CEDAR CITY — While year-end testing for public school students isn’t anything new, the introduction of SAGE comprehensive student assessment in Utah during the 2013-14 school year has been surrounded with controversy. One of these areas of contention deals with the fact that Student Assessment of Growth and Excellence, or SAGE, scores have been used for teacher evaluations. However, a bill that passed the Legislature Thursday, the Student Testing Amendments, will now restrict the use of end-of-level assessment scores for the evaluation and compensation of certain employees.

In her introduction of House Bill 201 during House debate, Rep. Marie Poulson, D-Cottonwood Heights, called it “very simple to understand.”

“We’re not meddling in local schools,” Poulson said. “This is something we did. SB64 (Public Education Employment Reform) in 2012 started this practice, and it’s been problematic ever since.”

Poulson said she has been working on this issue for a couple years and that the use of SAGE testing for teacher evaluations is a detriment to the Utah education system. She said:

One of the most interesting things about this issue is that it seems to be the major reason why so many teachers are telling me that they are leaving the classroom and retiring early. They are feeling like they are held accountable for things that they don’t have control over.

Problems with using SAGE testing for teacher evaluation

In her discussion of the bill, Poulson listed several problems with using this method of teacher evaluation, including students opting out of the test, the efficacy of the test for student evaluation, a lack of equity in teacher evaluations and the possible litigation that could result from such a policy.

The first problem area Poulson mentioned was the fact that parents can choose to have their students opt-out of the testing. While statewide opt-out rate has only been about 3 percent, Poulson said, many charter school and individual district rates are higher. She cited one charter school with up to 72 percent opt-out rates and the Provo School District with over 11 percent.

“If a teacher has 30 students in class … and five of their best students opt out, resulting data is artificially low,” Poulson said. “If five of their worst students opt out, it’s not representative of the class.”

Another problem with using the test scores to determine how well a teacher is performing is the fact that the state doesn’t allow end-of-level test scores to be part of the students’ final grade. Poulson said this makes it difficult to get students engaged in doing well and that often students will try to “game the test,” especially with an adaptive test such as SAGE where the difficulty of the test changes in real-time based on how well or poorly the students perform.

“In many of the computer-adaptive tests, (the students) will purposely miss the first few questions so the test doesn’t get harder.”

Additionally, Poulson said that teachers are reporting difficulties in administering the tests, something which ultimately affects the scores. To support this, she referenced grade school students, English as a second language students and students in special education.

Third graders are being administered the SAGE tests on computers, and the teachers are wondering whether we are testing keyboarding schools or the material on the test, as many of them don’t have the skills to adequately do the test. Also, reports where students are not speaking English at home; do they understand the material? Also, so many of the reports I’m hearing come from special ed. teachers … It’s very hard to just manipulate the taking or the administration of the test.

Poulson said another issue is the inequity in teacher evaluations, seeing as not all secondary education instructors are teaching subjects that are assessed by SAGE testing: math, English language arts and science.

However, Poulson said “the most damning evidence” is a January 2016 report from Regional Education Laboratory West.

According to the Regional Education Laboratory West website, the group serves Arizona, California, Nevada and Utah with a mission “to provide research, analytic support, and resources that increase the use of high-quality data and evidence in education decision-making.”

Poulson said that according to the report, the results being used to determine teacher evaluations are over 59 percent inaccurate and unstable.

“What this study suggests is that we should be very cautious in using these scores for teacher evaluations, dismissal and tenure.”

Poulson went on to reference statisticians from the Jordan School District who testified in the House Education Committee that using these scores for teacher evaluation would not hold up in court.

The House Education Committee gave the bill a favorable recommendation by a vote of 12-1 with 1 not voting.

Legislative process 

On March 1, the House passed the bill to enact Student Testing Amendments with a vote of 60-4 with 11 not voting, sending it to the Senate. From Southern Utah, only Rep. Brad Last voted against the bill; Reps. Mike Noel and Jon Stanard were absent or did not vote. Rep. Last was unavailable for comment.

On March 8, the bill went to the Senate Education Committee. During the committee hearing, Sen. Mark Madsen raised concerns about removing the testing from teacher evaluations, saying that he believed there needed to be some form of quantifiable measurement of student achievement. He said:

I know it’s imperfect. I think we’re removing an imperfect tool, and we’re replacing it with an even less imperfect tool. I know it’s used in these other subjects that don’t lend themselves to any kind of a standardized measurement. There’s got to be something else, and I’m not getting the impression that there is any other measurement … There is no way to measure student growth. We’re moving away now into the completely subjective realm.

Interim State Superintendent of Public Instruction Dr. Sydnee Dixon responded to Madsen’s concerns by saying that the points Poulson mentioned about students opting out and the variability had actually made using the test scores for teacher evaluations more subjective than objective. She also said that teachers were still going to be held to certain statewide Common Core educational standards.

“There is actually a very strict process that (teachers) have to follow,” Dixon said. “There’s a template that they plug their numbers into, and it’s really looking at the students they have, so they have to look at data when they come in. They can’t just sort of say, ‘Well, I think I’m going to teach them this, and we’ll see if it works or not.’ They really have to look at student data. … For some teachers, they might want to look at their SAGE assessment. For some, they might create another end-of-level assessment.”

Sen. Howard Stevenson raised the question of whether the SAGE test questions could somehow be embedded into instruction throughout the year and actually be part of teacher evaluation. Dixon said she didn’t believe the bill would prevent that from happening in the future.

The committee gave the bill a favorable recommendation by a vote of 5-0 with 3 not voting, with Sens. Madsen and Stevenson marked as two of the three absent votes.

The bill passed the Senate Thursday 27-0 with 2 not voting. All senators from Southern Utah voted in favor of the bill.

Read the full text of the bill as passed here: 2016 HB 201 2nd Substitute – Student Testing Amendments – passed 20160310

Email: [email protected]

Twitter: @STGnews

Copyright St. George News, SaintGeorgeUtah.com LLC, 2016, all rights reserved.