What Do Americans Think of School Choice? Depends on How You Ask the Question

Donald Trump’s nominee for education secretary will be grilled on a familiar but wide range of policy questions ranging from federal K-12 funding to the Common Core standards to college debt to early-childhood education. But the president-elect’s vision of education appears to center on large-scale expansion of choice programs, and since her nomination DeVos has become something like the poster girl of school choice. A billionaire evangelist of charter schools and school vouchers, she has, through generous political giving, helped bankroll a major expansion of Michigan charters and a failed 2000 state referendum on vouchers.

Her nomination has also bought her something of a bifurcated reputation: To teachers unions and traditionalists, she’s become a pariah. To defenders like fellow Michigander Mitt Romney, though, she’s a champion for kids stuck in failing schools. Trump clearly agrees.

Issues relating to choice could therefore feature heavily in DeVos’s hearing. Though Republican control of the Senate makes her confirmation probable, it won’t answer three questions raised by the nomination. To wit: Do most Americans actually support school choice? Can they agree about “school choice” means? And how is DeVos likely to affect these views from the Trump cabinet?

The answer to all three questions, frustratingly, is that it depends how they’re phrased and who’s being asked.

Public opinion polling on education isn’t especially plentiful, and questions about school choice have been only sporadically surveyed. These questions are also particularly sensitive to nuances of wording.

“It is challenging to ask about issues of school choice, and vouchers in particular, in a way that elicits a true barometer of public sentiment because the way in which you choose to present the issue to the public really matters,” said Martin West, an associate professor at the Harvard Graduate School of Education.

As editor in chief of the journal Education Next, West helps administer an education poll to more than 4,000 American adults annually. The poll’s findings demonstrate the extent to which respondents’ answers may be influenced by how questions are asked.

Consider school vouchers. For many, the very concept — a form of taxpayer funding for private education — flouts traditional ideas of public schooling. The EdNext pollsters devised two separate ways of raising the issue: One emphasizes the increased options available to parents, while the other stresses the use of public monies. Using the first construction, respondents in 2016 were evenly divided about vouchers (45–44). Using the second, support for the idea withered (29–57).

That kind of variability, and the nuances of wording that generate it, sometimes rouse accusations of bias. Phi Delta Kappa International, a professional organization for educators that has administered a respected poll annually since 1969, has varied the way it asks about vouchers, but its more frequent question has troubled some choice advocates: “Do you favor or oppose allowing students and parents to choose a private school to attend at public expense?”

That formulation, with its buzzkill finale, has yielded predictably negative responses over the years (31–57 disapproval in 2015, nearly the same as Education Next when using its less sunny description) and elicited accusations that the group puts its thumb on the scale. PDK Chief Executive Officer Josh Starr vigorously defended the survey’s integrity in an interview, pointing to the ideological diversity of the advisory panel that helps construct it. Still, he conceded, “It is true that with polls, it’s how you ask the questions [that determines] how you get the result.”

While more widely supported than vouchers across a variety of polls, charter schools also present a dilemma. After filtering out respondents who say they have no opinion, both the Education Next and PDK polls show strong and nearly identical support for charter schools (71 percent and 72 percent, respectively).

But those figures become less meaningful when we consider that the public may not understand what a charter school is. According to PDK’s 2014 poll, respondents believe that charters can charge tuition and select students based on ability; they are also divided evenly on whether charters can teach religion — and whether they even qualify as public schools. Those basic misconceptions remain common more than 25 years after the charter movement began.

No matter how carefully constructed, polls suffer from the deficiencies of their respondents, whose views are fluid and subject to incomplete or bogus information.

Amid the noise, however, it’s possible to decipher a growing trend.

Last August, Gallup released a survey on Americans’ views about public education. In contrast to previous years, the 2016 sample showcased a huge split in partisan conviction: 53 percent of Democrats described themselves as satisfied with the quality of education, while only 32 percent of Republicans agreed. That’s a startlingly large difference on an issue that doesn’t typically stoke political passions (at least not in recent history). The average satisfaction gap between the parties over the 18 years leading up to 2014 was just 4.13 points.

There’s some precedent for this phenomenon, especially in election years. In 2004 and 2012, the satisfaction gap between the parties was eight and 12 points, respectively. But 2016’s 21-point chasm began widening in 2015, when the Common Core standards emerged as a contentious (and widely misconstrued) topic in national media.

Partisan polarization is reflected in how Americans view their government, the economy and most policy concerns; its encroachment into education should alert lawmakers and educators — if they’ve somehow missed the fracas of recent years — to guard against greater ruptures.

EdNext’s 10-year polling averages present compelling evidence of growing division. In 2010, Democrats and Republicans both registered 70 percent support for charter schools. While Republicans are now even more likely (74 percent) to hold that position, Democratic support has flagged in the intervening years, now tallying just 58 percent. In the PDK poll as well, charter-friendly Democrats shrank from 62 percent to 50 percent between 2014 and 2015.

PDK’s Starr believes that media coverage and party messaging on the issue might be influencing voters who are less likely to peruse policy papers. “As charters have become a political football, there’s a trickle-down effect on the person who … just sees it as something that gets written about,” he said. “Like everything in America these days, it’s subject to that kind of political dynamic.”

A partisan gap has opened up on the voucher end as well, though it is narrower and perhaps more surprising. Support has declined significantly among the public at large over the past five years, with overall approval for income-targeted vouchers falling from 55 percent to 43 percent over that span. Notably, however, this softening has been concentrated among Republicans, whose political leaders are far likelier than Democrats to endorse such proposals. Regardless of how the question is posed, but especially when the dreaded matter of public financing is foregrounded, Democratic respondents are consistently more pro-voucher than Republicans.

EdNext’s West proposes an explanation grounded in self-interest: While the Democrats’ urban minority constituencies may covet scholarships to private schools as alternatives to their subpar district offerings, Republicans are more likely to dismiss such mechanisms as handouts. “It seems that many Republicans have come to see vouchers as a government program benefiting low-income families that is unlikely to do anything for them,” he said.

Of course, there’s no better measure of how the public views private school choice than the results of the voucher referenda that have actually gone before the voters. Dating back to 1967, the cause of freeing public dollars for private education has been defeated virtually every time it has appeared on a ballot — including losses for tuition reimbursement, tax credits, auxiliary services like textbooks and transportation, and full-fledged statewide voucher schemes.

Two of the worst defeats came in 2000, when voters in both California and Michigan comprehensively rejected ballot initiatives by margins of 41 and 39 points, respectively. Both found friends in civic-minded philanthropists, though these deep-pocketed allies couldn’t carry the day. In Michigan, as has been widely reported, the prime mover was Betsy DeVos.

Donald Trump has pledged $20 billion for “charters, vouchers and teacher-driven learning models,” an agenda that would suit his would-be education secretary very agreeably. The extent to which these mechanisms meet the expectations and preferences of the electorate is another question. The only poll conducted so far on the plan was released last week by the American Federation for Children, an advocacy group founded by DeVos that trumpets itself as “the nation’s voice for educational choice.” By a 51–35 margin, the survey sample backed the idea of “shifting twenty billion dollars in funding from other programs to school choice.”

Note the diction: the word “choice” tends to poll well, and the generic “other programs” displaces costs into the ether.

That’s a fairly good illustration of where we are in determining American attitudes on both traditional schools and reform alternatives. “Incoherence,” as education researcher Morgan Polikoff has written, “may well be the defining characteristic of Americans’ attitudes toward our public schools.” We like the school down the street, but everyone else’s are in dire shape. We admire charters but can’t be relied on to define what they are. We’re willing to fund private alternatives, but mostly when the money is redirected from “other programs” with no constituencies.

Americans speak increasingly less with one voice, and, divided, with less care.

“I think that most people, frankly, are pretty far removed from the realities that Marty [West] or myself think about all the time,” concluded Starr. “I don’t know how they’re making their calculations. But I think it’s likely completely different from how we think they are.”

Kevin Mahnken is an editorial fellow of The 74.