We Want to Give You a Choice of State Tests, Wis. Legislators Tell Districts
After his successful re-election campaign last year, Wisconsin Gov. Scott Walker, a Republican, pledged to make significant education policy changes in the state this year. The state legislature has taken that idea and run with it.
Speaker of the House Robin Vos and other GOP representatives have introduced legislation, Assembly Bill 1, with broad ramifications for schools, including a high-profile proposal to require traditional public schools deemed low-performing for four consecutive years to convert to charter schools. Vos told Associated Press reporter Scott Bauer that he wants lawmakers to pass the bill by the end of this month, and that, "This is the most important bill we have on our schedule right now."
My colleague Arianna Prothero noted the bill's implications for charter schools on her blog "Charters and Choice" last week. But I want to focus on another section of the bill that deals with student assessment. In short, it would provide traditional public schools, charters, and private schools participating in the state voucher program to choose from a menu of state assessments for certain grades.
New Choices for Schools
Current state law is that the state superintendent of public instruction (right now that's Tony Evers) approves or adopts assessments for traditional public schools, charters, and private schools participating in the state voucher program to use in the 4th, 8th, 9th, 10th, and 11th grades respectively. Those grades are specified because—the No Child Left Behind Act aside—those are the grades for which state law requires state assessments to be given.
Here's what the new Wisconsin bill would do with testing:
• First, the legislation would create an Academic Review Board, a new agency that would in turn come up with a new K-12 accountability system using A-F grades for schools.
• Next, among other responsibilities, the review board would be tasked with picking three nationally norm-referenced tests to serve as alternatives to the one the state chief selects for each grade level. (The Iowa Test of Basic Skills and the Stanford Achievement Test are examples of such tests.) Those assessments would first have to be validated by the University of Wisconsin-Madison Value-Added Research Center as appropriate for statistical comparison with those tests selected by the state chief.
• Finally, traditional public schools, charter schools, and private schools participating in the voucher program would be permitted to pick from one of those three alternate tests instead of the ones approved by the state chief. The U-W research center would then be required to "statistically equate" the results from those exams to those from the respective tests approved by Evers' office, so that the achievement and growth of students at the school can be measured. The corresponding data from schools wouldn't be collected until the 2015-16 school year for traditional public schools, and 2016-17 for charters and private schools.
• One key caveat: If a school decided to pick an alternate test, the state wouldn't cover the cost, and the school would have to pay for it.
Let's focus on the state's current plan for English/language arts and math tests, in part because they're the two subject-area tests that will be used in the bill's school accountability proposal, but also because of their connection to the Common Core State Standards.
For this coming spring, Wisconsin plans to use the Smarter Balanced assessments (aligned to the common core) in grades 3-8 for E/LA and math, except for the 1 percent of students with the most severe cognitive disabilties who will be taking the Dynamic Learning Maps Assessment, according to Evers' office. In the high school grades, the suite of tests from the ACT, including Explore and PLAN, are slated to be used.
Under the terms of the bill, what tests could the University of Wisconsin-Madison conceivably validate as comparable to Smarter Balanced? The legislation doesn't offer any specific examples. Brad Carl, an associate director at the U-W research center, told me in an interview that without even seeing an operational Smarter Balanced test, it's impossible for researchers to say which other nationally norm-referenced tests could provide comparable test results. (The first operational Smarter Balanced tests will be given in March, Luci Willits, a deputy executive director with the consortium, told me.)
What helps contribute to useful and accurate comparisons between results from different tests? There's a variety of factors, Carl told me, including whether the tests are paper-and-pencil or computer adaptive, whether they measure the same content, and whether there are accommodations for special populations like English-language learners.
In essence, Carl told me that strong comparisons between different, nationally norm-referenced tests are certainly possible and can be useful, but that they can go off the rails very easily because of the number of variables involved. He also said he's "not a big fan of requiring that there be three" options from the new advisory board, as the bill requires, since finding even that many nationally norm-referenced tests that can be equated to results from Smarter Balanced might be tricky.
"As always, the devil is in the details on these sorts of things. How close is close enough?" Carl said.
Carl noted that beyond technical issues, it also makes a difference whether the test results are being compared for the purposes of school and teacher accountability. Using the comparisons for those purposes, he said, makes the exercise more complicated.
And then there's the fate of common core itself in the state. Walker has vacillated between calling for its repeal and merely calling for districts to have the option of using the standards.
However, Wisconsin districts already have the option to use other standards—typically, districts decide to use the common core standards because the state tests will be aligned to them. If that sounds familiar, there's a similar dynamic at work in New Hampshire.
So if schools will soon have the option of picking their own state tests to use, but results from those state tests must be statistically comparable to those from Smarter Balanced, a common-core aligned test, what incentive would districts have not to use Smarter Balanced for the relevant grades and subject areas? It's not immediately clear, although pushback to the common core might spur some districts to reject Smarter Balanced, one of two consortia to receive federal funding to develop common-core aligned exams.
Private schools participating in the state voucher program, meanwhile, fought against a state law approved in 2013 that makes them publicly report student test scores on state exams. While this new bill wouldn't change that reporting law, it would give those private schools more choices about which tests it could use to satisfy it. One of the sponsors of the bill, GOP Rep. Jim Steineke, expressed the notion that the bill is supposed to have universal consequences for K-12:
No Role Models
If you're looking for a state that analyzes student performance on multiple nationally norm-referenced tests, Florida is perhaps the most prominent one. For years, the state has commissioned annual reports on how students using the state's tax-credit scholarship program have performed on the Stanford Achievement Test, the Iowa Test of Basic Skills, and TerraNova.
However, in the most recent such Florida report, researcher David Figlio stresses that student results on these tests should not be compared to test results from students in traditional public schools taking the state assessment, the Florida Comprehensive Assessment Test (FCAT).
Carl said he's not aware of any state that is already doing specifically what the bill is asking with respect to the intrastate comparison of test results that can subsequently be used for school and teacher accountability.
On a related note, keep in mind that there's a strong chance the Wisconsin voucher program will be expanded this year. It will be interesting to see how any such expansion plays with the push to overhaul K-12 accountability in the state.
UPDATE: A counterpart bill in the Senate to change school accountability in the state differs markedly from Assembly Bill 1. Among other differences, according to the Wisconsin State Journal, the Senate bill would not give schools a menu of state assessments to choose from. It also wouldn't assign A-F letter grades to schools, as the state Assembly's bill does. And Senate Majority Leader Scott Fitzgerald seemed to dismiss the idea that the two bills could be merged in a future conference committee.
In Jan. 14 testimony about the Assembly bill, Evers raised numerous concerns, including the need for "one, uniform assessment for accountability. Multiple tests reduce validity, transparency, and accuracy-significant problems for high stakes accountability."
Does this bill presage a new state in some states regarding student assessment, particularly as it pertains to school choice? As I wrote last month, testing promises to be one of the biggest education policy issues state legislators will grapple with this year, although exactly how states will change their approach remains to be seen, after months of a national debate about the nature and volume of student assessment. And testing is at the heart of discussions about the reauthorization of the No Child Left Behind Act.