A substantial majority of Americans support the idea of school vouchers. That’s what the First Amendment Center, a program of the Freedom Forum, a liberal institution, reported on August 1 in its annual publication,”State of the First Amendment.” In a survey taken earlier this year, the center found that 62% of Americans agreed that parents should have the choice of sending their children to nonpublic schools with vouchers provided by the government; only 35% disagreed with the proposition. Voucher supporters understandably will take heart from the result of this one-shot poll. No doubt voucher opponents will try to discount it. Neither will have the high ground here, however, because there has not been enough long-term polling to get a clear picture of the public’s views — a problem that will be put in sharp relief when the nation’s most influential annual survey on education attitudes is released later this month.

The best-known, most widely watched poll on education, the Phi Delta Kappa/Gallup Poll of the Public’s Attitudes Toward the Public Schools, has some of the only tracking data on public support for vouchers. Unfortunately, according to the poll’s co-author, Lowell Rose, it tells us little. Voucher support, as measured by Gallup, Mr. Rose said,”is jumping up and down.” In 2000, 39% supported vouchers in the PDK/Gallup poll; in 2001 it was 34%, and in 2002 it was up 12 points to 46% (the poll was conducted before the Supreme Court ruled in Zelman v. Simmons-Harris).As to why this is happening, Mr. Rose told the Sun recently,”I don’t have an idea in the world.”

Some, including Terry Moe of the Hoover Institution, have raised doubts about bias in the PDK/Gallup poll. The poll’s flagship voucher question reads: “Do you favor or oppose allowing students and parents to choose a private school to attend at public expense?” To Mr. Moe, the words “public expense”seem loaded. Past polling results lend weight to this concern. The editor of PDK’s print journal, Bruce Smith, recounted to us an experiment done with the poll in the 1990s, where the phrase “public expense” was replaced with “government expense.”Support for vouchers was eight points higher for the test group that heard that the “government,” as opposed to the “public,” was going to pay for choice.

What is needed is a better-phrased question, and one asked consistently over a series of years. Such a question turns out to be included in the PDK/Gallup poll. That one reads: “A proposal has been made that would allow parents to send their school-age children to any public, private, or church-related school they choose. For those parents choosing nonpublic schools, the government would pay all or part of the tuition. Would you favor or oppose this proposal in your state?” The misfortune is that this question comes directly after the previous — arguably biased — question, tainting the results. Still, even then, 52% of public-school parents favored this more neutrally worded proposal in 2001.

The newest PDK/Gallup poll will be released on August 20. While Mr. Rose, the poll’s author, didn’t want to give too much away, he told the Sun this much: The wild fluctuation on the voucher question continues. Perhaps this will tell Mr. Rose what is obvious: The poll isn’t terribly accurate. If next time Messrs. Rose and Smith ask the questions in a more neutral fashion, they will do a great service. They would do even more of a service if they expanded the questioning on this issue to more than two questions. “Legislators are moving now to enact voucher proposals, and they are having some success,”a senior scholar in religious liberty at the First Amendment Center, Charles Haynes, said recently. “We should ask parents if they would like alternatives in their community, if they think vouchers would be one way to accommodate them.”