It’s almost always correct that the devil is in the details, but it’s difficult to convey these details in short articles. Making things worse, headline writers have a tendency toward clickbait.
Both tendencies are evident in a provocatively titled article, ” The public doesn’t support restrictive voter ID laws, but many new ones will be in force in 2016,” by Herman Schwartz in the Reuter’s Public Opinion Blog.
The title, read in isolation, is wrong, or at best badly misleading. The public strongly supports a requirement to show a photo ID prior to voting. Support is strong whether the question is asked in a very generic fashion or when specific kinds of photo IDs are described. Schwartz looked at 37 polls for his commentary, including one that I collaborated on with scholars at MIT and Appalachian State University. We report detailed question wordings for 19 polls are reported in the Appendix to our paper.
Our data (shown below) indicates that support is uniformly strong among Republicans, while Democratic support varies by ideology, level of education, and attention to politics. Nonetheless, more than half of Democrats (and far more Independents and Republicans) support photo ID in recent polls.
Schwartz makes much of the fact that survey respondents were not asked specifically about different kinds of ID that correspond to what he describes as “new restrictive” laws.
This is true, but to interpret this to mean that the public “does not support” these laws, as the headline suggests, strains credulity. Schwartz is much more balanced in his opening paragraph and perhaps didn’t write the headline:
Defenders of photo ID laws regularly cite public opinion polls that show widespread support for their arguments. Yet these polls reveal no such support, and they prove nothing about this new restrictive legislation because the polls’ questions cover a far broader range of IDs than the actual laws accept as proof of identity. Many of the new laws do not accept a college student ID, for example, or an out-of-state driver’s license; but the polls drawing favorable responses encompass such IDs. As always, the devil is in the details.
However, my reading of the survey evidence is different from Schwartz. Given the consistently strong level of public support in many different polls, even when forms of ID are described, I would be surprised if the public suddenly turned toward opposition if a question included reference to a specific ID.
Schwartz also refers to a study conducted by scholars at the University of Delaware that shows, in his words:
Surveys show that respondents who believed that photo ID laws make it harder for some people to vote are less supportive of these laws. Yet none of the 37 polls examined for this article included information about the effects of the law on minorities and others.
Again, a true statement, but the devil remains in the details. Here’s what the study actually showed: when respondents were told that a voter ID requirement might stop some eligible citizens from voting, support for the law dropped 12%, from 81% to 69%. Yes, support dropped. From an overwhelming majority to just a strong majority. Not exactly opposition.
Schwartz rightly notes that survey respondents are never asked their opinions while being informed about the impact of these laws on minorities and other disempowered segments of the population. But what precisely would we tell respondents? That some voter ID laws have been ruled constitutional, others ruled unconstitutional, and that the impact of voter ID on turnout remains an open question among scholars? (The Brennan Center has an extensive list of scholarly research and there is another superb summary at the Journalists Resource Center.)
The point is this: since we have not reached a scholarly consensus about this impact of voter ID on turnout, survey researchers cannot present this as a fact.
It is true, as Schwartz points out, that public remains woefully uninformed about how photo identification laws operate, what specific kinds of IDs are required, and how ID requirements are often unequally applied. The public believes that voter fraud is far more frequent than it actually is. (And Schwartz does not add that support for voter ID among the public has a clear racial component, whether that be conscious or unconscious.)
It is also true that voter ID laws appear to be promoted primarily be Republicans attempting to hold on to political power in electorally competitive states, as my colleague Will Hicks and other collaborators have shown.
But much as opponents may dislike it, it is finally true that the public has long expressed strong levels of support for these laws, and the strong public consensus provides ample political cover for anyone advocating photo ID.
Misunderstanding public sentiments is not an option–instead, opponents to photo ID need to engage, inform, and move public sentiments.