PDK/Gallup Public Education Survey Has Earned Skepticism, and Here's Why
Seeing as how it’s back-to-school season, it must be time for my Third Annual “Let’s Take a Closer Look at the PDK/Gallup Public Education Survey” posting. It will teach you to take the headlines with a grain of salt.
Without time to delve into every issue and inspect every question, there are a few points worth examining about American public opinion on education. The greatest clarity perhaps comes from a result consistent between PDK/Gallup and the new Harvard/Education Next survey, and consistent with previous years: About half of Americans give their local public schools an A or B grade, but only one in five do the same for the schools nationwide.
Now that we have that out of the way, let the conflict begin! American Federation for Children responded quickly to the results of one particular question that alleges 70 percent opposition to private school vouchers:
The poll asked respondents about various other forms of educational choice, including charter schools, homeschooling and online education. In each of those instances, respondents overwhelmingly favored these educational options. When PDK asked respondents about their support for publicly funded private school choice, the question was worded, “Do you favor or oppose allowing students and parents to choose a private school to attend at public expense?” This question parrots the same false narrative that opponents of educational choice have been using for 30 years. It is not the school that is funded, it is a child, who then attends a school of their parents’ choice.
Wait. Skewing how a question is worded to achieve a result? Specifically, tacking on the phrase “at public expense” on a question about choice? It’s a familiar tactic used as recently as this summer by the American Federation of Teachers (not for Children), ably debunked by the Cato Institute’s Jason Bedrick. He pointed to the 2012 Harvard/Education Next survey, which came up with significantly different results.
This time around, 44 percent told the Harvard pollsters they support the proposal “that would give families with children in public schools a wider choice, by allowing them to enroll their children in private schools instead, with government helping to pay the tuition,” with 37 percent opposed. But school choice supporters like myself ought to take heed that opposition on that question rose from 29 percent last year, nearly all shifting from the neutral position.
(No doubt a different trend than the widespread support for scholarship tax credits, as other national surveys have shown.)
A couple other issues jump out from the new PDK/Gallup survey results. First, the published claim — touted by the Washington Post‘s Valerie Strauss — that “Fewer than 25 percent of Americans believe increased testing has helped the performance of local public schools.”
Two days earlier, an Associated Press poll found that 69 percent of American parents believe standardized tests do well at measuring the quality of public schools, and 75 percent hold that belief about their own child’s school. In addition, 61 percent think their students take about the right amount of standardized tests, and 11 percent more think the number of tests is too few.
So it appears a majority of the population like standardized tests as a good measurement of school performance, think students are tested about the right amount, but don’t think adding more tests will help improve that performance. Depends on which way you want to spin it, I guess. But taken together, it seems like a fairly sound, commonsense view to me.
One last example to raise the eyebrows. PDK/Gallup proclaims that 58 percent are against using those standardized test scores as part of a teacher’s evaluation. Specifically, the question says: “Some states require that teacher evaluations include how well a teacher’s students perform on standardized tests. Do you favor or oppose this requirement?” Even I can see the problem with the way that’s written. Just using a student’s raw achievement score would unfairly capture a teacher’s effectiveness.
PDK/Gallup should take some question-writing input from the Associated Press, which found that 53 percent of parents believe “changes in students’ statewide test scores over time” should factor into teacher evaluations “a great deal” or “quite a bit,” with another 26 percent saying they should factor into evaluations “a moderate amount.”
Harvard/Education Next offers up a slightly different twist to reinforce doubts about PDK/Gallup, finding 58 percent like proposals that would “require teachers to demonstrate that their students are making adequate progress on state tests” in order to keep tenure, somewhat like Colorado’s own SB 191 going into effect statewide this year.
There are many other questions that could be examined. But just “surveying” the results on school vouchers, standardized tests, and teacher evaluations, do you better understand now why the PDK/Gallup education survey results have to be taken with a strong dose of skepticism?