This column is the second of two dealing with the controversial “too much testing” going on in our public schools. This time, we’ll take a look at recent developments and trends in summative assessment.
Imagine going for an eye exam and being worried that the doctor will somehow punish you for not seeing well. Ridiculous, right? Well, that’s how we seem to treat students taking mandatory high-stakes assessments. As mentioned in the previous column, the goal of a valid summative assessment should always be to give educators the information they need to make decisions about how best to help all students learn.
I think the claim that too much testing is going on in schools is justified. For most school districts in Texas, local policy requires students be assessed periodically, an average of 4 – 6 times, on “benchmark” exams, which are used to predict how students will perform on the state mandated assessment. That practice is mostly a waste of time for those students who would likely meet the passing standard anyway (81% during 2015 for Belton ISD). But there is tremendous pressure on our public schools to operate under a capitalistic business model that demands metrics and accountability. So, educational leaders are forced to require these unnecessary periodic tests, and our teachers and students must comply.
But again, teachers should be allowed to be the professionals we expect them to be, and the annual state tests should simply provide the data they need to prepare curricula and guide instruction. Rather than being retained or kept from graduation for low scores on summative assessments, students should be rewarded for their achievements. My view, which is held by most assessment professionals, is that state-mandated tests should be used diagnostically to determine what is working in our classrooms and what needs to be improved, but never to punish students and teachers.
This year, for the first time, Texas schools will each be graded A – F, based primarily on how their students do on the STAAR exams, the annual state-mandated assessment. The STAAR tests are criterion-referenced, which means that the state has subjectively set a bar for students to clear. Most people are unaware that the state sets that bar low initially, and then moves it higher as students improve. In fact, based on previous performance by students, the State Board of Education already knows what percent of students will not “meet expectations.” That’s important, because it costs the state a lot of money to provide those students with extra resources and additional chances to pass the test.
When I was a student, my school district administered annual norm-referenced achievement tests. I suggest we take a strong look at rejecting the criterion model and returning to the older type of assessments. Rather than directly assessing the curriculum, achievement tests compare students’ abilities with one another, and give teachers and parents a quick snapshot of each student’s abilities and potential. We never studied or practiced for those tests. To me, that would have been like practicing for an eye exam.
Michael Brown is an education consultant and former teacher. He can be contacted at email@example.com.