From the Director: My semester with Specifications Grading
By Dr. Ann Johnson, Director of the Center for Faculty Development
For all of my twenty-plus years of teaching I’ve relied on a traditional point-based grading system. I used it because it was familiar and because most other professors I know use it; only recently did it occur to me to link my point system to my own frustration with student reactions to grades in my courses. I’d grown weary of all the small arguments over points on assignments and tests, and had a nagging sense that many students were more strongly motivated to ‘game the system’ than actually work toward my learning goals. And I didn’t like the way that grading seemed to undermine my relationships with students. Math professor and columnist Robert Talbert summarized perfectly my un-ease in this area: “[T]his points-obsessed, game-playing mentality feeds back into my relationships with students. I’m no longer the consultant to their client, but a Dungeon Master who rolls the dice and deals out the points — and I’m an obstacle to be overcome rather than a resource to be tapped. Tell me whether that’s a productive, healthy working relationship.”
So I was intrigued when I ran across education expert Linda Nilson’s new book, Specifications Grading: Restoring Rigor, Motivating Students, and Saving Faculty Time. Nilson’s approach reinforces many of the implicit goals we all try to incorporate in teaching, like motivating students to learn and excel, and making students feel responsible for their grades. By getting rid of points, it promises to reduce student stress, save faculty time, and minimize conflict between teacher and student.
How does it work? Nilson revives elements of the old “contract grading” concept popular in the 1970s and combines it with a pass-fail rubric system for evaluating assignments. The pass-fail structure for grading may be the most controversial aspect of Nilson’s plan because we’ve come to think of “passing” as mastering only the most minimal or mediocre set of requirements. In Nilson’s system, the bar is set much higher: She recommends building rubrics that require B+ level work. The student’s final grade for the course reflects the number and type of assignments and tests completed (this is where the plan resembles contract grading). Rubrics communicate an expectation that work submitted will be high quality. If it does not reach the passing bar, it fails.
If you’ve read this far, you probably have questions. This is just the skeletal structure of Nilson’s system -- there are nuances and details too numerous to explore here (see links below for more information, or contact me to check the book out from our Faculty Development library. For instance, if a student fails an assignment, all is not lost. In this system students receive a number of tokens (you determine how many) that can be used to ‘buy’ an opportunity to re-do an assignment (tokens can be used for other purposes, too). The system is not draconian, but the pass-fail aspect definitely creates a high-stakes atmosphere and I’ve found that this yields consistently higher quality work.
This semester I’m teaching an upper level undergrad psychology course and I decided to experiment with Specifications Grading. The first thing I learned was that it requires a good deal of up-front preparation. All assignment descriptions and rubrics must be available on day one. When I described the plan at our first class meeting, I was met with considerable skepticism and some anxiety. Students asked a lot of questions and were relieved (somewhat) to hear about the tokens. When I checked in with them about three weeks later, however, I found I had won quite a few converts.
My course meets twice a week and students hand in short, written reading responses at each session. I’ve developed brief guidelines for these, and one of the criteria for passing is “all guidelines are followed.” This requirement addresses a frequent phenomenon: the student who hands in slapdash and incomplete work, hoping to get partial credit. A few students tried this approach early on, but after one ‘fail’ they quickly adapted and I’ve seen an impressive increase in quality.
For longer reflection papers (3-4 pages), rubrics allow no more than 5 spelling/grammatical errors; I also build in specific outcomes that I want to see. I will admit: Issuing a “fail” verdict feels harsh, and I always hesitate to do it. But I’m learning here, too – several students have responded by quickly setting up meetings with me where we can talk over what happened and how to avoid it on the next assignment. I have seen dramatic improvement in some cases and feel more like I’m a collaborator in their success. I’ve become a “resource to be tapped” and have learned the value of clear expectations and consequences.
The system is not universally loved by my students (does that ever happen?) but many have replaced their skepticism with enthusiasm. I recently administered a mid-semester evaluation via Qualtrics and received overwhelmingly positive marks for “organization and clarity.” One student commented: “I thoroughly enjoy having clear and concise expectations of me.” Another said, “The opportunity to be successful and create your own grade in this course is something I really enjoy.”
One last observation: Nilson’s system really delivers on the promise to reduce faculty time spent grading. I was curious about this because of the many, many complaints I hear from faculty about time spent grading. I still include personalized notes while grading student work because I like the personal touch, but overall, the simplified rubric has streamlined my grading considerably.
I’m still gathering student feedback but my mid-semester impression is strongly positive. I will tweak some of my assignments and rubrics next time (live and learn) but look forward to refining and repeating this grading scheme in future courses.
Curious? Nilson’s book provides many examples of assignments, grade contracts, and rubrics from a variety of disciplines, and I’d be happy to share the ones I’ve developed. The links below take you to articles written by others who have tried this approach.
- Interview with Linda Nilson about Specifications Grading
- Two Chronicle of Higher Education blog posts by Robert Talbert on implementing Specification Grading: Thinking about grading and Putting grades to work.