Less Assessment, More Learning

Academe, Nov/Dec 2003 by Barrington, Lowell W

Increasingly, popular "learning assessment" efforts in higher education distract professors from their essential work. It's time to rethink the assumptions behind these measures.

My university, like many colleges and universities around the country, is in the middle of a crisis. The crisis involves academic freedom, faculty morale, inefficient use of time and resources, and, most crucial, the nature of a liberal arts education. This crisis has emerged thanks to a seemingly innocuous, even valuable, goal: that of finding a way to "assess" how much our students are learning.

What's wrong with this endeavor? A lot, as it has developed in practice. It is beneficial for faculty to consider how they can improve student learning. Although most professors already do so on a regular basis, assessment is supposed to make such considerations more structured and regular. But it fails to improve teaching for two reasons: (1) it duplicates existing efforts while taking time away from activities more advantageous to student learning, and (2) it leaves professors with the unfortunate choice of either fabricating assessment data or teaching things that are easily assessed. Fabricating data is demoralizing at best, and teaching easily assessed subject matter is a direct attack on liberal arts education.

As a result of assessment, instructors spend much time on things like developing course-specific assessment plans, regrading assignments using assessment criteria (and assessment's unique language), and writing assessment reports. Such time could certainly be better spent on research or class preparation. Furthermore, champions of assessment have never indicated what activities faculty should spend less time on now that they are spending additional hours on assessment. Teaching? Research? Their families? Assessment supporters argue that the question is irrelevant, because once assessment is implemented, its time costs are minimal.

Whether or not that is true depends on your definition of "minimal." Anything that duplicates existing efforts without achieving the desired goal is an inefficient use of time. In my department, for example, we already have several policies in place to review faculty performance. Students evaluate us, faculty members sit in on each other's courses and suggest ways to improve them, and external reviewers examine the department and its approach to teaching. Most important, professors themselves think about ways to improve their courses.

The even more distressing consequence of these new assessment policies is the choice they offer to professors: produce misleading (or dishonest) assessment data or abandon the core ideals of liberal arts education. We have been told that assessment is being pushed mostly to address grade inflation. If three-quarters of the students at some universities graduate with 4.0 grade point averages, grades are meaningless; if that is so, we cannot rely on grades to judge what students have learned. Grade inflation is a problem. But assessment leaves the traditional grading system in place and adds a new (and even more flawed) system alongside it. Grade inflation will soon be joined by "assessment inflation," as many professors choose to mislead those in charge of reviewing the assessment data. It is distressing that no one who supports assessment understands this consequence.

An analogy may help clarify the problem. The Soviet system of economic planning resulted in economic inefficiency, partly because planners at the top needed to get information from factory managers and others below them. In this setting, managers had an incentive to deceive those above them. They said that they needed more raw materials than they really did and that they could produce fewer finished products than they really could. This deception gave factory managers a degree of security about their ability to "fulfill the plan."

Similarly, some professors may feel pressure to deceive academic administrators by generating data to fulfill the assessment plan. Although assessment data will not be used to evaluate whether or not a student should get into graduate school, the information can (and will) be used in decisions about salaries and teaching lines. In today's poor economy, universities are dealing with tight budgets. Pools for salary increases are smaller than in the past, and proposals for new faculty hires face greater scrutiny. In this climate, will professors or departments generate assessment data indicating they are doing a poor job? In many ways, academia already operates like the Soviet planning system-with its five-year plans and large, rigid bureaucracies. Given that the push at the federal level to require assessment has come mainly from conservatives, it is ironic that one of assessment's consequences is inefficient, Soviet-style planning.

Although assessment may force many professors to generate fraudulent or misleading data to prove that their students have learned something, that is not assessment's most dangerous consequence. Rather, it is that many other professors will not present inaccurate data. Instead, they will "teach to the tests." Taking seriously the seriously flawed process of assessment, these instructors will teach what is most easily measured. By imparting to their students a bunch of facts they may not remember a week after the semester ends, such professors will ensure that the "learning" of this "knowledge" can be demonstrated.