Policies and Procedures
The Logic of Makeup Exams
The purpose of any makeup exam is to allow students, with legitimate reasons for missing a scheduled exam, to fulfil the requirements of a course, and hence avoid being penalized for factors beyond their control. Legitimate reasons at most universities, including UW, include
If such a reason is determined to exist, a student is entitled to write a makeup exam comparable to the final exam written by the rest of the class, but at a more convenient time.
Since documentation of illness is costly but not difficult to obtain, normal makeup exam procedures are essentially a way to pay for extra studying time. (The costs may include money, but are more likely to consist of time, trouble, and certain psychic costs). Hence, normal procedures are unfairly favourable to those students who are better-endowed with the means to pay the required costs. Simply put, some students will always be better than others at getting sick, and the normal procedures governing makeup tests give those students better chances than other students have.
In particular, imagine two students, equally and both badly prepared to write an upcoming final exam, but one better than the other at getting sick. One student writes the scheduled exam and fails; the other gets sick, writes a makeup exam some time later, and passes. Under normal makeup exam procedures, the failing student has to re-take the course in order to get credit for it. The cost to the failing student is monetary--about one-fifth of a term's tuition--plus whatever time costs there are, from re-taking the course or delaying graduation by a term. In this hypothetical scenario, the inequity and iniquity are stark and egregious. To get somewhat less grossly unjust scenarios, just substitute pairs like "gets a B--" and "gets a B+", in place of "fails" and "passes".
In normal practice, no-one can possibly know how often these scenarios occur. Whenever someone passes a course with a deferred exam, and someone else fails or gets a low mark without a deferred exam, you cannot tell for sure whether or not such an injustice has happened. The problem is something that seems only natural, but is in fact noteworthy: In normal practice, only students who have raised legitimate reasons for missing the final are allowed to write the makeup exam.
A reasonable question is: If some students write the scheduled final exam, and fail the course, and you are going to set a makeup exam anyway, then why not let the failing students write the makeup too, if they want? They all had some reasons for not passing the course--but the reasons were just not quite legitimate enough. If only they had been a little better at getting sick, maybe their reasons would have been recognized as legitimate too.
Many students find that Economics 372 is a difficult course. For some, it is a first course in logic, in the sense that rigorous proofs are required for the first time; for others, it is a crash course in statistics. Moreover, the class is large, with an average of over 100 students. The upshot is that someone always gets sick, and therefore there is always a makeup final exam.
It is costly to create a makeup exam. The amount of work, for an exam involving 5 students, is over a full day: 2-2.5 hours to make up the questions, 3 hours of proctoring, about 2 hours of marking, and about 2 hours of paperwork (photocopying, posting marks, changing grades).
But the amount of work involved, for an exam involving 10 students, is very little more--only the marking time increases appreciably, but by much less than double. After marking the first four or five exams, the amount of marking time per exam decreases dramatically, to perhaps 10 minutes per test. So, a sensible answer to the above reasonable question is: Everyone who fails the course gets to write the makeup final.
Having reasoned this far, a further step becomes necessary. You cannot tell the difference between someone who writes the final exam and fails, and someone who writes the final exam planning to fail (and succeeding in this easy mission). Both students get to write the makeup final.
The problem is the waste of time and travel. Save it! Do not show up to the final exam, planning to fail. Just don't show up at all, but make sure you are on the list of rewriters.
Scheduling Makeup Exams
A makeup final exam can occur on any weekend during the four months after the term in which I teach a course, provided that the university is open (which rules out dates between about the 21st and the end of December). In most cases, makeup exams are held on Saturdays, in HH280, from 3PM to 6PM. In any case, you can find out the time and room from my voice mail message (extension 3413), or from your course's announcements page, starting about a week before the exam.
The rewriters choose the weekend of the test as follows. Each rewriter sends me a list of eligible weekends, not necessarily distinct, from "least favourite" at the top to "most favourite" at the bottom. By putting a weekend at or near the top of your ranking, you are asking to have that weekend removed from consideration, as a possible time for the exam.
Then there is the selection process, or more accurately, the deselection process: In the first stage, eliminate a weekend, if it appears at the top of two-thirds or more of the rankings.
In the second stage, eliminate any weekend that appears among the top two items of two-thirds or more of the rankings.
In general, in the nth stage, eliminate any weekend that appears among the top n items of two-thirds or more of the rankings.
At some stage, say stage N+1, there will be no weekends left. Immediately before then, at stage N, there will be one or more weekends left that are, in a sense, the "least disliked" of all the weekends. If there is just one weekend left after stage N, choose it. If there are 2 or more left, make up some sensible tie-breaking rule, and use it to choose one of the weekends that remains at stage N.
Note the quote marks, in "least favourite" and "most favourite": You do not have to indicate your true preferences in your ranking. (How could anyone require you to do so? They would need to read your mind somehow). Your ranking is a strategy that you play in the selection game, and will only by chance coincide with your preferences. For example, if there is a particular weekend you absolutely cannot stand, but all the rest are reasonably convenient, then you should name the bad weekend in every position on your list, so that it will almost certainly get vetoed.
Here is another tip. You can probably count on most of the other rewriters to veto weekends chronologically: earliest is worst, followed by next earliest, etc. This means that if there are two weekends that you cannot stand, and one of them is early in the term, then there is no need to vote against the early date. You can probably prevent either of your bad dates from being selected, just by voting against only the later date, and counting on the others to veto the earlier date.
Format and Marking
By checking the old exams file, you can see that makeup exams have exactly the same format as regular final exams. In many cases, they have exactly the same questions as regular final exams, though some of the questions have changes in notation and wording. Makeup finals are as close as I can make them, to the length, content, and level of difficulty of regular final exams. However, it is a fact that makeup final exams have significantly lower class averages than regular final exam averages, most likely because
For these reasons, it is difficult to estimate a makeup final's relative difficulty, compared to the midterm exams you wrote, and compensate appropriately for difficulty. Recall that if A1, A2 and AF are the class averages on the midterm and final exams for a course, and X = max[A1, A2, AF], then your adjusted midterm and final exam marks are
M1’ = M1 + X – A1
M2’ = M2 + X – A2
F’ = F + X – AF
an adjustment that equalizes the class averages of the three tests. The adjustment X – AF, to the final exam mark, is a measure of how difficult that exam was, compared to the two midterms. How should such a number be calculated, when due to reasons 1 and 2, and not due to the nature of the test, AF is biased significantly downward? It would be a mistake to infer, from an unusually low AF, that the makeup final was especially difficult. I estimate the magnitude of this mistake to be about 10%. That is, writers of makeup exams, exactly as difficult as a normal final exam, on average get marks that are that much lower.
There is no good answer. All I can predict, based on past experience, is that the adjustment will be embarrassingly large, but small enough so that the makeup final average will still be less than the normal final average.
Q: How do I become eligible to write the makeup final?
A: Fail the course, or do not show up for the regularly scheduled final. There is no need to ask permission, or bring a doctor's note or death certificate.
Q: What if I fail the makeup final?
A: You fail the course, and will have to take it again.
Q: What if I can't make it to the makeup final?
A: You fail the course. Why didn't you veto the date?
Q: But I got sick.
A: You have to prove it, and you have to prove you had a good reason to miss the original final exam. It is not good enough to get a doctor's note saying you claimed to have some symptoms of something. You need two notes, one for each test, saying that the doctor(s) actually found symptoms of something serious. Only in this case you will get a third chance to write the final exam.
Q: Is there going to be a makeup midterm exam?
Q: Can I write the midterm at some more convenient time?
A: No. The class was scheduled last year, and the midterm was scheduled weeks or months ago, so if you can't make it, you must have knowingly scheduled something in its time slot. Besides, your worse midterm doesn't count, and a 100% final is always an option.
Q: Can I write the makeup exam at some time different from the other rewriters?
Q: But I'm sick.
Q: What if I get my own exam proctor?
Q: What if....?
Q: Will you hold extra office hours before the makeup exam?
Q: Why don't you answer my e-mail?
A: If you had a reasonable question, the answer is in the course outline, or in this document somewhere.
Q: Why don't you return my phone call?
A: See above.
Do not take the makeup exam option on purpose, if you hope to graduate any time soon after the term is over. The makeup can occur up to almost five months after the course is over! The larger the number of people writing the exam, the later the exam is likely to be. The kind of person who writes a makeup exam is not the kind of person who can be rushed into anything.
My makeup exams exist in direct or indirect conflict with various UW policies. They will continue to be tolerated, perhaps barely, as long as they do not create any extra work for anyone but me. Therefore, you must not harass the staff or other faculty members for any reason related to makeup exams! For example, you are not allowed to argue to some other professor "Redekop always has makeup finals. Why don't you?" You are not allowed to try to persuade anyone (for example, the undergraduate chair, the department chair, the registrar, the university president) to reschedule a makeup exam, once the date has been set.
Rewriters who fail the final exam are relatively rare. Almost everyone who writes the regular final exam, with a sincere effort, ends up passing the course. Writers of makeup exams do not do so well. (See reason 2, above). What is worse, many people who are eligible to write makeup exams do not even show up to write them! There seems to be a fading over time, of both memory and commitment. So, I advise you: When the time comes to write the final exam, show up, grit your teeth, and give it your best shot.