A Selected Disagreement: Ranking Students

As people respond to the automated SBG persuader, I get anonymous emails with their disagreements.  Here’s a big one.  It is in response to the premise that “final grades would more accurately reflect current understanding if we could only use recent scores to calculate them.”

“How is that fair to the student that learned addition quicker?  Shouldn’t the grade ultimately rank my students according to their ability?”

Sure, pitting kids against each other until they’re tearing at each other’s eyes is fun. But is it right? To me, the answer is “obviously not.”  Is it effective as an educational tool? Obviously not.  Is it the principle off of which we want kids to model their lives?  Obviously not!

Is the point of education to figure out which kid is the best?  Say it with me: obviously fucking not.

Sorry.  I was being a little sarcastic up there, but let’s be reasonable.  The argument for competition in a classroom is that a competitive atmosphere is motivating and can raise the overall level of achievement.  What better way is there to get Sarah to raise her grade to a 95 than telling her that Rebecca has a 94?

The problem I have with this is that it focuses solely on the grade.  Ranking students by grade, so that they compete for grades, makes them care about the grade.  I want them to care about learning.  As Alfie Kohn says in The Schools Our Children Deserve, “The difference between learning and achievement is hard enough to grasp; the difference between doing well and doing better than others is especially confusing in a society so obsessed with being Number One that the ideas of excellence and winning have been thoroughly conflated.”

Now I’m going to slam you with some research that Alfie Kohn compiled.  Watch out.  Just read the bold sections unless you’re going to check my references.

  • Susan Nolen’s study titled “Reasons for Studying: Motivational Orientations and Study Strategies” concluded that students who equate success with surpassing others are more likely to think in a “surface-level” way.

  • Carole Ames published “Children’s Achievement Attributions and Self-Reinforcement: Effects of Self-Concept and Competitive Reward Structure” in the Journal of Educational Psychology  with a conclusion that students are more likely to feel empowered to affect their own achievements when those achievements are not linked to competitive results.

  • P.S. Fry and K.J. Coe concluded in “Interaction Among Dimensions of Academic Motivation and Classroom Social Climate: A study of the Perceptions of Junior High and High School Pupils” that competitive environments “cause students to dislike school and show less interest in a given subject.”

  • From David and Roger Johnson said in Cooperation and Competition: Theory and Research, Kohn concluded that when a group includes members of different ability levels, they “learn more effectively on a range of tasks when they’re able to cooperate with one another than when they’re trying to defeat one another.”

I’m glad that there’s research I can use to bolster my intuition.  I don’t think that competition should play a significant role in education, and, since grades are often viewed as the “conclusion” or “you know, the point” of school, I really don’t think that grades should be based on competition.

If you use competition in your class, I’d love to hear how you justify it.  Again, I’m open to the idea that I’m wrong, or missing a subtlety.  I hope my intro wasn’t so caustic that you aren’t reading this.  I guess that’s impossible, since you’re clearly reading this if you’re reading this.  But you know what I mean.

18 thoughts on “A Selected Disagreement: Ranking Students”

  1. Jo Boaler’s book, What’s Math Got to do With it?, also gives evidence and analysis about why having students of different abilities work together cooperatively helps all of them do better, and enjoy math more.

  2. It’s important to be aware of cultural differences, and where competition is applied. I’m a volunteer English teacher at a Chilean high school (long story), and Chilean kids *love* competition. It’s a *very* socially-oriented culture: Chileans form teams and support them aggressively at the drop of a hat (intramural competitions, for example). It’s a lot like the Drazi Green-Purple episode of “Babylon 5″ (“The Geometry of Shadows).

    I would never assign them a grade compared to each other, but as a technique to drive their interest in an activity, it’s great.

      1. I am also a privileged, white, American male teaching to small groups of easily interested kids. However, I think that it is worth pointing out that there is a difference between “students like to compete” and “competition helps students learn.”

        My college has done some internal research that concludes that our students like competition. However, my students are also exactly like the time of students studied in Johnson and Johnson’s meta-analysis of cooperative vs. competitive vs. individualistic learning. If memory serves me correctly, their results (based on 115 years of research) is that students learning cooperatively perform 0.88 standard deviations better than those learning competitively. This difference is huge.

        I think that we should be careful about the distinction between “like” and “need.” All things being equal, I think that we should accommodate our students’ likes as much as possible; in this case, I would like to see evidence that the Chilean students learn significantly differently from the students in Johnson and Johnson’s research.

  3. Riley,
    I love this post and agree 100%. But I have had colleagues who say “my job really is to figure out which kids are ready to take the next course.” (Which, in this case, is an AP science course in 1 year with no previous exposure.

    How do you address this to the separate the wheat from the chaff model of teacher, if you will. After all, they might say, “not everyone can be a mathematician, scientist or artist, and better to give them an indication of that early, right? “

    1. Sure, you shouldn’t pass a kid who can’t take the next course. When you think of it as separating the wheat from the chaff, you’re basing your decisions based on rank. Wouldn’t it be great if we could make that decision based on what each kid knows, instead?

      What do you think about the statement that “not everyone can be a mathematician, so let’s tell them early?” I’m somewhat inclined to agree, actually. If a student hasn’t learned to add, let’s either keep them in adding class or let them double up on other classes, you know? But all of these decisions can be made individually. We do not have to rank multiple kids in order to know if one of them can add.

      This brings up the amount of work we as teachers and as a system are able to do. We can’t give each kid an individualized experience (maybe). We have to group them by skill level. But can’t we define the skill levels and sort kids based on SKILLS, instead of sorting on PERCENTILES?

  4. I think that small amounts of competition (in-class games) is motivating, but high-stakes competition (like curving grades) is corrosive. I think that pre-med programs at some schools illustrate the problems of high-stakes competition, with rampant cheating and even vandalism of other students’ work.

    Grading should reflect how well the students are preforming at the end of the course and be predictive of how well they will do in future. I have yet to see a study that shows SBG as doing better at that task than other grading systems, though the theoretical justifications for SBG suggest that it should.

    I do have problem with your example about the student who learned addition quicker. If one student consistently gets the material months before another student, and at the end of the year the two students end up at about the same place, then you have seriously shortchanged the faster student, holding them back so that the other student could catch up. If the two students do end up with different levels of mastery, it seems reasonable that the one who is further ahead in learning should get a higher grade to reflect that difference. Otherwise the grades are pretty meaningless.

    1. When one student is consistently faster than another student, you won’t be able to engage both students meaningfully for the whole year if you are determined that they should learn the same set of things. I don’t think SBG tries to address that problem. The grades don’t lose their meaning, though. Even when all students get an A, the A means “this student mastered the skills set forth in this syllabus” or whatever.

      What advantages are there to moving ahead with the faster student, give him an A, and the slower student a C? In this scenario the A means “this student mastered the skills set forth in this syllabus,” and the C means “this student showed some proficiency in the skills set forth in the syllabus.” The course was still incorrectly calibrated for one of the students. The grades mean the same thing. Can you elaborate?

      1. In my courses, a B means that students have basically mastered the required content of the course, and an A means that the student has gone beyond basic skills, either by doing everything extremely well or by going beyond the content of the course.

        Most of the SBG standards I’ve seen have been aimed at getting all students to a “proficient” level, rather than to an “advanced” level. This is realistic if you expect all students to master all standards—the stated goal of most SBG practitioners. For my interpretation of grades, mastering all the standards in such a course is a B, not an A. If the standards are written so that mastery of all is *not* expected of most students, then SBG might produce the sort of calibration I expect.

        My courses always have somewhat open-ended assignments, that students can go beyond the requirements in ways that I can’t anticipate ahead of time, making it rather difficult to capture the description of an A in SBG-style standards.

        1. How does a student in your class earn his grade? I ask because it seems like the difficulty of attaining the top grade in SBG could be adjusted just like the difficulty of attaining the top grade in a conventional class. For example, on a five point scale, you could use 4 to mean “master of expected skill level” and 5 to mean “went further than I expected.” Or just grade each standard out of 100 and do your grading in precisely the same way – just lump the grades together by topic instead of by assignment type.

          1. In the class for which I’m considering SBG this fall, I currently grade the students on 9 assignments: 6 computer programs and 3 papers. The three papers are a fellowship application, a report on building molecular models (where the correct construction of the models is the goal), and a library/bioinformatic research paper, which requires both library research and use of bioinformatic tools.

            Each of the programs tests 3 things: programming skill, documentation skill, and the bioinformatic concepts and algorithms being implemented. I’ve always provided feedback on all three (to the extent that I can—bad documentation skill can make it impossible to determine the other two), but I’ve used a letter grade to summarize my overall impression of the program. I’ve been trying to decide whether splitting the grade would be beneficial to the students.

    2. That’s a course problem not a grading problem. If you have a kid who mastered all the stuff in the first month of class, they shouldn’t be in the class. If they have to be in your class, that doesn’t mean you make them the A and move everyone down.

      “If two students do end up with…etc” No, it’s not reasonable because you’re assessing students on their mastery of course content/standards/proficiency levels of whatever you want to call it. You’re going back to norm based if you establish your highest achiever as your top score.

      If you’ve got kids who are going above and beyond, congratulations, you’re a good teacher. I have yet to see anywhere on this blog that says you should hold back the faster students so other students can catch up. If “A” means above and beyond (which I agree with) then all students who are above and beyond get the A. There is no Super Duper A or whatever. Ranking kids in K-12 is crap and doesn’t need to be done.

  5. Some competition is healthy but not a determiniation for grades. I had a professor in college who used the bell curve for his class. It didn’t matter if you truely performed at an A level if you fell in the bottom 50% of the class. Eventually the students in his class slacked and realized they didn’t have to do their best, just better than the rest of the class.

    Also, Malcolm Gladwell provides some great research in his book Outliers about a cap on intellegence. He says that smart is smart and once you reach a certain level, there is no indication that someone with a higher IQ will be more successful. This informaion has greatly influnced how I view my honors classes.

  6. I’m at a university, but I’m trying to move towards my expectations being about a length of time for work, and supporting my students in using that time effectively. Rather than a set of exercises, which is probably only suitable to a middle slice of the students.

    For the faster students, we are cheating them if we don’t open the door for them to do more. Whether that more be quantity of exercises, challenge of problems, or opportunity to work on communication or representation.

    Obviously different teachers have different purposes for grades, but I like mine to be feedback for students if they are successfully meeting objectives.

    1. Length of time is good for calibrating how much work to give the students, but is a bit risky for any sort of grading. Giving faster students lower grades for better work (a consequence I’ve seen sometimes called “holding them to a higher standard”) is perceived by the students as punishment for being smart, and results in them losing respect for the teacher and deliberate dumbing-down.

  7. “Is the point of education to figure out which kid is the best?”

    That depends on who you ask. Many organizations (post secondary institutes and employers chief among them) do want students ranked on the basis of their ability. For them, one of the points of education is to figure out which kid is best.

Leave a Reply