I want to call attention to a paper that is quite related to this week’s readings:

It is actually discussed in more layman’s terms here (good for me as I am far from a psychologist).

We saw in most of the readings, and the TED talk, that the traditional carrot and stick method harms creativity and performance in anything other than mechanistic tasks. Also, grades force students to narrow their focus onto achieving high marks rather than learning, and can even average grades can discourage students expecting (especially if the peers do better). The Rogers et al. (2016) paper goes beyond that, suggesting that even without grades, seeing their peers do better on a graded assignment harms performance.

The authors actually ran two experiments for this paper. In the first, they collected grade data from a MOOC they offered. The course assignments included writing essays, as well as reading and grading their peers’ essays. This allowed the authors to track the “quality” of each individual as well as the quality of each peer essay read by each individual. They found that students exposed to excellent mid-term essays generally turned in less impressive final essays. These same students were even less likely to finish the course than those who graded poor peer essays. Their second experiment involved writing and grading SAT essay questions, and again those exposed to high quality writing reevaluated their own skill and were less confident in their abilities. Moreover, they were more likely to feel that essays were an unfair measure of skill, but still felt ashamed by their lack of success. The conclusion they drew was that even encountering superior work can have an incredibly demotivating effect.

To be honest, I have felt this a lot. Obviously imposter syndrome is a big part of graduate school, and we are bombarded with exceptional work all the time (“read this Nature paper while you’re trying to publish your masters work”). But I would suggest it is even worse in an interdisciplinary PhD program. Most of my colleagues have different backgrounds, and have many skills far beyond my own; I’ll never compete with the computer scientists in writing code or the statisticians in model design. Being forced to face this routinely, especially in our department seminars, is often crushing. I spend half of those seminars barely able to follow along, even when listening to junior colleagues – it took a year before I realized some of the other students were just as lost when I gave my talks.

Accordingly, even if we eliminate grades in favor of more detailed evaluations, the competitiveness and the feeling of self-doubt will still haunt many students. It makes no difference if it is a graduate seminar, or a grade-school arithmetic problem on the chalk board. I don’t know how we can get away with this.

Incidentally, about 2% of the 150,000 registered students completed the MOOC, which lends support to the opposite argument: a lack of incentive to finish (and penalty for failure to do so) totally guts motivation.

Second point, Alfie Kohn’s essay was fantastic. I want to agree with virtually every point, but as one might have guessed, I feel stirrings of pessimism. One must never forget that a student’s entire career depends on these grades. You can eliminate letter grades, and replace them with the analytical grid Dr. Elbow described, or you can use detailed assessments. All good in theory, but if an employer, college, grad school is going to base hiring or admissions decisions on those assessments, you can be assured that the student will focus on optimizing them, rather than learning.

In centuries past, it may have been the case that most students attended college pursuing a classical education simply for their own benefit. But today’s middle-class is generally built upon these degrees. The quality of the school and the “rank” of the student are literally of life-affecting importance. Until this changes, students will always prioritize ranking over learning. I don’t agree with Elbow that they want to be ranked, it is simply inevitable.