The Costs of Overemphasizing Achievement
By Alfie Kohn
Only extraordinary education is concerned with learning; most is concerned with achieving: and for young minds, these two are very nearly opposite.
— Marilyn French
I. Common sense suggests we should figure out what our educational goals are, then check in periodically to see how successful we have been at meeting them. Assessment thus would be an unobtrusive servant of teaching and learning. Unfortunately, common sense is in short supply today because assessment has come to dominate the whole educational process. Worse, the purposes and design of the most common forms of assessment–both within classrooms and across schools–often lead to disastrous consequences.
Part of the problem is that we shy away from asking the right questions and from following the data where they lead. Instead, we fiddle with relatively trivial details, fine-tuning the techniques of measurement while missing the bigger picture. Take grading, for example. Much of the current discussion focuses on how often to prepare grade reports or what mark should be given for a specified level of achievement (e.g., what constitutes “B” work). Some educators have become preoccupied with the possibility that too many students are ending up with A’s.
From another perspective, though, the real problem isn’t grade inflation–it’s grades, which by their very nature undermine learning. The proper occasion for outrage is not that too many students are getting A’s, but that too many students have been led to believe that getting A’s is the point of going to school. Specifically, research indicates that the use of traditional letter or number grades is reliably associated with three consequences.
First, students tend to lose interest in whatever they’re learning. As motivation to get good grades goes up, motivation to explore ideas tends to go down. Second, students try to avoid challenging tasks whenever possible. More difficult assignments, after all, would be seen as an impediment to getting a top grade. Finally, the quality of students’ thinking is less impressive. One study after another shows that creativity and even long-term recall of facts are adversely affected by the use of traditional grades.
The data to support these findings are available to anyone who cares to look, and the practical problems of eliminating grades–including the challenge of helping parents understand the benefit to their children of doing so–are solvable for anyone who is committed to the task. That commitment, however, entails some serious reflection about why we are assessing students in the first place.
If we are primarily interested in collecting information that will enhance the quality of learning, then traditional report cards are clearly inferior to more authentic models. Unhappily, assessment is sometimes driven by entirely different objectives–for example, to motivate students (with grades used as carrots and sticks to coerce them into working harder) or to sort students (the point being not to help everyone learn but to figure out who is better than whom). In either case, the project is doomed from the outset, not because we haven’t found the right technique but because there is something fundamentally wrong with our goals.
The practice of sorting children is accomplished not only by grades (the most egregious example being grading on a curve) but also by standardized testing. So-called norm-referenced tests, like the Iowa and Comprehensive Tests of Basic Skills, are not intended to tell us whether teaching and learning have been successful. They are designed not to rate but to rank, to artificially spread out students’ scores. Not only are the results reported in relative terms (rather than assessing how well each did according to a fixed standard), but the questions on the tests have been selected with that purpose in mind. The test designers will probably toss out an item that most students manage to answer correctly. Whether it is reasonable for students to know the answer is irrelevant. Thus, to use a test like the ITBS to gauge educational quality, as assessment expert W. James Popham recently remarked, “is like measuring temperature with a tablespoon.”
Standardized tests often have the additional disadvantages of being (a) produced and scored far away from the classroom, (b) multiple choice in design (so students can’t generate answers or explain their thinking), (c) timed (so speed matters more than thoughtfulness) and (d) administered on a one-shot, high-anxiety basis.
All of these features represent the very opposite of meaningful assessment. But that doesn’t mean these tests are irrelevant to what goes on in classrooms. To the contrary, they have a very powerful impact on instruction, almost always for the worse. Teachers feel increasingly pressured to take time away from real learning in order to prepare students to take these dreadful tests. Some of this pressure originates from state capitals, of course. However, school district administrators often compound the harm by adding additional tests, sometimes those that are least informative (by virtue of being norm-referenced) and most destructive (by virtue of how teachers end up creating a dumbed-down, test-driven curriculum). All of this is done, of course, in the name of tougher standards and accountability, but, as any good teacher could tell you, the practical result is that the intellectual life is squeezed out of classrooms.
In fact, researchers could tell you this, too. In a study conducted in Colorado, some 4th-grade teachers were asked to teach a specific task. About half were told that when they were finished, their students must “perform up to standards” and do well on a test. The other teachers, given the identical task, were invited simply to “facilitate the children’s learning.” At the end, all the students were tested. The result: Students in the standards classrooms did not learn the task as well.
Why? For one thing, when teachers feel pressured to produce results, they tend to pressure their students in turn. That is exactly what was found in a second study, conducted in New York. Teachers became more controlling, removing virtually any opportunity for students to direct their own learning. Since people rarely do their best when they feel controlled, the findings of the Colorado experiment make perfect sense: The more teachers are thinking about test results and “raising the bar,” the less well the students actually perform–to say nothing of how their enthusiasm for learning is apt to wane.
II. The implications of taking seriously these concerns about grades and tests obviously would be enormous. But even this critique doesn’t get to the bottom of what’s wrong with the current approach to assessment. The underlying problem concerns a fundamental distinction that has been at the center of some work in educational psychology for a couple of decades now. It is the difference between focusing on how well you’re doing something and focusing on what you’re doing.
Consider a school that constantly emphasizes the importance of performance! results! achievement! success! A student who has absorbed that message may find it difficult to get swept away with the process of creating a poem or trying to build a working telescope. He may be so concerned about the results that he’s not all that engaged in the activity that produces those results. The two orientations aren’t mutually exclusive, of course, but in practice they feel different and lead to different behaviors. Without even knowing how well a student actually did at a task or how smart she is supposed to be, we can tell a lot just from knowing whether she has been led to be more concerned about layers of learning or levels of achievement.
Doesn’t it matter how effectively students are learning? Of course it does. It makes sense to sit down with them every so often to figure out how successful they (and we) have been. But when we get carried away with results, we wind up, paradoxically, with results that are less than ideal. Specifically, the evidence suggests that five disturbing consequences are likely to accompany an obsession with standards and achievement:
1. Students come to regard learning as a chore. When kids are encouraged constantly to think about how well they’re doing in school, the first casualty is their attitude toward learning. They may come to view the tasks themselves–the stories and science projects and math problems–as material that must be gotten through. It’s stuff they’re supposed to do better at, not stuff they’re excited about exploring. The kind of student who is mostly concerned with being a top performer may persevere at a task, but genuine interest in it or excitement about the whole idea of learning often begins to evaporate as soon as achievement becomes the main point.
This is related to the discovery by psychologists that intrinsic motivation and extrinsic motivation tend to be inversely related: The more people are rewarded for doing something, the more they tend to lose interest in whatever they had to do to get the reward. Thus, it shouldn’t be surprising that when students are told they’ll need to know something for a test–or, more generally, that something they’re about to do will count for a grade–they are likely to find that task (or book or idea) less appealing in its own right.
2. Students try to avoid challenging tasks. If the point is to succeed rather than to stretch one’s thinking or discover new ideas, then it is completely logical for a student to want to do whatever is easiest. That, after all, will maximize the probability of success–or at least minimize the probability of failure.
A number of researchers have tested this hypothesis. Typically, in such an experiment, kids are told they’re going to be given a task. Some are informed that their performance will be evaluated while others are encouraged to think of this as an opportunity to learn rather than to do well. Then each student is given a chance to choose how hard a version of the task he or she wants to try. The result is always the same: Those who had been told it’s “an opportunity to learn” are more willing to challenge themselves than are those who had been led to think about how well they’ll do.
It’s convenient for us to assume that kids who cut corners are just being lazy because then it’s the kids who have to be fixed. But perhaps they’re just being rational. They have adapted to an environment where results, not intellectual exploration, are what count. When school systems use traditional grading systems–or, worse, when they add honor rolls and other incentives to enhance the significance of grades–they are unwittingly discouraging students from stretching themselves to see what they’re capable of doing. It’s almost painfully ironic: School officials and reformers complain bitterly about how kids today just want to take the easy way out. . . while simultaneously creating an emphasis on performance and results that leads predictably to that very outcome.
3. Students tend to think less deeply. The goal of some students is to acquire new skills, to find out about the world, to understand what they’re doing. When they pick up a book, they’re thinking about what they’re reading, not about how well they’re reading it. Paradoxically, these students who have put success out of their minds are likely to be successful. They process information more deeply, review things they didn’t understand the first time, make connections between what they’re doing now and what they learned earlier, and use more strategies to make sense of the ideas they’re encountering. All of this has been demonstrated empirically.
By contrast, students who have been led to focus on producing the right answer or scoring well on a test tend to think more superficially. Consider just one of dozens of studies on this question, which concerns the ability to transfer understanding–that is, to take something learned over here and apply it to a new task or question over there. As a group of 8th graders were about to begin a week-long unit in science class, researchers gauged whether each student was more interested in understanding or in being successful. When the unit was over, the students were tested on their ability to transfer their new knowledge. Regardless of whether their earlier test scores had been high or low, the success-oriented students simply did not do as well as those who were more learning-oriented.
4. Students may fall apart when they fail. No one succeeds all the time, and no one can learn very effectively without making mistakes and bumping up against his or her limits. It’s important, therefore, to encourage a healthy and resilient attitude toward failure. As a rule, that is exactly what students tend to have if their main goal is to learn: When they do something incorrectly, they see the result as useful information. They figure out what went wrong and how to fix it.
Not so for the kids who believe (often because they have been explicitly told) that the point is to succeed–or even to do better than everyone else. They seem to be fine as long as they are succeeding, but as soon as they hit a bump they may regard themselves as failures and act as though they’re helpless to do anything about it. Even a momentary stumble can seem to cancel out all their past successes. When the point isn’t to figure things out but to prove how good you are, it’s often hard to cope with being less than good.
Consider the student who becomes frantic when he gets a 92 instead of his usual 100. We usually see this as a problem with the individual and conclude that such students are just too hard on themselves. But the “what I’m doing” versus “how well I’m doing” distinction can give us a new lens through which to see what is going on here. It may be the systemic demand for high achievement that led him to become debilitated when he failed, even if the failure is only relative.
The important point isn’t what level of performance qualifies as failure (a 92 versus a 40, say). It’s the perceived pressure not to fail, which can have a particularly harmful impact on high-achieving and high-ability students. Thus, to reassure such a student that “a 92 is still very good” or that we’re sure he’ll “do better next time” doesn’t just miss the point; it makes things worse by underscoring yet again that the point of school isn’t to explore ideas, it’s to triumph.
5. Students value ability more than effort. How do we react when a student receives a score of 100 on a quiz? Most teachers and parents treat that as news worth celebrating. Those who are more thoughtful, by contrast, are not necessarily pleased. First of all, they will be concerned about the “bunch o’ facts” approach to instruction and assessment that may be reflected by the use of traditional quizzes. Even successful students are not well-served by such teaching.
But even when better forms of assessment are used, perceptive observers realize that a student’s score is less important than why she thinks she got that score. Let’s ask how a student might explain doing especially well on a test. One possibility is effort: She tried hard, studied, did all she could to learn the material. A second possibility is ability: If you asked her how she got a hundred, she might reply (or think), “Well, I guess I’m just smart.” Yet another answer is luck: She believes she guessed correctly or was just having a good day. Finally, she might explain the result in terms of the level of task difficulty–in this case, the fact that the test was easy. (Notice that these same four reasons could be used by another student to make sense of his grade of 23 on the same quiz: I didn’t try hard; I’m just stupid; it was bad luck; or the test was difficult.)
Which of these four explanations for doing well (or poorly) do you favor? Which would you like to see students using to account for their performance in school? Almost everyone would vote for effort. It bodes well for the future when kids attribute a good score to how carefully they prepared for the test. Likewise, those who attribute a low score to not preparing for the test tend to perceive failure as something they can prevent next time. So here’s the punch line: When students are led to focus on how well they are performing in school, they tend to explain their performance not by how hard they tried but by how smart they are.
Researchers have demonstrated that a student with a performance focus–How am I doing? Are my grades high enough? Do I know the right answer?–is likely to interpret these questions “in terms of how much ability [he or she has] and whether or not this ability is adequate to achieve success,” as educational psychologist Carol Dweck and a colleague have explained. In their study of academically advanced students, for example, the more that teachers emphasized getting good grades, avoiding mistakes and keeping up with everyone else, the more the students tended to attribute poor performance to factors they thought were outside their control, such as a lack of ability. When students are made to think constantly about how well they are doing, they are apt to explain the outcome in terms of who they are rather than how hard they tried.
Research also demonstrates that adolescents who explain their achievement in terms of their intelligence tend to think less deeply and carefully about what they’re learning than do those who appeal to the idea of effort. Similarly, elementary school students who attribute failure to ability are likely to be poorer readers. And if children are encouraged to think of themselves as “smart” when they succeed, doing poorly on a subsequent task will bring down their achievement even though it doesn’t have that effect on other kids.
The upshot of all this is that beliefs about intelligence and about the causes of one’s own success and failure matter a lot. They often make more of a difference than how confident students are or what they’re truly capable of doing or how they did on last week’s exam. If, like the cheerleaders for tougher standards, we look only at the bottom line, only at the test scores and grades, we’ll end up overlooking the ways that students make sense of those results. And if we get kids thinking too much about how to improve the bottom line, they may end up making sense of those results in the least constructive way.
If all of this seems radical, it is–in the original, Latin sense of the word radical, which means “of the root.” Indeed, cutting-edge research raises root questions, including the possibility that the problem with tests is not limited to their content. Rather, the harm comes from paying too much attention to the results. Even the most unbiased, carefully constructed, “authentic” measure of what students know is likely to be worrisome, psychologically speaking, if too big a deal is made about how students did, thus leading them (and their teachers) to think less about learning and more about test outcomes. As Martin Maehr and Carol Midgley at the University of Michigan have concluded, “An overemphasis on assessment can actually undermine the pursuit of excellence.” That’s true regardless of the quality of the assessment. Bad tests just multiply the damage.
Most of the time students are in school, particularly younger students but arguably older ones too, they should be able to think and write and explore without worrying about how good they are. Only now and then does it make sense for the teacher to help them attend to how successful they’ve been and how they can improve. On those occasions, the assessment can and should be done without the use of traditional grades and standardized tests. But most of the time, students should be immersed in learning.
For a more detailed exploration of the difference between focusing on performance (achievement) and learning, as well as citations to the relevant research, please see chapter 2 of The Schools Our Children Deserve.