The basic theme of this blog is my disdain for the letter grade. As a professional educator, evaluation of students is a big part of my life. So, if I were to not use letter grades, then how would I evaluate my students?
The first idea that comes to mind is a narrative comment. Allow me to consider a hypothetical situation to better explain what I mean. I teach math, so students take many tests (and no, I do not like tests either, but that seems to be a discussion for a different blog... at the very best it's a tangent, and being a math teacher, I should know all about tangents, both the mathematical kind as well as those meandering ones such as when I start going off about tangents when I am supposed to be considering evaluation... )
Anyway, let us pretend that I give my ninth graders a test and the test includes problems on algebra, some on trigonometry, a few on matrices, some graphing, and some transformation problems. As this is math, I can create an answer key and I can correct a student's work against this answer key and I can quickly see a student's mistakes. I do not merely consider a student's answers; I also look at the supporting work, that is, the steps a student took to find, for example, the inverse of a matrix or to solve an equation. So, regardless of a student's answer, I can see the methodology and, really, the thinking a student utilized to arrive at their answer. Thus, I can evaluate not only the correctness of their answers but also the appropriateness of their method and their ability to communicate this method.
So, I go through and correct their test. I mark which answers are right, which are wrong, and I indicate where their method is appropriate and where it is not. Like most math teachers I know, I can award points for correct answers as well as for supporting work. So, then I add up all of these points and arrive a number. Now, many math teachers would know the possible number of points one might earn on a test, and at this point they would divide the number a student earned by the total possible to get a percentage which they would then convert magically into a letter. (See my last post to read some of my opinions on that.) But, the point of this post is to consider what to use in place of such a letter grade.
Back to the student's test... they have earned a number of points. Some of these points are for correct answers and some are for correct methods. Now, that number of points itself is feedback for a student and is thus a form of evaluation. Moreover, if I indicate clearly where they earned points and where they did not, then that is another level of evaluation. A student can see if they used the proper method and if their use of that method led to a correct answer. That, much more than a letter grade, tells a student if they have correctly learned how to find the inverse of a matrix or if their equation solving techniques are strong.
So, without any narrative comment (and without a silly letter), I have already provided a student with copious information regarding how they did. But, I have not really passed any sort of judgment yet as to my opinion on the quality of their work.
So, now the narrative comment becomes useful. I could convey to a student if I think they did well or not. I could judge their performance compared to earlier work. I could specifically state if they were strong on matrices, but weak on algebraic equations. I could clearly point out whether they chose the best methods or not and how accurate their answers were. I could also judge if they have been learning as much as I think they should have up to that point. Additionally, I could outline areas that they ought to focus on as they strive to improve.
Such a narrative could fill half a standard sheet of paper or more. Such a narrative could take 10-15 minutes to write. Such a narrative would provide far more useful feedback to a student than any letter grade could ever hope to.
But, what is the cost of such a narrative, filled as it would be entirely with my opinions? Does my opinion on the quality of the work really matter? Couldn't my corrections stand alone as all the evaluation I do and couldn't I then let a student make up their own opinions on the quality of their performance? That is a discussion for a future post.
When I entered my doctoral program, I brought with me a Masters of Fine Art in Writing and Literature that was evaluated entirely by means of narrative comments. I remember well the day that we brought our masters degree transcripts: everyone else had their work neatly inscribed onan official-looking piece of paper with an embossed seal. Mine was a sheaf of papers that looked like the manuscript copy of Notes from Underground! This posed no particular problem in the circumstances because all we were obliged to do was to prove we had a masters. But if they'd needed to make a comparison between my work and that of my cohort...that would have been tough.
ReplyDeleteYour critique of the letter grade goes to this dual function we have as teachers. We're both mentors and--like it or not--gate-keepers. We are obliged both to cultivate dispositions and habits of mind, and to render an account of who's achieved mastery, who's in the middle, and who hasn't come to a median standard. Why? Because we fear that colleges won't respond well to opacity on that count.
I'm going to duck the opacity question for now. But I will say that the narrative comment seems eminently appropriate to the mentor function. And though the elements of opinion raises the specter of subjectivity, I've found that it settles more questions because it tells students what we really think. In my experience, it's what they really want to know.
One way to align the gate-keeper more fully with that desire of students to know where they stand might be to shift from norm-referenced grading to criterion-referenced grading, otherwise known as grading on a curve. This could be done with or without letter grades, though some form of legible benchmark would enable students (and outside entities that might need to evaluate what the grades mean) to understand where each individual performance fell along the continuum.
Grading on a curve has a bad rap for being "squishy," but it doesn't have to. It probably would place a greater burden on the instructor to use the results of testing to plan future instruction, thus creating more work day-to-day.
I agree with your points about the narrative, of course--I'm not sure anyone would disagree that feedback is a good thing (except for Alfie, perhaps?!) I flatter myself to think that my students read the comments that I've written on there laboratory assignments and tests, and use that feedback (both positive and negative) to guide their future work in the course. A big part of my reason for wanting to come to a school like Poly is that I feel that public school teaching--true teaching--currently places teachers in a completely untenable position. It's tough to even give an accurate letter grade to a student's work, let alone written comments, when you see 200 students a day.
ReplyDeleteThat's no exaggeration: the year I left Berkeley High, I was seeing 200 students a day, every day.
Poly has provided me with the opportunity to work with a reasonable number of students, and provides me with enough time in my work week to fairly assess their work, including written comments. This experience has had an enormous effect on my ability to work with students, and I suppose if we weren't all part of a bigger system, I might be content to walk away from the letter grade.
The reality, though, is that we all seek metrics on which to base our decisions, in just about everything we do, and that's not unjustified. A decision on whether or not to buy Apple stock should not be based on how cool the products are, and a decision to buy a car for one's family should not be based on the flowery prose of the salesman. There are metrics for these things that serve a valuable purpose, and which are perfectly appropriate.
I think that trying to assess a student, one often relies on the same thing: a quantitative measurement that can be used to indicate the student's status, either as a means of measuring his own growth or as a way of comparing him with other students.
(Full disclosure: I and many of my science colleagues look at letter grades as a measure of a student's ability with some skepticism, which is why we require "instructor approval" for admission into AP-level courses; the grade does not reveal all, obviously, and we spend time interviewing each student's previous teachers to attempt to determine how well-suited we think they are to the AP course they've signed up for.)
Jamie,
ReplyDeleteI love that you referenced criterion-based grading (though I'm not surprised as I find it more common among English departments). One of my future posts is going to be on the use this style of grading which I employ in my Senior Stats course. You seem to imply that such a grading style is interwoven with the use of a curve (another topic for another post and another means I have used), but this is not necessary. The two can (and likely often do) go together, but it's kind of a Reese's Peanut Butter cup symbiosis.
Richard,
ReplyDeleteYou talk about purchasing cars and how a score or grade is needed. I beg to differ. A grade doesnt tell you anything, however an explanation of its positives and negatives tells you much more. If all I see is a letter grade, how do I know if it has the features most important to my like high gas mileage and good safety features. Maybe it got the higher score for looks and high tech features. A narrative is much more help to me.
And Alfie would support what Josh has written on this post which is a form of evaluation.