The troubling question of grades

I wasn't in charge of a college course until I was 62.  So please forgive me for yapping about something that many readers of this blog have thought about for years.

As I was grappling with the problem of grading my students ("How much should they be able to do in order to rate a "B+?"), I talked with my old friend Tom Angelo, who reassured me that grading was the shameful secret of teaching. 'No one can really tell you how to do it properly.' For example, faculty will disagree forever on such profound questions as "Are there many courses and circumstances in which it would be appropriate for all students in a large class to earn an A?"  At GW and many other institutions many faculty are upset about 'grade inflation' (such a large fraction of students being given A's or B's.)

Against this backdrop, here are three pieces of information.

The first I encountered years ago: there is virtually no relationship between undergraduate GPA and anything about the student's experience in life after college (except for graduate school grades). Grades don't predict job success, voting behavior, likelihood of winning a Nobel Prize, or anything else.

• 1984      Cohen, Peter, “College Grades and Adult Achievement: A Research Synthesis,” Research in Higher Education, XX:3, 281-293.

I've only seen one study of GPAs in professional schools - it was about a business school - and they didn't find any relationship between an MBA student's grades and later success in business, either.

The second piece of information comes from  Ken Bain's classic book from 2004, What the Best College Teachers Do.   On page 32, he writes (I'm paraphrasing) , "For years, psychologists have studied what happens when someone has a strong intrinsic interest in something, and they're also offered an extrinsic reward to do it (e.g., a high grade, money, etc.).  Would that reward increase their fascination with the subject, leave it unaffected, or decrease it?"

In one set of experiments by Deci and others, students were asked to play an unfamiliar game under supervised conditions.  Some students were paid money for winning, while the others received no rewards for their performance.  When left 'alone,' the students who'd been competing for prizes stopped playing the game. In contrast, students who had not played for any reward often continued playing after the supervisor left.  Variations of this experiment showed that, in order to provide external feedback while doing minimal damage to the student's own interest in the subject, it's best to combine feedback about performance 'Here's what you did right; here's how do it better.') with selective praise for performance ('The way you castled your pieces was quite ingenious!').

Bain reports that decades of research by different investigators indicate that, 'when given a reward for good performance, students' own interests in the subject actually go down and sometimes disappear.'

But why should instructors worry about whether grading is sapping the student's own interest in the subject if students are doing well enough in the course?

If literature students no longer love novels so much, or if engineering students lose their taste for creating things, they're less likely to keep using their academic learning on their own.  We're trying to help students develop their 'thinking muscles.'   If students leave a course (still) loving what they've learned, they're more likely to find ways to use that capability and, in the process, get even better.  (If the receipt of a final grade also stops the use of those thinking skills, on the other hand, it seems possible that those skills will atrophy.)

Those great college faculty that Bain studied do everything possible to increase the students' intrinsic motivation, and tap that energy.  That fits what I've seen over the years.  For example, I've written about Jon Dorbolo's experience as a novice teacher of introductory philosophy.  When his first course went sour, Dorbolo asked his students to rate every assignment anonymously. To his dismay, a majority of his students rated every single assignment useless and boring. However Dorbolo then noticed that every assignment was rated useful and interesting by a minority of students.  Different students liked different sets of assignments. Some students liked only religious philosophers.  Some liked only utilitarians.  Dorbolo realized that his most important goal was to help students learn the rudiments of philosophical thinking. And an implicit goal was to not teach them to hate philosophy.

So, by the next time Dorbolo taught the course, he'd created different, overlapping reading lists. Each set of assignments helped students learn skills of close reading and analysis.  But each student got to choose a set of readings most likely to motivate them to think hard, work hard, and appreciate what they were learning.  Dorbolo also converted the course from mostly lecture (since it made no sense to lecture about one reading when most students in the room hadn't seen it).  Instead class sessions were mainly devoted to small group discussion. Dorbolo saved a lot of time that would otherwise have been devoted to creating and delivering lectures.  Instead he focused on managing the discussions, assessing learning, and studying what helped and hindered student learning.

But what about grades?  Is it possible that grade inflation is a good thing, if it motivates us to provide feedback and documentation of student learning in other, less destructive ways?

Do you know of any faculty who've looked at research on assessing and valuing student performance? motivating learning? Tried any experiments?  If so, please add a comment.

PS. If you have any trouble posting comments on the blog, please email me.

Post slightly revised on May 19, 2012