How can teaching assistants deal with grade disputes?

Question

From the answers to this question, it is clear that TAs in the US are usually responsible for grading students' work.

In my country, most of the TAs face the following challenges:

1. Students confront TAs and attempt to negotiate for extra marks in order to pass, or to improve a grade;

2. Students confront TAs and attempt to negotiate for extra marks by using a friend's answers as evidence to support their case;

3. Students inquire about the subtle methods used in grading: the amount of marks awarded for each step, the reason for a decrease of even 0.5 marks, or they ask to see the solution manual, etc.,

And then these conflicts evolve into personal enmities, which may eventually cause the TA serious consequences (a heated confrontation, loss of students' respect, etc.).

The issue is worst if the assignment's answers are quantitative or mathematical. There is no solution manual and sometimes no way to guarantee consistent and fair marking over a great number of student assignments (even for a single particular question).

So I want to know how TAs in the US—or in other countries—avoid these complications. What are the key techniques?

Note: Some say it's a minor issue. No. It is a major issue and often arises with TAs in my country, where the professor is not involved in the distribution or evaluation of grades.

Answer 1

In the United States, the professor responsible for the course should take responsibility for resolving disputes (but not requests for corrections and clarifications). There is an expectation that they will be heavily involved in grading, even if the bulk of it is done by teaching assistants.

In some cases, the professor may choose not to involve themselves in grading at all, and leave it up to teaching assistants to resolve disputes. I do not agree with this practice, but if you are stuck with it, you can handle disputes by:

• Using a rubric to grade.
• Resolving all disputes using the rubric.
• Refusing to compare the work of different students. In the USA, comparing students' papers runs a risk of violating privacy laws.

Ultimately a portion of students will never agree with you about how you grade (they want to learn something different from what you are expected to teach) and this simply must be tolerated.

Answer 2

Tell students that, "a TA's grades are final and non-negotiable," and that, "any disputes should follow the usual channels." Or, ideally, have your boss (the prof or whoever) or their boss (whoever runs, or has control over, the department) tell them.

Answer 3

The ultimate solution to this problem is for the professor to spend time and set question papers that are not ambiguous or amenable to varied interpretation by students.

You state:

The issue is worst if the assignment's answers are quantitative or mathematical.

I think it is actually the other way around. Mathematical/statistical/OR/Computer Science and other computational or proof-based courses are rather easy to grade. The reason is that these subjects tend to have exactly one right answer as long as the question is framed with sufficient care.

As an example of a carelessly worded question which will eventually lead to student and TA/professor grading disputes is the following.

There are two identical machines. There are two jobs waiting to be processed on these two machines. Job 1 takes 2 hours and Job 2 takes 3 hours. How quickly can both jobs be processed?

Now, when setting the question paper, the professor may have in mind an unstated assumption that a job cannot be broken down into smaller jobs that can be parallelly processed. Hence, in the professor's mind, the right answer is 3 hours.

However, students may interpret this problem in such a way that the jobs can be broken down into smaller components that can be processed parallelly. In that case, the right answer is 2.5 hours. Students are not wrong in this case. The question was just silent on whether the job can be broken down into smaller components or not.

In social sciences, such as linguistics, art history, gender studies, development economics, there are very few "right" answers. A clever student can dispute any grading key set by the professor and implemented by the TA. Hence, you are likely to see higher amounts of grade inflation in such fields as compared to fields like math/physics/OR/statistics/CS/Engg, etc.

Finally, the professor should clarify these details in the very first session of the course that it is difficult to differentiate too finely between different wrong answers. Professors are not omniscient, and neither are the TAs. So, the student is not expected to dispute scores/grades too aggressively.

Inspite of all of this, there will still arise grading disputes and this is certainly one of the more unpleasant experiences of being a professor.

Answer 4

One way to resolve this for the instructor to specify that assignment marks will not be reviewed unless there is an obvious error by the marker or a mechanical error in entering or adding the mark, and that all reviews will be done by the instructor once at the end of the term.

The effects of this are:

1. that students typically stop fighting for fractional marks that have practically 0 impact on the final grade at the end of the term, and
2. that students do not negotiate with the TA.