Polls do not measure public opinion, because public opinion is not an objective reality. It depends on how you ask the questions.
Steve Chapman has an excellent column that illustrates this idea using the abortion issue:
But the decision in Dobbs v. Jackson Women’s Health Authority shattered that complacency. Many if not most people were shocked to see a basic liberty go up in smoke. A Pew poll taken days after the ruling found 57% of Americans disapproved of it, while 43% strongly disapproved. More Democrats strongly disapproved of the ruling than Republicans strongly approved of it. Support for legal abortion has registered a significant and lasting increase. Gallup found that 69% of Americans now believe the procedure should be legal through the first three months of pregnancy, up from 60% in 2018. It also found that 37% think it should be allowed in the second three months, up from 28% in 2018.
So which is it? If 57% oppose the Supreme Court decision, then presumably at least 57% favor legal abortion. (Maybe more, I don’t oppose the Dobbs decision but am also pro-choice.) And yet we are told that only 37% favor legal abortion in the second trimester.
Chapman points out that the behavior of voters does not seem to reflect these polls:
You might figure that if 63% of Americans oppose allowing second-trimester abortions, the 15-week cutoff would be an easy sell. But the court’s reversal has engendered great suspicion of measures that fall short of restoring the old status quo. When Virginians heard Youngkin propose a ban after 15 weeks, they apparently didn’t focus on “15 weeks”; they focused on “ban.”
If the polls were accurate, then abortion should be a winning issue for less extreme Republicans like Youngkin. It isn’t. Once voters are faced with the reality of the issue, opinions change rapidly:
It’s not surprising that support has grown for permitting second-trimester abortions. The post-Dobbs experience has given Americans a new understanding of the dangers of curbing access, thanks to heartbreaking stories about women whose fetuses suffer from severe and even fatal defects but who are forced to carry them to term. Even women with complications that put their own lives at risk have found doctors unwilling to perform abortions for them. After passing its strict ban, Texas saw an 11.5% increase in infant mortality.
Many people don’t have a settled view of abortion—it depends how you ask the question. And this is true for many other issues. Consider these two poll questions:
1. Should prostitution be legal?
2. Should prostitutes be put in prison?
Will the responses be consistent?
(Yes, a person might say “Keep it illegal but have them pay a fine.” But that’s not really any different from “Make it legal and put a tax on the activity.”)