Support The Bulwark and subscribe today.
  Join Now

The Data Have Spoken: Abortion Was a Decisive Issue in the 2022 Midterms

Insights from the exit polls and AP’s VoteCast survey.
November 11, 2022
The Data Have Spoken: Abortion Was a Decisive Issue in the 2022 Midterms

On June 24, in Dobbs v. Jackson Women’s Health Organization, the Supreme Court overturned Roe v. Wade. The Court abolished the constitutional right to abortion, thereby empowering states to outlaw the procedure. Since then, many states have done just that.

On Tuesday, four and a half months after Dobbs, Republican candidates did surprisingly poorly in the midterms. The GOP lost crucial governorships, failed to capture the Senate, and—against the norm for parties out of power—made hardly any gains in the House.

Are these two events related? Did Dobbs torpedo the GOP?

It certainly did. Here’s the evidence from voter surveys and the election results.

The political environment going into the midterms favored Republicans in almost every way. President Biden’s approval ratings were underwater, Republicans were the out party, and in pre-election surveys, the GOP had the advantage on most issues. But on Tuesday, in state after state, pro-choice Democrats facing pro-life Republicans won races they had been expected to lose. The pro-choice side also won every statewide referendum on abortion, even in Montana and Kentucky.

These results don’t prove, by themselves, that abortion made the difference in elections for office. But two surveys of the electorate—the national exit poll sponsored by several TV networks, and VoteCast, a separate study overseen by the Associated Press—strongly support this conclusion.

In the network exit poll, 60 percent of the respondents said abortion should be legal in all or most circumstances. Twenty-seven percent of voters said abortion was the most important issue in casting their ballots. Democrats won both of these groups—the 60 percent and the 27 percent—by about three to one. Sixty percent of voters said they were angry or dissatisfied about Roe being overturned, and more than 70 percent of this angry-or-dissatisfied majority voted for Democrats.

The network exit poll couldn’t measure the increasing salience of abortion, because in the 2018 midterms, abortion was so low-profile that the exit poll didn’t even ask about it as a voting issue. But VoteCast did, and its new survey documents a colossal shift. In VoteCast’s poll of the 2018 electorate, nearly 80 percent of Republicans said abortion was a factor in how they voted; only about 20 percent of Democrats said the same. This year, those numbers turned upside down, giving Democrats a four-to-one advantage on the issue.

Wall Street Journal analysis of VoteCast data.

Like the exit poll, VoteCast found that about 60 percent of the electorate—63 percent, in the VoteCast sample—said abortion should be legal in all or most cases. But unlike the exit poll, it directly measured the effect of Dobbs. In the VoteCast survey, pro-choice voters (those who said abortion should be legal in all or most cases) were far more likely than pro-life voters (those who said abortion should be illegal in all or most cases) to say that the overturn of Roe had a “major impact” on which candidates they voted for. The gap was more than 20 points: 55 percent of pro-choicers said Dobbs was a major factor, compared to 32 percent of pro-lifers. When analyzed by party, the gap was more than 30 points: 65 percent of Democrats said Dobbs was a major factor, compared to 32 percent of Republicans.

In reading exit polls, it’s common to look at the subset of voters who said abortion was their “most important” issue. That can be misleading, because a lot of voters who care about abortion also care about inflation, crime, and other competing issues. It’s more useful to look at the “major impact” group, which logically encompasses the “most important” subset and is relatively unaffected by other issues the voter cares about. In the VoteCast survey, 47 percent of respondents said the overturn of Roe was a major factor in which candidates they voted for. And this half of the electorate chose Democratic House candidates over Republican House candidates by a margin of about 30 points. Among the smaller subset of voters who named the overturn of Roe as their most important issue, the gap was 43 points.

Kaiser Family Foundation analysis of VoteCast data.

But these numbers don’t tell the whole story. Dobbs didn’t just influence which candidates people voted for. It also influenced whether they showed up at the polls at all—and this provided a crucial boost to pro-choice candidates. In the VoteCast survey, pro-choice voters were twice as likely as pro-life voters (48 percent to 23 percent) to say Dobbs had a major impact on their “decision whether to vote” in the election. In partisan terms, the gap was even bigger: 57 percent of Democrats, compared to 23 percent of Republicans, said Dobbs had a major impact on their decision about whether to vote.

VoteCast data published by the Kaiser Family Foundation show significant turnout and candidate-selection effects in state after state. In Michigan, 68 percent of Democrats said the overturn of Roe had a major impact on which candidates they voted for. Only 40 percent of Republicans said the same. And Democrats were twice as likely as Republicans—59 percent compared to 30 percent—to say that the Supreme Court’s ruling had a major impact on their decision to vote at all.

In Arizona, Georgia, and Pennsylvania—in races for governor as well as for the U.S. Senate—people who voted for the Democratic nominees were about twice as likely (compared to those who voted for the Republican nominees) to say that Dobbs had a major impact on their choice of candidates. And they were more than twice as likely, compared to those who voted for Republicans, to say that the ruling had a major impact on their decision to cast a ballot in the midterms. On average, the partisan gap was about 35 points.

In a separate analysis of the VoteCast data, KFF found that in key Senate races, abortion drove a significant number of Republican voters to support the Democratic nominees. In Pennsylvania and Arizona, 12 percent of Republican and Republican-leaning voters said the overturn of Roe was the most important factor in their vote, and about 20 percent of these usually Republican voters cast their ballots for John Fetterman or Mark Kelly. In close elections, such defections—which subtract from the Republican column while adding to the Democratic column—can be decisive.

In Nevada, the state that might end up deciding control of the Senate, the Dobbs effect was enormous. In the VoteCast survey, people who voted for Democratic Sen. Catherine Cortez Masto were four times as likely as those who voted for Republican challenger Adam Laxalt—59 percent compared to 14 percent—to say that the Supreme Court’s ruling had a major impact on their decision about whether to vote. And they were more than three times as likely—67 percent to 20 percent—to say that the Court’s decision had a major impact on which candidate they voted for. That’s a 45-point gap on both questions. If Cortez Masto wins, Dobbs will almost certainly have cost Republicans the Senate.

None of this proves that the Court was wrong to overturn Roe or that Republicans shouldn’t continue to oppose abortion. You can make principled arguments for both positions. But politically, the result is clear. Most voters are pro-choice. They don’t like what the Court did. In an election they might otherwise have skipped, many of these people showed up at the polls to vote for Democrats. Others, who might have voted Republican, crossed over to block the GOP’s abortion agenda. And Republican candidates paid the price.

William Saletan

William Saletan is a writer at The Bulwark.