 18 to 29 | 6695 | 46 | 1 in 146 | 10% | 9% |
| 30 to 64 | 25900 | 298 | 1 in 87 | 64% | 62% |
| 65 and older | 8256 | 121 | 1 in 68 | 26% | 29% |
| Male | 19139 | 219 | 1 in 87 | 47% | 47% |
| Female | 21725 | 246 | 1 in 88 | 53% | 53% |
| White | 29937 | 352 | 1 in 85 | 76% | 75% |
| Nonwhite | 6354 | 65 | 1 in 98 | 14% | 14% |
| Cell | 26116 | 286 | 1 in 91 | 62% | — |
| Landline | 14748 | 179 | 1 in 82 | 38% | — |
Based on administrative records. Some characteristics are missing or incorrect. Many voters are called multiple times.
Pollsters compensate by giving more weight to respondents from under-represented groups.
Here, we’re weighting by age, primary vote, gender, likelihood of voting, race, education and region, mainly using data from voting records files compiled by L2, a nonpartisan voter file vendor.
But weighting works only if you weight by the right categories and you know what the composition of the electorate will be. In 2016, many pollsters didn’t weight by education and overestimated Hillary Clinton’s standing as a result.
Here are other common ways to weight a poll:
Our poll under different weighting schemes
| Our poll result | |
|---|---|
| Weight using census data instead of voting records, like most public polls | Stevens +9 |
| Don’t weight by education, like many polls in 2016 | Stevens +8 |
| Don’t weight by primary vote, like most public polls | Stevens +8 |
| Our estimate | Stevens +8 |
All estimates based on 465 interviews
Just because one candidate leads in all of these different weighting scenarios doesn’t mean much by itself. They don’t represent the full range of possible weighting scenarios, let alone the full range of possible election results.
Undecided voters
About 17 percent of voters said that they were undecided or refused to tell us whom they would vote for. On questions about issues, these voters most closely resembled Democrats.
But if they were to break 5 to 2 in favor of Republicans, that alone would be enough to change the lead in our poll, assuming we did everything else perfectly. (We could also be wrong on turnout or our sample could be unrepresentative. Or other voters could change their minds.)
Issues and other questions
We're asking voters whether they support Brett Kavanaugh's nomination to the Supreme Court and whether they believe the sexual assault accusations against him.
We're also asking voters about health care.
Do you approve or disapprove of the job Donald Trump is doing as president?
| Approve | Disapp. | Don’t know | |
|---|---|---|---|
| Voters n = 465 | 42% | 52% | 6% |
Would you prefer Republicans to retain control of the House of Representatives or would you prefer Democrats to take control?
| Reps. keep House | Dems. take House | Don’t know | |
|---|---|---|---|
| Voters n = 465 | 41% | 50% | 9% |
Do you support or oppose Brett Kavanaugh’s nomination to the United States Supreme Court?
| support | oppose | Don’t know | |
|---|---|---|---|
| Voters n = 465 | 43% | 48% | 9% |
As you may know, Supreme Court nominee Brett Kavanaugh has been accused of committing sexual assault when he was a teenager. Would you say you believe the allegations, you do not believe the allegations, or you simply are unable to come to a conclusion?
| Believe | Do not believe | Don’t know | |
|---|---|---|---|
| Voters n = 465 | 33% | 30% | 37% |
Do you support electing more people who describe themselves as feminists?
| support | oppose | Don’t know | |
|---|---|---|---|
| Voters n = 465 | 49% | 35% | 16% |
Is it important to elect more women to public office?
| agree | disagree | Don’t know | |
|---|---|---|---|
| Voters n = 465 | 73% | 16% | 11% |
As you think about your member of Congress, would you prefer your representative to support President Trump and his agenda, or to serve as a check on the president and his agenda?
| Support | Check | Don’t know | |
|---|---|---|---|
| Voters n = 465 | 39% | 52% | 9% |
Percentages are weighted to resemble likely voters.
What different types of voters said
Voters nationwide are deeply divided along demographic lines. Our poll suggests divisions too. But don’t overinterpret these tables. Results among subgroups may not be representative or reliable. Be especially careful with groups with fewer than 100 respondents, shown here in stripes.
Gender
| Dem. | Rep. | Und. | |
|---|---|---|---|
| Female n = 246 / 53% of voters | 50% | 32% | 18% |
| Male 219 / 47% | 40% | 44% | 16% |
Age
| Dem. | Rep. | Und. | |
|---|---|---|---|
| 18 to 29 n = 44 / 9% of voters | 61% | 20% | 19% |
| 30 to 44 98 / 18% | 57% | 28% | 15% |
| 45 to 64 202 / 43% | 41% | 43% | 16% |
| 65 and older 121 / 30% | 39% | 41% | 20% |
Race
| Dem. | Rep. | Und. | |
|---|---|---|---|
| White n = 377 / 83% of voters | 43% | 40% | 16% |
| Black 15 / 4% | 91% | — | 9% |
| Hispanic 11 / 2% | 34% | 60% | 5% |
| Asian 26 / 4% | 71% | 18% | 11% |
| Other 17 / 3% | 27% | 25% | 48% |
Race and education
| Dem. | Rep. | Und. | |
|---|---|---|---|
| Nonwhite n = 69 / 13% of voters | 62% | 21% | 17% |
| White, college grad 221 / 40% | 43% | 42% | 14% |
| White, not college grad 156 / 43% | 43% | 38% | 19% |
Education
| Dem. | Rep. | Und. | |
|---|---|---|---|
| H.S. Grad. or Less n = 35 / 16% of voters | 49% | 34% | 16% |
| Some College Educ. 131 / 31% | 41% | 38% | 20% |
| 4-year College Grad. 133 / 29% | 43% | 44% | 13% |
| Post-grad. 160 / 23% | 50% | 31% | 19% |
Party
| Dem. | Rep. | Und. | |
|---|---|---|---|
| Democrat n = 139 / 30% of voters | 90% | 2% | 7% |
| Republican 119 / 28% | 9% | 82% | 9% |
| Independent 183 / 37% | 39% | 34% | 27% |
| Another party 10 / 1% | 26% | 46% | 28% |
Primary vote
| Dem. | Rep. | Und. | |
|---|---|---|---|
| Democratic n = 126 / 27% of voters | 78% | 8% | 14% |
| Republican 178 / 40% | 21% | 65% | 14% |
| Other 161 / 34% | 48% | 28% | 24% |
Intention of voting
| Dem. | Rep. | Und. | |
|---|---|---|---|
| Almost certain n = 299 / 67% of voters | 48% | 39% | 13% |
| Very likely 118 / 26% | 45% | 33% | 22% |
| Somewhat likely 19 / 4% | 36% | 42% | 22% |
| Not very likely 14 / 2% | 28% | 30% | 42% |
| Not at all likely 10 / 1% | 6% | 31% | 63% |
Percentages are weighted to resemble likely voters; the number of respondents in each subgroup is unweighted. Undecided voters includes those who refused to answer.
Other districts where we’ve completed polls
About this poll
- Most responses shown here are delayed about 30 minutes. Some are delayed longer for technical reasons.
- The design effect of this poll is 1.24. That’s a measure of how much weighting we are doing to make our respondents resemble all voters.
- Read more about the methodology for this poll.
- Download the microdata behind this poll.
This survey was conducted by The New York Times Upshot and Siena College.
Data collection by Reconnaissance Market Research, M. Davis and Company, the Institute for Policy and Opinion Research at Roanoke College, the Survey Research Center at the University of Waterloo, the University of North Florida and the Siena College Research Institute.
By Michael Andre, Larry Buchanan, Matthew Bloch, Jeremy Bowers, Nate Cohn, Alastair Coote, Annie Daniel, Richard Harris, Josh Katz, Rebecca Lieberman, Blacki Migliozzi, Paul Murray, Adam Pearce, Kevin Quealy, Eden Weingart and Isaac White