The dilemma of standardized testing and the achievement gap
Since No Child Left Behind was passed in 2001, trying to close the achievement gap has been on every educator’s mind.
Key to that law has been the requirement of measuring achievement through the administration of standardized tests to determine the extent to which schools are making “adequate yearly progress” (AYP) toward that goal.
NCLB is now being dismantled, but its impact will be felt for a long time to come. No one, I imagine, ever anticipated that standardized testing would become such a juggernaut, or that schools would become test-prep factories. Thus we have the dilemma of being able to define the achievement gap only as a disparity in standardized test scores between minority students and their white counterparts. That bit of information makes good fodder for political speeches, but it does precious little to aid in improving teaching and learning.
It isn’t difficult to understand the appeal of standardized tests to those seeking educational data. There is a definite allure to the efficiency by which such tests can be administered simultaneously to large numbers of students and then quickly scored. It’s a process that traces its roots to the tests given to new recruits when the United States entered World War I.
What was seen as a need for military expediency in 1917, however, has become entrenched in our society, and the collection of such data has exploded in its frequency, in its undue influence on the curriculum, and in its use for making life-impacting decisions about children, teachers, administrators, and schools.
For school children, the stakes increase dramatically when standardized test scores are used as the primary determinant for placement in classes or programs, promotion from grade to grade, and for graduation from high school. And for teachers and school administrators, the hot button issues are tying test scores to teacher evaluations and the consequences of failing to achieve AYP goals.
If standardized tests are good at anything, it’s measuring the achievement gap. That’s because standardized tests allow educators to compare large populations of students, with the capability of disaggregating the data by demographics (race, economic status, gender). The human element means no standardized test is completely reliable. Students’ performance will vary on any given day due to factors which cannot be replicated or standardized—what they ate for breakfast, how much sleep they got, how they interpret questions, and their ability to stay focused for long periods of time. The relatively stable information we are able to obtain about large populations of students becomes useless when we consider individuals and small groups.
In all of the noise about the “achievement gap,” what has been missing is a conversation about what “achievement” itself might mean. Standardized testing has—intentionally or not—imposed a de facto definition of achievement as “the score on a standardized achievement test.” Let me posit that this is a pretty silly definition. But silly or not, it has had the effect of narrowing the curriculum to those bits that can be captured on a bubble sheet or in a short essay.
There are, of course, other models of assessment: portfolios, protocols for analyzing student work samples, projects scored through the use of a common rubric, and opportunities for student self-reflection. At the very least, these measures could be structured so as to confirm or refute the limited data that standardized tests are able to provide. Teachers already use these kinds of assessments to make immediate informed decisions about their students so they can adjust their teaching accordingly. Standardized tests lie outside of the teaching and learning process, and their use represents a disruption and an imposition.
Perhaps it is time to examine what we actually mean by “academic achievement.” Maybe if we can come to an agreement on that, we’ll be ready to consider how best to measure learning. Then, perhaps, we’ll finally be ready to tackle the achievement gap. Until then, we’re just spinning our wheels with tools that do not do what we need them to do—which is to provide useful and comprehensible information for educators, parents, and students.
Howard M. Miller is chair of the Department of Secondary Education at Mercy College School of Education.