Grading is a concept that almost no two teachers agree upon. Some believe in including effort while others believe only performance should be considered. Some believe in many A’s while others believe A’s should be rare.
In this post, we will look at absolute and relative grading and how these two
Absolute Grading
Absolute grading involves the teacher pre-specifying the standards for performance. For example, a common absolute grading scale would be
A = 90-100
B = 80-89
C = 70-79
D = 60-69
F = 0-59
Whatever score the student earns is their grade. There are no adjustments made to their grade. For example, if everyone gets a score between 90-100 everyone gets an “A” or if everyone gets below 59 everyone gets an “F.” The absolute nature of absolute grading makes it inflexible and constraining for unique situations.
Relative Grading
Relative grading allows for the teacher to interpret the results of an assessment and determine grades based on student performance. One example of this is grading “on the curve.” In this approach, the grades of an assessment are forced to fit a “bell curve” no matter what the distribution is. A hard grade to the curve would look as follows.
A = Top 10% of students
B = Next 25% of students
C = Middle 30% of students
D = Next 25% of students
F = Bottom 10% of students
As such, if the entire class had a score on an exam between 90-100% using relative grading would still create a distribution that is balanced. Whether this is fair or not is another discussion.
Some teachers will divide the class grades by quartiles with a spread from A-D. Others will use the highest grade achieved by an individual student as the A grade and mark other students based on the performance of the best student.
There are times when institutions would set the policy for relative grading. For example, in a graduate school, you may see the following grading scale.
A = top 60%
B = next 30%
C = next 10%
D, F = Should never happen
the philosophy behind this is that in graduate school all the students are excellent so the grades should be better. Earning a “C” is the same as earning an “F.” Earning a “D” or “F” often leads to removal from the program.
Grading Philosophy
There will never be agreement on how to grade. Coming from different backgrounds makes this challenging. For example, some cultures believe that the teacher should prepare the students for exams while others do not. Some cultures believe in self-assessment while others do not. Some cultures believe in a massive summative exam while others do not
In addition, many believe that grades are objective when there is little evidence to support this in academic research. A teacher who thinks students are low performers gives out such grades even if the students are high achievers.
As such, the most reasonable approach is for a school to discuss grading policies and lay out the school’s approach to grading to reduce confusion even if it does not reduce frustration.
the teacher must be capable to evaluate the amount of knowlwdge necessary to the student go ahead in the course. With yhese knowledge the teacher can select the minimum minimorum that can be used as basis to generate all of them. Them, it is possible to do a test composed of two portions: the former is the minimum minimorum and the students wathing them received a grade enough to continue the course, the last portion is the maximum possible on which the students arre measured by the skill they shown, the questions are hard. But these are the beginning of evaluation. The teach must through the text fom the students, analyse his/her own teaching. How to do this? Noting the answer can detect: 1) how the students had be compreesion of the question, 2) the creativity in the answers, 3) the way to solve questions were similar to teacher or, by thre other side, had shown a seek in knowledge beyond the class and books, and so on. The discursive test are very useful for this purpose. In summary the evaluation of studys in order do assign a grade is an excelent way to promote the grow of the students.
The problem I encountered in medical school with relative grading is that it was counterproductive to team work. Specifically, you benefit from the missed points of others. my introduction was a graduate student taking medical classes, after finish my notes I went to leave class only to see a medical student take someone’s notes and throw them in the trash. I put them back on the desk after he left. Then, in neuroanatomy lab, I shared a half brain with a medical student. After he learned each cranial nerve he would break it off so I would do worse (too bad, I did better). As a medical school professor, I’ve seen and heard of this several additional times. Also,in the four institutions I taught medical students at, the students would not help one another in the two that used relative grading and the the students freely helped each other in the two that used absolute grading. In my experience, relative grading encourages and rewards academic misconduct and discourages teamwork. In my opinion, relative grading has no place in an ethical institution of higher learning.
Yes, this is a clear problem with relative grading. However, if you fix the proportion of letter grades the overall performance may go down but the distribution of letter grades will still be the same.
In 33 years of college teaching in both engineering and geology at two different universities, I have always used relative grading. I call it the natural distribution of total points. I plot a graph with the total points vs the number of students with that point total. The top two grades are always A’s, and any grades within 3 or 4 points are also A’s. Point totals 10 points lower are B’s. 20 points lower are C’s. At my school, we also have intermediate grades such as BA, CB, and DC. The BA range is roughly 5–9 points lower than the top of the A range, unless the top A is clearly superlative and a curve buster. I look for natural gaps between the clusters of total point scores, to establish the boundaries for the grades.
I read Dr. McCabe’s description of medical school where relative grading prompted unethical behavior. I have encountered cheating in my career, but apparently not as rampant as in his experience. I ask students to do projects and homework in teams which they suggest and I approve. My tests are open book, open notes, and are done individually. In my experience, the best students are usually willing to help other students.
I feel that grading relatively is fairer than a 90-80-70-60 scale or any other absolute scale. My tests typically result in average scores varying from the 70’s to the 80’s. Some of my objective tests average in the low 70’s. The relative scale recognizes that tests vary, teachers vary, but the best students always do A work, which should be accepted as such, regardless of how low my tests average. Honestly, I consider my grades less subjective or subject to manipulation than those based on so-called objective scales.
Dr. Duane
Thank you for your comment. I guess another factor to consider is the degree being pursued. Maybe in medical school relative grading doesn’t work but it is appropriate in other areas.
I was teaching at a place with relative grading. I used to teach core courses in computer science, and programming courses. It never happened that the whole class earned between 80 and 90 percent, and it never happened that I was awarding a C to students with a 60 percent score. It was usually the case that relative grading covered the fact that the class average score was quite low-usually close to 40 percent, sometimes approaching 50.It actually lead to a very calculated complacence among the students- they would aggressively tract the class average, and perform just about it.No matter how low the class average was (sometimes between 30 and 35 percent), we had to plot the histogram, and award the top 10% As. It would be justified as ‘being a tough course’, while it would not be unusually tough, and in any case, we had the policy of setting a graded question paper, where certain fraction of the questions are quite easy, certain fraction are moderately difficult, and certain fraction were harder, and needed the student to a small amount of creative thinking. It was frustrating to see that students would routinely ignore the harder questions, and not attempt them. Only 4-5 in a class of 120 would attempt these questions with any seriousness, and their scores would be the outliers in the graph, not affecting the grades of the lower performers. These few would get the satisfaction of seeing their top ranks in the class, but as far as the class grades were concerned, these students did not make any difference. It was as if the entire grading was done by setting aside those (moderately) challenging questions. The fact that most of the A graders did not attempt the top questions would not show through. Relative grading covers up a lot, mostly it covers up. The syllabus, the textbooks used, and the question papers and assignments used do not tell the true story. A fair sample of the graded scripts would tell the real story, but they are never scrutinized with the aim to check the overall quality of grades.That cruel situation of a student scoring 70% and still ending up with a C would never arise, at least in that institution.
This is quite an extreme example of relative grading abuse. All I can suggest is perhaps a modified version of it in which certain thresholds must be met for relative grading to be employed. For example, students have to earn at least 60% to benefit from relative grading or some other modified approach.
Relative grading is stupid. It sounds like it would break if one clever dumbass decides to cheat on an exam and achieves the highest score in the class. So everyone else’s grades would be adjusted accordingly based on that guy’s so-called efforts? Absolute is better
If you use relative grading without thinking it would be stupid. If one person gets 100% you can remove them from the curve as an “outlier” and go with the second highest score if you wanted. The only problem with relative grading is if you implement it absolutely.
Good read! Dr Darrin.
Thank you for sharing
Pingback: Adjusting Class Curve When Removing A Question From An Exam -Blackboard -Canvas - BlackboardSub