Two Polls Diverged On The Issue Of School Choice: An Explainer
By Jennifer Wagner
Pardon me while I put my ex-political hack hat on for a moment to talk about polling.
When you work in politics, you don’t just love polling; you live or die by it. You have to constantly know where you stand with likely voters so you can fine-tune your campaign to attract more support.
If you ask any politician who’s down in the polls how they’re feeling, they’ll almost certainly give you some version of this response: “Well, a poll is just a snapshot in time, and we’ve still got a chance to move those numbers before the election.” Or they’ll go on a rant about how their supporters aren’t represented in the sample. Or they’ll rail on polling as bunk science. Whatever the case, you can also bet that if they have sufficient funds in their bank account, they’ll be out in the field testing out new messages to move the needle.
Obviously, that’s the world of political polling, where it’s much easier to judge accuracy based on the results of a campaign. (Fun fact: After you’re done winning or losing, pollsters themselves get rated based on how well they did predicting the outcome of your race. It’s a cutthroat world, and there are a lot of players trying to get it just right.)
Issue-based polling exists in a more nebulous environment. That’s not to say the data are questionable, but we don’t have any way in the short-term of determining whether what we see in our survey work matches up with where parents decide to send their kids to school. We can only figure that out using longer-term trend lines.
All of which is important to keep in mind as we look at the large — and growing — body of research on K-12 education in America.
Last week, three national organizations, including EdChoice, released annual surveys that touch on K-12 issues. (You’ll find our 2021 Schooling in America survey, now in its ninth year, here. PDK’s 53rd annual poll is here. And the 2021 Education Next poll is here.)
This year, PDK took a slightly different path than in the past, choosing to focus solely on pandemic-related schooling questions and less on broader topics like quality of instruction, student discipline, campus safety and other top-of-mind issues for parents and families.
Accordingly, for purposes of this post, we’ll just be looking at Schooling in America and the EdNext Poll. There are some notable differences between them on the topic of school choice.
On the topic of charter schools, for example, parents in our poll reported 74 percent favorability, where only 43 percent of parents offered support in the EdNext poll:
When asked about school vouchers, 48 percent of parents in the EdNext poll reported support for a universal program (no income limits), and 50 percent reported support for a means-tested program. Support for vouchers among the general public dropped 10 points over the past two years — from 55 percent in the 2019 EdNext Poll to 45 percent in the 2021 EdNext Poll.
Schooling in America, on the other hand, showed 67 percent support for vouchers among the general public and 78 percent among parents.
These data points might lead an outside observer to conclude one of two things: (1) support for school choice has taken a huge hit over the past two years; or (2) one of these polls is way off-base.
Support for school choice appears to be high not just statistically, but legislatively: Over the past year, 18 states have passed seven new school choice programs and expanded an additional 12 existing ones. Families have spent the last 18 months trying to bring order to the chaos that enveloped K-12 education during the pandemic. Quantitative and qualitative research alike indicates they are eager for options, and many were disappointed in how their children’s schools responded.
Then the polling must be messed up, right?
I’m a professional communicator with a law degree, which makes me far less qualified than our research team to answer that question. So I asked a couple of them why they thought the numbers might be so far apart.
Among the reasons posited that might explain the differences between EdNext’s annual survey and ours was that we allow people to answer “don’t know” and move on to the next question; less than 10 percent of respondents typically select this. On the other hand, EdNext includes “neither support nor oppose” when reporting their results, an answer that winds up at 20 percent or more of respondents, depending on the question. These are very different options. One aims to identify people who don’t know or are unsure about the answer; the other allows people who might lean one way or the other to identify in the squishy middle. EdNext’s results make it clear who is strongly in support or strongly opposed to a particular topic, but the way their data are reported makes it potentially more difficult to see the leaners — people who offer soft support or opposition for an issue.
While we’re on the topic about language, it’s worth noting that we ask questions about school choice differently. EdNext asks one version of each question. Their language on vouchers:
A proposal has been made that would give all 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. Would you support or oppose this proposal?
When we ask about vouchers, we ask two versions of the question, one without specifics to get a baseline of support:
Based on what you know, or have heard from others… In general, do you favor or oppose “school vouchers”?
And one that offers a definition of the policy so we can find out how people feel when they have more details:
A school voucher system allows parents the option of sending their child to the school of their choice, whether that school is public or private, including both religious and non-religious schools.
If this policy were adopted, tax dollars currently allocated to a school district would be allocated to parents in the form of a “school voucher” to pay partial or full tuition for the child’s school. In general, do you favor or oppose a school voucher system?
There’s nothing wrong with either approach, but it’s easy to see how they could lead to very different outcomes.
Another reason the results might differ is timing. EdNext fielded their poll in late May and early June. We fielded ours from June 14 through July 8. The former date range is when many kids were getting out of school; perhaps choice wasn’t top of mind for parents looking forward to summer break. As summer evolved — and it became clear the Delta variant might further disrupt K-12 education — perhaps they began thinking about the issue again.
One thing missing from EdNext’s poll: education savings accounts or ESAs. This type of school choice encourages families to customize an education for their child using government-allocated funds that are placed in a safe, secure “savings account” for use on approved items and services. This is the most flexible — and popular — form of school choice. Our monthly and annual polling shows ESAs are far more popular than anything else, routinely clocking in at support levels of 70 percent or more. We’ve seen record-high levels of support for ESAs over the past five years in Schooling in America. This year, the policy clocked 78 percent support among the general public and 84 percent among parents:
EdNext isn’t wrong to exclude ESAs, but it would be interesting to see in future polling how their results would stack up against ours.
All of which brings us back to the original theme of this post: Polls are a snapshot in time. One isn’t necessarily better than the other. Different polls complement each other, and these both offer great insights into how Americans think about the K-12 system (of particular note in the EdNext poll are questions about school boards and how and when they are elected).
There is no such thing as too much polling when it comes to issues like school choice. The more we know about how people, especially parents, think and what matters to them, the better advocates, policymakers and community leaders can respond to those needs. That’s why we poll monthly and quarterly, as well as annually, to make sure we’re capturing as much data as possible to inform the conversation.
Jennifer Wagner is a mom, a recovering political hack and the Vice President of Communications for EdChoice, a national nonprofit that supports and promotes universal school choice.
