“Ummm… why are your final grades 15, 26, 14, 32, 9, 41, and… 1?”
The course total is out of 100. 60 represents a minimal passing grade. Anybody who scores under 30 has to not only repeat the same course next year but will also be held back a full year in order to concentrate on this one course.
The teacher’s response to my question was rather defiant in tone. “Because that’s what they got on the test!”
I was not this teacher’s boss, but I was responsible for collecting the grades and submitting them. I didn’t want a confrontation but the situation was setting off alarm bells the size of Big Ben inside my head. The impact these awful grades would have on the students who took this class were one thing. The dubious rationale behind giving a final test score as a course grade was another.
I subsequently found out that this teacher’s approach to course grading is not entirely uncommon. But it is highly flawed logically, pedagogically unsound, and generally bad practice. Let me list thy reasons why…
Let me start off with an indirect response. Some teachers like to show how tough they are, or how challenging their courses are, by giving low grades, which they believes marks them as a ‘serious teacher who demands a lot from students’. This ‘tough love’ manifests itself in high failure rates, which the teachers assume somehow gives them more classroom credibility as tough educational taskmasters.
“A high failure rate means… your standard are out of whack with learner capabilities”
It’s mostly BS though. An extraordinarily high failure rate means that your standards are out of whack with actual learner capabilities. You’ve set the bar inappropriately. It also means that what you are teaching, or what you think you are teaching, to these students is not being learned. Sometimes students are at fault for this. But imagine if every Olympic performer on the Japanese judo team finished last in their respective weight categories. Would it not suggest that the primary problem is (and prepare yourself for a shock here) the coaching?
But let’s get to today’s real point – valid testing, course grading and the “But that’s what (s)he got on the test!” motif.
First, conflating a final grade with the ‘score on the test’ implies that you have performed and graded a proficiency test, not a course achievement test. When learners take TOEIC, TOEFL, IELTS et al test-takers receive a score — and that score is what they get. Ta-tum!Proficiency tests are designed and expected produce a single score. Period.
Achievements tests, on the other hand, are supposed to represent a measurement of knowledge and or skills developed over an entire course. The notion that all of this can be reduced to a single score derived from a single course-ending test is shakier than a Kardashian twerk. In order to have this single test be an accurate and meaningful measure of course achievement it would have to be extremely comprehensive, incredibly well-balanced and inclusive, wholly representative of the course goals and contents, and be constructed in such a way that reliability and/or validity could not be questioned.
Good teachers realize that most courses are about the process of learning, not merely the product. Proficiency tests are all about the product, and not about diagnostics or wider aspects of development and achievement. If we measure only some product there is little reason to value actual attendance or the crucial hypothesizing that comes through engaging the actual course contents, with others in the classroom. Such a ‘course credit’ would have the same value as a placement test — at best it might indicate that you are already good at English, not what you achieved through participating in this course.
This is why most competent teachers will grade a course on a series of separate evaluations (of varying types and styles) and also generally include factors such as attendance, participation, homework, effort etc. in the grading mix. The criteria (yes, plural) for grading should be made clearly known to students in advance. A course-ending random series of old-school discrete-point questions in the proficiency test mold will not suffice — doubly so if you are teaching a course that proposes to inculcate skills or develop productive communicative abilities.
“This approach is the inverse of conflating test scores with course grades…”
Taking the process-as-measure approach means that a course grade of 60% will actually have a meaning. It should not and will not simply be ‘the number that they got on the test’. Rather, ’60’ is the number you write as the student’s final grade if, in your professional opinion, that students has completed the minimum requirement to earn a passing credit. This approach is the inverse of conflating test scores with course grades – it’s not that you ‘got’ an 85 therefore you must have done well, as it is that you did very well so I’m going to give you an 85. Because (wait for it!) that’s a number I give to students who… have done very well. Double ta-tum!
To score otherwise would be, again for emphasis, to treat your course achievement grade as if it were a proficiency or placement test result. It would assume that your whole course is about teaching to that test and, further, that final test score is the only thing that matters, the only item that defines the entire course performance.
How bad was that test performance?
Now this doesn’t mean that in-class test scores should not be added to make up part of that course grade total. In my own course grades I include the scores derived from three different in-oncourse evaluations (very distinct in style and scope in order to better encapsulate the nature of the course). However, none of these assessments are of the type that could conceivably produce a score of ‘1’, ’19’ or ’26’. If, in my professional opinion, the performance of a given student was so bad on any given test as to scrape the crud off a trawler’s hull, I would give that student a 30 — once again, not because ‘that’s what he got’ but because that’s what the grade ’30’ represents.
A course grade should not not be something that simply happens to the teacher or student (“Well, it looks like your course grade is 96. Wow! Whodathinkit?“), it is something that should be carefully and actively decided by the teacher to accurately represent the student’s performance — which would automatically include the results of any tests administered during the course as a partial measure of performance.
Subjective/Objective: “A course grade based on a single, final test… is far from being ‘objective'”
Let me conclude by anticipating a possible objection. By placing emphasis on the teacher actively producing the course grade, as opposed to it being the numerical product of a test alone, I might be seen as placing too much value on the subjective evaluation of the teacher over the more objective raw test score.
The problem here is that the notions of subjectivity and objectivity are elusive. A teacher who actively chooses and assigns a course grade but includes the results of various in-course tests (the scores of which might also contain aspects of ‘subjectivity’ — such as role-plays) in that grade is not just pulling subjective ‘I like the color of your eyes’-type scores straight out of his or her butterdish.
Likewise, a course grade based solely on a single final test that features discrete-point random questions that do not address the course comprehensively, or fail to adequately measure the skills the learners are supposed to be developing, is far from being ‘objective’ — even if one can somehow assign a number to it.
So if I give you 100% in my course it means that I think you did the course perfectly. It isn’t what you ‘got on the test’.