This is an archived article that was published on sltrib.com in 2016, and information in the article may be outdated. It is provided only for personal research purposes and may not be reprinted.
The Salt Lake Tribune's Feb. 8 front page focused on Utah's teacher shortage and the fact that young people are no longer choosing education as a career. Many factors contribute to this: low pay, large class size, loss of retirement benefits and lack of respect. However, a major contributing factor for the teacher shortage is that teachers are now being held accountable in areas where they have little or no control. Unreliable data is used to determine teacher evaluations.
In 2015, the Utah Legislature passed SB 204, which allowed parents to opt out their children from year end assessments. Statewide, the opt out is about 3 percent, but in many charters and districts, the opt-out rate has been significant, up to 72 percent in some schools and over 10 percent total in some districts. For example, if a teacher has 30 students and five of the best students opt out, the resulting data is artificially low.
If five of the worst students opt out, the data is not representative of the class.
My experience as a teacher administering these assessments is similar to the reports of many other educators. It is difficult, if not impossible, to motivate students to do well on assessments when it is evident that they "have no skin in the game." This is further complicated by Senate Bill 204, which prohibits the use of scores on a student's grade or advancement. However, even before the application of this law, test-fatigued pupils have clicked away — all A's or all C's — knowing that the results do not affect them. Computer adaptive tests are no different. Students purposely miss the first few questions so the test does not become more difficult.
Another troubling problem in tying these test scores to evaluations is the fact that not all tests are curriculum-aligned. They may measure general aptitude, but not necessarily what is taught in that class. Also, the rules involving evaluations and test scores are applied unevenly. In secondary school, a high percentage of teachers teach non-tested subjects. Rules in some charter schools differ greatly from those in the public schools. Teachers report difficulties administering the tests that could affect results. For example, do third graders have adequate keyboarding skills to take a test on a computer? Have we made the proper adjustments for special education students? Do students who don't speak English as their first language understand the questions?
Perhaps the most damning evidence, however, is research published in January of this year by Regional Educational Laboratory West, which serves Arizona, California, Nevada and Utah. The study states that in more than 50 percent of the cases, the growth percentile model in these states, which includes assessment scores, is due "to random and unstable sources. ... Effectiveness estimates are unlikely to provide sufficient reliability for high stakes decisions, such as tenure or dismissal." This study further warns that states should be cautious is using such scores for teacher evaluations. Unless this situation changes, a lawsuit is inevitable.
Clearly, inaccurate year end assessment scores should not be a factor in determining teacher evaluations. House Bill 201, introduced in the Legislature this year and awaiting a standing committee hearing, is an attempt to accomplish this goal. Evaluations should be a fair reflection of what occurs in the classroom, not something based on random and unstable data.
Rep. Marie H. Poulson, D-Salt Lake City, represents District 46 in the Utah House of Representatives.