I have a vivid memory of the last time I got marks. It was in the 8th class. We had a special course on technical drawing. I had got 60 marks in the class exam and almost 60 in the final. However, I felt that the final was a very unfair exam. I had topped the class exam, however, in the final there were a dozen boys ahead of me. My problem was that I could solve difficult problems but couldn’t draw a straight, smooth line. So, I did not like an easy paper such as in the final where my weaknesses became more pronounced. The teacher, on the other hand, had to ensure that enough students got sufficiently high marks!
We need to have some measure of performance and achievement. Marks and grades are intended to serve that purpose. When we test human beings, the process of testing has an impact. What a student learns and how a teacher teaches are strongly influenced by the testing methodology. Grading systems are not just a mapping of a range of marks to a grade. The first and foremost issue is that grading must not be a centralised process with a common board of examination. An example may help to understand why. Consider a subject like science which may have a lab component.
Let us say, students will earn 20 marks for homework assignments. If there are 5 of them in a term, this means 4 marks per assignment. Another 20 marks can be for the lab. Each student has to demonstrate to the teacher that the experiment is done and the results are properly written. The number of marks per experiment, however, will be very few. A mid-term exam can be worth 25 marks and the final, 35 marks. The mid-term and final have to be set by the teacher.
Statistics indicate that there will be a clustering of students within a class. The clusters determine the conversion of the total marks into grades. A teacher may examine the borderline students to decide on the dividing lines between grades.
In a subject like English or Hindi, the marks can be distributed over essays, book reports and oral presentations, in addition to the home assignments and exams. Each teacher may decide to change the proportion of marks for an activity. A common category is quizzes. Instead of giving homework, a teacher may decide to have a quiz on the lesson she planned to give as homework and grade that instead.
By the time final exams come around, both students and parents are aware of the students’ performance, therefore, the result of the final exam is not likely to be a surprise. None of the elements is a major component. Consistent performance is what matters.
Incidentally, suppose the final exam was common across schools and grading was applied to all students, imagine the pressure on teachers to give high marks for each assignment because the teacher in the other school is very liberal. No wonder, the current systems with a small portion of internal assessment are failures.
Concerns against a distributed evaluation system
How will we admit students to colleges if there is no common exam for all students in various schools? The centralised board exams are not serving this purpose. Professional colleges already have common entrance exams. Other courses are also feeling the pressure when trying to admit students from different school boards. It would seem that the board exams are no longer reasonable differentiators given the very high marks obtained by many students. I am confident that our children are not substantially smarter than us.
How can we be sure that schools are fulfilling the tasks expected of them? Here a centralised testing for some skills like mathematics, and communication may be desirable. The analysis for such testing is not to measure the student but to measure the performance of an institution. We need not have a centralised exam but can instead use indirect measures like how many go to college and how they perform in college.
Our primary goal has to be to help students learn
Now, let us consider a simple quiz on learning. How many children can tell the names of the presidents of India? Compare that to how many children can tell the names of the captains of the Indian cricket teams. Did schools ever teach the latter?
Children are obviously capable of learning, including memorising, a lot. How do we motivate them? The instrument for motivation is the teacher. Yet, in a centralised system of examination, the teacher is merely a cog. In fact, a less significant cog than the tuition classes. In the tuition class, the students learn the critical skills necessary to score very high marks.
Let us consider the implications on students and teachers if the testing measure is based on what happens in the classroom. If two teachers teach the same subject, even in the same school or college each sets his own tests. The following are modelled on my experiences or memories of my experiences:
- Teachers ensure that it is easy to pass with a ‘C’ but hard to get an ‘A’.
- Feedback on getting every answer paper back and arguing with the teacher is an essential part of learning. In my experience, teachers never hesitate to increase the marks in a test if a student convinces them that (s)he had misunderstood the intention or the teacher had misunderstood the student. The correction is immediate.
- No single test has a make or break capability. The grade is based on performance throughout the semester in various types of activities. In my experience, in English, essay writing and book reviews mattered a lot. In mathematics, problem solving mattered more. Again discussion with teachers was the norm and, by implication, a teacher could not afford to be arbitrary and wasn’t.
- There may be surprise tests. Attendance was implicitly assured, not enforced. (Except in the case of problem students.)
- Mass tuition classes would not make any sense as there is no common paper for which to prepare. One may take tuitions to help one understand.
- Teachers can play on their strengths and the needs of the students.
There are significant social implications of a non-centralised grading policy as well. Affirmative action in the US raised issues of reverse discrimination. Yet, a society cannot afford to ignore the needs of the socially disadvantaged groups. Invariably, the disadvantaged students will belong to the less well known schools or regions. A distributed grading policy allows one to pick the ‘big fish from the small ponds’. For instance, “Texas, California, and Florida have all instituted a percent-based admission policy as a replacement for affirmative action. In Texas, for example, this policy guarantees the top ten per cent of Texas’s high school graduates admission to a state school as long as they have taken a list of required courses.”1
Until recently, BITS Pilani was using a similar practice of picking the best students from various universities to fill their seats and ensuring an all India impact.
In Goa, I noticed many well-off parents moving from ‘rural’ locations to ‘urban’ areas to ensure that their children went to better schools and had a better chance of getting into professional colleges. A policy of picking the best students from each school may even reverse the migration. This will also give more importance to an assessment method that is led by teachers who observe a student over time.
It is not enough to say that a teacher is important in a society. We must give teachers the necessary importance. Grading policy at the class level is an important tool and a signal in the right direction.
The author can be reached at [email protected].