The Questionable Forms of Student Assessment

Executive Perspective

When I first started teaching many years ago in New York, we administered a standardized test and received results the following year. The delay meant the results had no value to teachers because those students were long gone and may have had as much as an additional year of instruction.

The results were useful at the school district level to assess districtwide performance compared to the norm. It was the teachers who developed the quizzes and tests that were used to inform instruction.

Things changed dramatically with the arrival of No Child Left Behind. No longer was it just students being assessed. Teachers, entire schools and districts were held accountable for the test results. Comparisons could be made of districts within states, schools within districts and teachers within schools. Teacher evaluations often were based on student performance on the tests taken, in many cases on reading and math performance for teachers who did not teach either subject.

Parents began to compare the performance of their district with neighboring districts. Comparisons also were made of schools within the same district. The effect on many districts was to focus on reading and math instruction at the expense of other subjects, resulting in a less comprehensive and rich curriculum. Subjects such as art, music and history suffered the consequences.

Predictable Results

Throughout that period, little attention was paid to per-pupil expenditures and the inequity in resources, factors that significantly affected student performance. Stories appeared about teachers categorized as exceptional one year to then be chastised as performing below expectations the following year when assigned to students with greater needs and fewer resources.

No Child Left Behind gained great support from the civil rights community because it exposed the very factors referenced above: inequity and lack of resources. It has been popular to say that you can predict a school’s performance on standardized tests by looking at the school’s zip code. Schools serving low-income students will in most cases underperform schools serving high-income students. Should we interpret that to mean that low-income students are not as capable as wealthier students? The gap in performance is confirmed by the test results but little is done to correct the inequities causing the gap.

Expecting all students to achieve above a certain level without considering the resources available was a major flaw of NCLB and thus many children continued to be left behind. A veritable rebellion on the part of educators led to the passage of the Every Student Succeeds Act that allowed states to consider factors beyond test scores to assess the effectiveness of schools. At least 50 percent of how states assess must include performance on a standardized test, but that leaves room for other forms of assessment that could focus on factors affecting student achievement such as social and emotional needs. Unfortunately, few states have taken advantage of that ESSA provision.

Another major development is affecting how and why students are assessed. Many universities recently stopped requiring students to take college-entrance exams. This has resulted in a significant reduction of students taking the SAT and ACT. The rationale behind this is the belief that entrance exams may be denying low-income and minority students the opportunity to receive a college education at the school of their choice. Some universities have reversed themselves and again are requiring the entrance tests because they find them necessary to determine whether applicants will be able to succeed with college-level courses, which is exactly what the tests supposedly measure.

Mastery Achieveable

The ongoing debates may have less to do with the assessment instruments and more with how they are used. Today’s assessments are developed using the sophisticated test and measurement practices required to ensure the test is valid and reliable. The tests consistently measure what they purport to measure. The low performance of a student may be accurate, but the test does not tell us why.

A standardized achievement test is not an intelligence test, but the results are affected by the student’s experiences and the resources available.

Assessments always will be part of the education process but perhaps it’s time to stop blaming the tests and rectify the factors that influence the results. We know that students can achieve mastery when they proceed at their own pace under the tutelage of a teacher who engages them in a system where no child is marginalized.

Dan Domenech is AASA executive director. @AASADan