I just wanted to follow up on a question Andrew Coulson raised last week about a poll showing a plurality of South Carolina African Americans in support of school choice. Andrew notes:

A new poll released today reveals that 43 percent of African Americans in South Carolina support private school choice while only 40 percent oppose it. What’s even more interesting, however, is that 53 percent said that “giving parents a tax credit or scholarship to choose the best school for their children — public or private — would improve the state’s dismal high school graduation rate.”


So an additional 10 percent of respondents think the program will work but don’t currently support it. Why? Perhaps because many black religious and political leaders in South Carolina have criticized the concept for years.

Certainly opposition from black leadership has probably softened support, but I don’t think that explains the difference in support between the first and subsequent questions. As Andrew notes, the other results peg pro‐​choice responses consistently at 53 percent.


Here’s the question in full: “Should parents, grandparents or custodial relatives be allowed to receive state scholarships for their children to go to private school if they feel the public school is not meeting their children’s needs?”


First, the description of the tax credit program instead implies a state voucher program. This is bad wording, but probably doesn’t drop support since black support for vouchers tends to be equal or higher than support for credits.


I think the real problem here is the phrase phrase “state scholarships.” This sounds to me like there very well could be conditions, such as academic merit, placed on who is eligible for the “state scholarships.” There are need‐​based and merit‐​based scholarships, but they are typically not available to all, and the question is at the very least confusing. This ambiguity, with the suggestion of limited availability, might have softened support/​increased undecideds.


In the context of consistent 53 percent support on other, better‐​worded choice‐​related questions, I think we can reasonably conclude that poor question wording on the first question likely dropped support for school choice about 10 points.


We really need to be careful with public policy questions … small changes can have a serious impact on the results.