Pushing and polling on abortion

Marjorie Connelly first got into polling as a college student when she worked as a phone interviewer asking questions of survey respondents. After more than 30 years working in the business at The New York Times and more than six years at the University of Chicago’s National Opinion Research Center (NORC), she’s no longer the one on the phone. Connelly now manages research projects at NORC, formulating questions and sorting through and analyzing data when it comes back from the team doing the fieldwork.

Crafting a well-made survey requires precision. “You want to make sure that the question is balanced,” Connelly said. “You want to give both sides, so that’s why it’s always like, ‘Do you approve or disapprove of the president?’” She also pays attention to the order of questions. Earlier questions can affect later answers in a survey, which is why pollsters start with general questions before getting specific.

She and her team took that approach when writing questions for a recent AP-NORC poll on abortion. One portion of the poll asked 1,125 respondents to share their views on whether abortion should be legal or illegal in all or most cases during each of the three trimesters of pregnancy. The results, released last month, conflicted with other recent polling on similar issues but affirmed what pro-lifers have long observed about the views of Americans, revealing a nuanced public perspective on the politically polarizing topic.

“It’s not a black-and-white kind of issue,” Connelly said. “We ask the question on timing, and then you see there’s large differences even among people who think it should be legal.” According to the results, 61 percent of respondents approved of legal abortion in the first trimester of pregnancy. Meanwhile, only 34 percent said abortion should be legal in all or most cases in the second trimester, which begins in the 13th week of pregnancy. That percentage trickled down to 19 percent in the third trimester, weeks 29 to 40.

“It produced results that most people are not aware of: There has never been majority support for what Roe v. Wade and Doe v. Bolton actually did,” said David O’Steen, executive director of the National Right to Life Committee. The pro-life movement has known for decades that a majority of Americans do not support abortion on demand for any reason—especially not late in the pregnancy. But he said other polls show strong support for upholding Roe and its companion case, Doe, because people don’t understand the reach of the abortion rights it established.

Gallup released data from similar questions last month but with different results. It showed 56 percent of respondents said they would oppose “a ban on abortions after the 18th week of a pregnancy,” which falls within the second trimester. In contrast, 66 percent of respondents in the AP poll said abortion should be illegal in all or most cases during the second trimester, and 81 percent thought it should illegal in the third trimester. Those results also differed from a similar Gallup poll from 2018, as March for Life president Jeanne Mancini pointed out last week. According to the 2018 results, only 28 percent of respondents said abortion should be generally legal in the second three months of pregnancy and only 13 percent said it should be generally legal in the last trimester.

How can the same public that doesn’t support “second trimester” abortions also oppose “18-week” bans? Lydia Saad, director of U.S. social research at Gallup, said the differences come down to wording. She said the language of a “ban” could be “more onerous to people,” eliciting stronger responses, whereas wording like “generally legal” or “generally illegal” is a little less provocative. On top of that, Saad added, “We’re not saying ‘trimester.’ We’re saying 18th week. And I don’t know if people know what trimester that refers to.”

That doesn’t mean one set of survey results are invalid, though. “It’s just telling you that people are very sensitive depending on the framework and what are the factors that trigger those feelings,” Saad said.

That’s why O’Steen thinks sometimes survey questions should include more background information.

“You want to find out how people feel relative to what they’re going to hear from the press or the other side,” he said. “But you also want to find out how they’re going to feel if the question is framed absolutely accurately.” Polls initiated by National Right to Life sometimes include starker questions that make people think. Survey results released by Knights of Columbus in 2019 showed 59 percent of respondents supported a “ban on abortion after 20 weeks, except to save the life of the mother.” But O’Steen said a later poll conducted on behalf of National Right to Life in November 2020 found that 71 percent of respondents strongly or somewhat supported such laws when the question pointed out that unborn babies can feel pain at about 20 weeks of gestation.

“We try to stick very closely to how is the public hearing about this issue from leaders, from the news media,” Saad said. “We’re trying to understand, how do Americans feel just sitting out there in their homes on this issue without being given any additional information than what they have or they’re hearing on the news?”

But to O’Steen, that’s just the problem: Many Americans don’t know crucial information that could help them see the problems with the pro-abortion position, and they won’t get it from mainstream culture. In its next term, the Supreme Court will hear a case about a law that protects babies from abortion after 15 weeks of gestation. O’Steen pointed out the results from the AP-NORC poll “would indicate majority support for that Mississippi bill that’s being considered.”