Punditry and sensationalist headlines

 A recent op-ed in the New York Times by Cecile Richards (former president of Planned Parenthood) says "Despite years of punditry and sensationalist headlines, the myth that Americans will always be divided on this seemingly intractable issue [abortion] is just that: a myth." Of course, no one knows how things will "always" be, but the standard view is that Americans have been pretty evenly divided on the issue for the last fifty years and that there are no signs of change, so the best guess is that things will stay that way.  According to Richards, the reality is that there's a solid majority in favor of legal abortion:  "According to Gallup, 80 percent of the country believes abortion should be legal under any or certain circumstances. And last year, American Bridge 21st Century, in partnership with Planned Parenthood Action Fund and Emily’s List, found that a majority of the voters polled (71 percent of women and 64 percent of men) felt Republicans were 'out of step with their own views' on abortion."  The first statistic clearly doesn't support her case, since "certain circumstances" can be very restrictive (for example, only if necessary to save the woman's life).  The second one sounded more impressive, so I followed the link.  It turns out that the survey involved "base and ambivalent voters on abortion; the samples excluded registered voters who are strong Republicans or are staunchly anti-abortion."  Although her statement is technically correct, since she said "the voters polled," it also doesn't support her case.  So the punditry and sensationalist headlines give a pretty accurate summary of public opinion on abortion. 

Although the survey excluded an important part of the public, it did raise some interesting questions.  The purpose was to see how supporters of legal abortion could appeal to persuadable voters.  One recommendation was to "remind voters what is at stake in the impending Supreme Court decision . . . including the potential for restrictive laws in their own state."  That led me to look for surveys asking people what they expected to happen if Roe v. Wade was overturned.  A 2019 Kaiser Family Foundation poll asked "If Roe v. Wade were overturned, it would be up to each state to decide if abortion would be legal and in many states abortion would immediately become illegal. Do you think that abortion would continue to be legal in your state if Roe v. Wade were overturned, or would it no longer be legal, or do you not know?"  34% said it would continue to be legal, 29% that it would become illegal, and 37% said that they didn't know, but there were substantial differences among the states--for example, in California, 64% expected it to remain legal and 12% thought it would become illegal, while in Texas 12% expected it to remain legal and 55% expected it to become illegal.  These examples suggest that expectations are related to state laws, or the climate of opinion in the state, or both.  

The Pew Research Center compiled information about five kinds of laws:  bans on abortion in the first trimester (which would go into effect if Roe v. Wade is overturned), requiring doctors who perform abortions to have hospital admitting privileges, mandated counseling, mandated ultrasound, and a waiting period after counseling or ultrasound.  They also showed state level figures on the percent who said that abortion should always or usually be illegal.  I computed an index of general restrictiveness based on those laws--the most restrictive was Louisiana, and eighteen states had none of them and tied as being least restrictive.  The correlation between the percent saying that abortion should  always or usually be illegal was .78.  On the state level, both existing laws and opinions about abortion predicted expectations--that is, people in states where opinions were less favorable were more likely to expect abortion to become illegal, even controlling for current state-level laws.  That leads to a question of whether people's expectations are based on their own wishes--expecting things to turn out the way you want them to--or on sensing the opinions of other people--either one could account for the relationship.  On the individual level, opinions about whether abortion should be legal have no discernible effect on expectations after you control for state.*  That is, it seems that people pick up the local climate of opinion rather than projecting their own wishes.  Of course, sensing the climate of opinion isn't that hard, because opinions on abortion are strongly related to partisanship--people just have to know if they live in a Democratic, Republican, or swing state.  However, there are some plausible reasons to expect that one's own opinions might matter--there are the tendencies to think that you are on the "right side of history," and to have a distorted impression of public opinion as a result of living in "bubbles" of like-minded people.  On the other hand, activists often motivate their side by pointing to worst-case scenarios--in this case, supporters of abortion rights saying that overturning Roe v. Wade would take us right back to the 1950s.  So the absence of a discernible effect for one's own opinions is interesting. 

*I also looked at a number of other factors--gender, ethnicity, religion, education, and party identification--and there was no sign that any made any difference except maybe religion (Catholics, evangelical Christians, and "other Christians") more likely to expect it to become illegal compared to mainline Protestants, non-religious, and non-Christian religions.  

[Date from the Roper Center for Public Opinion Research]