Grades – A Measure or a Rank

Is a course grade a measure of a student’s learning? Is a course grade merely a way to rank students for whatever gateway comes next? Can a course grade be both and do a good job at each task?

Recently, a good colleague sent me a link to a New York Times Education article on grading at University of North Carolina, Chapel Hill. The article begins like this…

It could be a Zen koan: if everybody in the class gets an A, what does an A mean?

The answer: Not what it should, says Andrew Perrin, a sociologist at the University of North Carolina, Chapel Hill. “An A should mean outstanding work; it should not be the default grade,” Mr. Perrin said. “If everyone gets an A for adequate completion of tasks, it cripples our ability to recognize exemplary scholarship.”

The article author Tamar Lewin goes on to explain that UNC is considering adding more context to its grading – a median average to accompany the mean average, as one example. In my opinion, the more context that can be added to a grade, the better! A grade devoid of meaningful context is not worth much. Why? My answer: because grades mean so many different things to different people.

Just read the article to get a snapshot of the confusion surrounding grades. I am not sure that such a thesis was intended by the author, but that is what the article concisely points out for me. For the schools that declare something like “no more than 35% of students can receive A’s,” they are clearly siding that grades are merely rankings. Relative achievement numbers so that the students can be lined up against a wall from “best to worst” – at least as well as such can be done with the likely imperfect instruments used to evaluate student learning (in theory).

But shouldn’t grades be real measures of learning? In the ideal, shouldn’t a grade give a learner an indication of how he/she performed relative to a clearly articulated standard…a mutually visible quality criteria (mutual in that the teacher and the student can see the same target equally well)? In this case, a grade is an absolute achievement number, not a relative achievement number. Such a grade would indicate to a student how he/she measured against a clear learning goal – NOT how he/she measured simply relative to other people.

What if Mr. Perrin taught a seminar course of 20 students, and all twenty students happened to be the “best” 20 students on campus? (Just go with me for a minute.) If Mr. Perrin had established and made known to his students (better yet…established with his students) the clear learning targets, AND if the 20 students had all achieved at the highest levels according to those clearly established targets, THEN shouldn’t all 20 get “As?”

Granted, if all 20 got As, it would make it very difficult for a receiving institution to know which one of the 20 students was “the best of the best.” But don’t grades fall short of their true meaning and intent when we use them as simple, imperfect rankings? If used for their true meaning and intent, shouldn’t the grades be for the primary purpose of the RECEIVER of the grade?! Should not the grade be an indicator to the receiver – the earner – of the grade that “Oh, I understand this material and this set of skills – the grade indicates that to me’? Or should the grade just be…”Oh, I am ranked 8th of 20 in this course. I was not in the top 35%. I must not know the material that well”? What if all of the top 8 MASTERED the content targets…the course objectives…the quality criteria? Is it just “too bad” for that 8th student?

Maybe the course instructors don’t really know what the clear targets are. Or maybe they don’t know how to articulate clear targets to the students. Or maybe they just didn’t articulate any targets to the students. Or maybe the targets are complex enough that simple numerical averages are not enough to communicate one’s understanding of the material and skills. Does my weight, alone as a number, indicate my level of health? Does my BMI, as a stand-alone number? Or would more context be helpful?

More context of the learning objectives is needed. Kudos to UNC for realizing that more context is needed. I hope Mr. Perrin considers that grades might be best used for communicating to the learner rather than to the next gateway through which the learner must pass. No wonder students “grade grub.” If the grade is just a relative benchmark for lining me up so that I can be “assessed” for the next program, then I would grade grub too! I am a learner – why would I not learn such a behavior if the grade just indicates a ranking. To grub a point is the most simple way to increase my rank – a decent plan A to move up the ranks. How can we blame students for such a behavior if we create the system that promotes such behavior?

BUT…if grades were a measure of my learning compared to clear targets, then “grade grubbing” might come more in the form of such questions as, “What did I not know as well on this content?” “How could I learn it better next time?” I love it when I hear these types of questions!

11 thoughts on “Grades – A Measure or a Rank

  1. Pingback: SBG Gala #5 « Quantum Progress

  2. From an article by Alfie Kohn:

    * Never grade on a curve. The number of good grades should not be artificially limited so that one student’s success makes another’s less likely. Stipulating that only a few individuals can get top marks regardless of how well everyone does is egregiously unfair on its face. It also undermines collaboration and community. Of course, grades of any kind, even when they are not curved to create artificial scarcity — or deliberately publicized — tend to foster comparison and competition, an emphasis on relative standing. This is not only destructive to students’ self-esteem and relationships but also counterproductive with respect to the quality of learning (Kohn 1992). As one book on the subject puts it: “It is not a symbol of rigor to have grades fall into a ‘normal’ distribution; rather, it is a symbol of failure: failure to teach well, to test well, and to have any influence at all on the intellectual lives of students” (Milton et al. 1986, p. 225).

    Full article: http://www.alfiekohn.org/teaching/grading.htm

  3. Pingback: Grade Reporting: To comment or not to comment…that is the question! | Experiments in Learning by Doing

  4. Pingback: A Few of my Favorite Blogs « 21k12

  5. Bo:

    I think your blog posting on grades hit the mark. You told the truth. I think we (Educators) need to get our heads out of the sand on this question. The world will pass us by if we continue to use grading to “rank” students. The truth is that is what we do grading for. Sit around the cafeteria table and listen to most teachers talk about grading, tests, comments, etc. It is ALL about ranking for the majority of them. I don’t mean to point fingers, but really. If we are interested in learning, deep understanding of the learning targets (see my post on what is understanding at http:\\rryshke.wordpress.com), then we have to transform our approach to assessment and grading. I am thinking more about the MASTERY approach to learning. If we are interested in student learning, why wouldn’t we be interested in mastery. I think the reason is because we have structured school to be 36 weeks, seven-hour days, and 50 minute periods. As a result not all students can master the learning targets in the time slot they are given and therefore our assessment and grading promotes ranking. The system is not FRIENDLY to the learner. So Now what?

    How does this conversation get traction in most schools, especially good independent schools that want to promote ranking?

    Thanks for the post!

    Bob Ryshke
    Center for Teaching

  6. Bo — I read your recent post about grades with interest. If you search for “grade deflation” on the Princeton website, you’ll see all sorts of articles related to the “grade deflation” that Princeton instituted about 6 years ago (one example below). Departments have limits on the number of A’s. If a professor wants to recognize outstanding work by a student, he/she actually has to petition the head of the department and dean of the college and write a letter justifying the rationale for the A+. Because that requires extra effort, bureaucracy, and additional approval from higher up, professors have admitted to students that they just don’t do it, even if deserved. Likewise, professors have expressed frustration to students that they just couldn’t give them an A, even if deserved, because they had to be careful about the number of A’s awarded. Princeton has gone so far as to send to the parents each year a multi-page booklet on the “grading policy”.

    When Princeton changed the grade policy, they expected the other Ivy League schools to follow suit — except they didn’t. One unintended consequence was Princeton graduates not making the cut on GPA for grad schools because Harvard, Yale et al. continued to give high grades. Now Princeton sends out a letter with each transcript explaining their grade policy to mitigate the effects that the grade deflation might have on grad school and job applications. (Anecdotally, Princeton kids don’t think the explanatory letter has proved helpful.)

    This is all crazy. In an effort to better correlate grades to the quality of the work, they have instituted a system that rewards intense competition and can penalize outstanding work. If every student in a class masters the material, every student cannot be awarded an A. The ramifications are interesting.

    If grades and assessments are going to be meaningful, leading academic institutions should be more thoughtful and employ the very critical analysis they are teaching their students. And graduate schools and employers shouldn’t be using “grade cut-offs” (though that’s a whole other issue).

    Anyway, it’s all very interesting.

    I was just discussing this topic with my future son-in-law who is at Wharton getting his MBA. He said that at Wharton only 25% of the grades may be A’s, 60% may be B’s and the remainder below. Professors can give fewer A’s but cannot give more — regardless of the mastery of the material. He said that the way professors handle this is to make tests so difficult that they get the necessary grade differentiation. This semester he took a course where the mean grade was a 91. The professor complimented the class but then told them that the final was going to be extremely difficult because he couldn’t give that many A’s.

    Crazy…

    Olga

  7. Yesterday, I spent a few minutes writing to my blog in order to reflect a bit on a recent NYT Education article about grades. A few people replied via email, and two colleagues posted comments to the post. Thanks to those who are engaging in the conversation – for me, blogging is an opportunity to think out loud with the potential that others will think with you…to help refine and clarify a deeper understanding by collaboratively discussing.

    If I am reading and interpreting jgough’s comment as she intended it, then I fear that I was not as clear in my writing yesterday as I had hoped or thought. So thanks to jgough for giving me a chance to clarify by asking some good questions. I think some of her questions would be better for Mr. Perrin at UNC, but I can use her good questions to clarify what I was trying to say.

    1. I was trying to say that much confusion surrounds grades. Grades mean many different things to different people. At a most simple dichotomy, grades are measures of learning for some and ranks of comparison for others. I wish grades were clearer in meaning – that we held shared understanding for what they are and what they communicate. If we are going to have grades, then grades should be outcomes of assessment with clear PURPOSE.

    2. I was trying to say that grades should be measures of learning over ranks of comparison. As measures of learning, I think grades can become intrinsically motivating – to help one improve oneself with good feedback about what one knows, wants to know, and how much of a gap exists between those two nodes. As rankings, I think grades fail to serve their best potential for the learner – the earner of the grade. Again, if we are going to have grades, then grades should be outcomes of assessment with clear PURPOSE.

    3. I was trying to say that grades are best when they correlate closely to well established learning targets that are clearly defined. When a grade serves as a measure of current performance against a clearly articulated standard, or quality criteria, then I believe students and teachers (and others) can use the grade to get a relatively accurate picture of learning compared to the learning target and clearly defined quality criteria. If we are going to have grades, then grades should be outcomes of assessment with clear TARGETS.

    4. I was trying to say that additional context for any grade is helpful. A grade devoid of context is not worth much. Recognizing that medians can add some positional context to the mean average, which can give undue weight to extremes, I was happy that UNC was, at least, communicating that mean averages are a tiny bit more contextualized with accompanying median scores. I hope that UNC and other institutions will keep striving for even more context. I was NOT trying to say that adding median scores was enough addional context. But it is a start, when many educational institutions think mean averages are the ONLY way to report grades. I do not think numbers are great summaries of student learning – they fall way too short! If we are going to have grades, then grades should be outcomes of assessment with clear and thorough COMMUNICATION.

    I think a grade is a numerical/alphabetic symbol, used for efficiency not effectiveness, and I believe that many people have internalized what an A, a B, a C, and an F symbolize. HOWEVER, we don’t agree on what those symbols indicate. One person can see John’s A as an indicator that he mastered the course content, and that same person can see Susie’s B+ as an indicator that she did not master the course content. But what if John and Susie attend a school that limits the number of A’s that an instructor can give? If the institution has a practice of assigning grades, and if the grades are NOT indicators of student learning relative to clear targets, then the grades are merely rankings for someone else to use for future admission or evaluation of a candidate. I think using grades for such a solitary purpose is wrong. But if a school decides to use grades for that purpose, then the school should at least make that clear! Don’t allow people to think that the grade communicates clearly about a student’s learning relative to the course objectives. Because some people will see Susie’s B+ and jump to that conclusion, when Susie may just have failed to get into the 35% who can receive A’s – whether she (L)EARNED one or not!

    It’s no wonder that teachers and professors think that some grade harder and easier than others (as indicated in Lewin’s article) – we DON’T AGREE on WHAT, HOW, WHY, or WHEN to grade!

  8. If we agree that a single number without context is meaningless, and even a simple statistic like a median doesn’t really add much context, how can we get back to emphasizing to all audiences (colleges, employers, parents, teachers) that the measure of learning really comes from the conversation, not the number? We write comments with every grade, but many students and parents may overlook these as insignificant compared to the hard number. This also ties back to my musings about wanting to deepen the conversation about learning beyond grades and scores .

    Do things like honor roll help this emphasis or hurt it?

    • Good comments and question. Thanks. With so many different people, I imagine that Honor Roll helps some and hurts others. Probably, there is not a one-size-fits-all solution. However, I agree that we need to continue working to add context to grades, if we are going to give them. As an institution and as an industry, it would be great if we could agree, to even a moderately greater extent, what is in a grade. I find this resource at ATI interesting for stimulating more thinking about what’s in a grade: http://www.assessmentinst.com/wp-content/uploads/2009/04/arkstudyguide-09.pdf

  9. I’ve read your post and the article A Quest to Explain What Grades Really Mean, and I have questions.

    You said “In my opinion, the more context that can be added to a grade, the better! A grade devoid of meaningful context is not worth much. Why? My answer: because grades mean so many different things to different people. “
    While I agree completely that the final number gives no meaningful context, I have to wonder if a median average to accompany the mean average adds any context? Let’s take a look at real data from real kids. Let’s hold some variables constant. These data are from two sections of the same course with the same teacher.

    Here are 3 examples, remember same teacher, same test, same curriculum (overall combined median is 90):
    GD: 83 (B) for the semester, 83 (B) median score for this section
    AH: 88 (B) for the semester, 91 (A) median score for this section
    SQ: 90 (A) for the semester, 91 (A) median score for this section

    You asked “But shouldn’t grades be real measures of learning?” Does any of the above information tell whether real learning has occurred? Will it ever?

    About AH…Does 88 tell anyone that AH has always been a nervous, reluctant student who has finally found success and the confidence to ask questions and reveal needs that make learning possible? Does AH’s 88 with a median of 91 describe her growth? Does it say how hard she has worked to come from a 61 on the first test to this average of 88? AH knows that learning has occurred! Daily, I had to explain that I cannot release exam scores until everyone has completed every exam as is our policy. And daily, AH returned to inquire again because of the success felt by this child – a learner no longer hiding; a learner with a real desire for feedback.

    About SQ…Does 90 indicate that this child is coasting through this course without much effort or learning? Does SQ’s 90 with a median of 91 inform others that not much new learning has occurred because of natural talent? We work to assess/grade/measure to a common standard. This child is doing just enough to get by. Does SQ’s 90 make him more of a learner than AH? Who would you want in your class, at your school?

    About GD…Does 83 give any information about the learning of this child? Could it be that this child wants an A, but is not willing to do what it takes to have one? Or, could it show that this child works with the teacher every afternoon after school to learn and might actually be over-achieving? And, which median score do we use as the comparison number? Doesn’t GD’s 83 compared to the section 83 give a different perspective on “learning” than GD’s 83 compared to the combined sections 90?

    If it is about learning and growth, why would we offer more opportunities to compare one child to another child?
    Returning to your GREAT question “But shouldn’t grades be real measures of learning?” can one number or letter ever show a true measure of learning? If the answer is yes, then what number or letter should we be reporting? Should we report the mean of the grades for the entire semester? Do we report the most current grade? Should the most recent cumulative work be weighted more than early work?

    How do we adequately represent what a child has achieved by a determined deadline?

    I suppose I am asking for help to clarify a couple of things. What do I need to learn, change, and improve about grading? How can I report my learner’s grades better next time?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s