A new Pew Research Center survey on religious liberty found that the public is split on the question of whether businesses that provide wedding services should be able to refuse same-sex couples if the owner has religious objections: 48% are in agreement and 49% are not. Catholics believe, by a margin of 54% to 43%, that businesses should be required to provide services to gay couples.
Other surveys provide a different outcome. Does this mean the Pew survey is flawed? No. It means that the wording of the question strongly influences the respondent’s answer. What also matters is whether self-identification is an accurate measure of reality.
For example, last December an AP and NORC Center for Public Affairs Research survey found that 82% of Americans said religious liberty protections were important to Christians. Similarly, 8 in 10 Americans said that it was very or extremely important for people like themselves to be allowed to practice their religion freely.
So when Americans are asked about their support for religious liberty in general—when there is no competing value they are asked to weigh—their commitment shines through. But in the real world, there is usually a conflict between rights.
Last fall, the Catholic League commissioned Kellyanne Conway of The Polling Company to survey Catholics on a range of issues, one of them being religious liberty. Catholics were asked, “Do you agree or disagree that private businesses with religious objections should be forced to provide services that violate their beliefs?” By a margin of 63% to 30%, they opposed compelling private businesses to provide services that violate their religious beliefs.
What about when the question is narrowed to wedding-related businesses? It makes no difference: 62% say it is mostly unfair and 29% say it is fair.
How can these differences be explained? Americans prize religious liberty but they also support equal treatment. When these values conflict, much depends on whether the respondent is being asked to defend government coercion or support equal treatment: the former is not popular, but the latter is.
In the Catholic League survey, we disaggregated on the basis of several criteria, among them being ideology. For example, when respondents were asked about whether businesses should be required to provide health coverage that violates their religious beliefs, Catholics in general took the side of the owner. There was one segment that favored coercion: they were identified as being the most liberal Catholics in the sample.
We know from other surveys that the most liberal Catholics are also the most likely to be non-practicing. Yet the Pew survey treated them as equals—no attempt was made to distinguish them from others. Therefore, it is likely that if non-practicing Catholics were factored out of the Pew survey, Catholics would appear less liberal on this issue.
Should non-practicing Cath-olics be included in samples of the Catholic population? Should Americans who identify themselves as vegetarian be included in a survey of vegetarians even if they occasionally eat steaks or hot dogs? It depends on whether self-identification is seen as a satisfactory measure of reality. To put it another way, if a man has male genitals, is he a woman because he says he is?
Postmodern sages who think truth is a myth are entitled to live in their world of make believe, but they are not entitled to our respect. Reality may be interpreted differently, but not all interpretations are equally valid.
