Advice for new Teaching Assistants

This is based on a short talk I have given several times to new TAs (mostly new graduate students) at Carleton. It's very basic. I think it should work for most subjects, not just economics. (Though maybe not as well for science and engineering where TAs run labs?). I think it should work for most Canadian universities too, and maybe outside Canada.

I have no special expertise in this area. But I do have a lot of experience, because I have taught a lot of big courses with a lot of TAs. You will probably be working for someone like me, so you need to know what I expect.

1. This is a job. It isn't just a way to give money to graduate students. Sometimes it will be a really challenging job that requires all your ability, like trying to explain some difficult economic (whatever) concept. Sometimes it will be a really boring job that anyone could do, like adding up grades. In both cases it matters that you do this job well. If you add up the grades wrong we might fail a student who should have passed.

2. Your reputation matters. You should do your job well because it's the right thing to do. But I'm going to appeal to your self-interest as well as to your moral sentiments. For most of you, even though you might have flipped burgers in the past, this is your first professional job as an economist (or whatever). And it's a very small world. If you don't care about doing this job well, word will probably get out, and you might find it harder to get a good job later. If you do your job well, you can ask the prof for a reference. Future employers don't just care about your grades; they want to hire someone who isn't a pain to work with. A good reference from a prof you have TA'd for may count more than a good reference from a prof who has taught you. Plus, if you take this job seriously, you should find it fun. Update: And by teaching others economics you will learn and understand economics better yourself.

3. Profs (the ones you TA for). You are the Teaching Assistant. The prof is the one in charge of the course.  Most profs will give you a fair amount of discretion in how you do your job, but if the prof tells you to teach it this way, or to grade it that way, that is what you do. (Though if the prof tells you to do something seriously crazy, have a quiet chat with the department chair).

One thing you need to remember about profs is that nobody hired them for their managerial skills. And managing you is a long way down their list of priorities. Your manager at the burger place was probably a much better manager. Don't expect too much, and try to keep yourself organised.

4. Students (your students). Go easy on the students, especially first year students. They don't know their way around universities, and universities are very weird places, with a very weird culture. Don't expect them to know all the rules. Remember what it was like for you when you first arrived. And don't expect them to know any economics either. Get one of your own first year essays. Read it, and die of embarrassment. Yes, you really were that bad once. And then remember: you were one of the brighter ones.

Your students are certainly not your enemies, but they aren't really your friends either. Don't put yourself in a conflict of interest where it would be difficult to give an F to a student who deserves an F. Or be seen as difficult. And if you fail to follow this advice and do find yourself in a conflict of interest, then tell someone (like the prof), so it can be fixed.

5. Grading. Grading tests, assignments, exams, and maybe essays, is usually the main part of a TA's job. A lot of new TAs are worried about grading. "How is it even possible to decide what letter (or number) grade to put on each of this pile of exams?" Don't worry. I don't know how it works either, but it does work. As soon as you have read two exams, you will be able to tell if one is better than the other, and whether it's just a little bit better or a lot better. "But how do I translate that into a specific letter grade?" Here's how I do it. I follow the "functional theory of grading": a grade is what a grade lets the student do. Just read the text you have to grade, ignore everything else, and ask yourself: "What sort of student wrote this?"

F means: "This student learnt nothing in the course, and should not be at university. A random person off the street could have done as well".

D means: "This student shows some signs of having learned something, but really doesn't get it, and should switch to another subject."

C means: "This student sort of gets it, and is OK to continue in Economics."

B means: "This student should be in Honours economics, and maybe should think about doing a Masters.

A means: "This student is future PhD material".

A+ means: "This student should become a prof."

Some profs give you very strict answer sheets and grading schemes. Follow it, but be ready to show an exam to the prof if the grading scheme makes no sense in this particular case. Others just let you get on with it and use your own judgement.

If you are grading with other TAs, you want to try to ensure you are all grading consistently. I suggest you do this. Get together with the other TAs, plus the prof if possible, and all go through a few exams together. All read the same exam, then on the count of three, all TAs call out a grade at the same time. Briefly discuss your differences, then go on to the next exam, and repeat. After 3 or 4 exams, you will converge and all be giving about the same grade

And don't lose the grades! Make copies.

6. Leading Discussion groups/tutorials. The hardest part of this job is getting all the students to participate. We all fail, and you will fail too. Rather than just throwing a question out there, and waiting hopelessly for a student to pick it up (it's either the same student every time, or none at all) try breaking it up into bits, and go round the room asking each student one part. Sometimes this works.

Some new TAs are scared of being in front of a class full of students. Don't be. Most of the students are more scared of you than you are of them. And it's not like high school. The ones who sat at the back making trouble at high school mostly never went on to university anyway, unless they got their act together. And they aren't forced to come to class. But if worse does come to worst, and some are making trouble, and you have told them twice to stop it, then tell them to leave, and if they don't leave, then you leave the class yourself, immediately. Call security; it's their problem, not yours. This almost never happens. But knowing in advance what you would do if it did happen gives you a lot more confidence.

Some new TAs are scared they might be asked a question they don't know how to answer. Don't worry. This will happen, but it's nothing to be scared of. Just say you don't know. Nearly all students are very forgiving, if you just admit that. You can tell them to ask the prof. Or you can say you need to think about it, and will try to answer next week. Or you can say "Hmmm. I don't know. Let's see if we can figure it out together. I wonder if the textbook says anything that might help us?" (Actually, this can sometimes be a good thing to say even if you do know the answer.) Just don't BS the answer. And if you do say something wrong, which probably will happen, it's perfectly OK to say you said something wrong and are going to try again to get it right.

Speak slowly and clearly and loudly enough to be understood. This is especially important if you have (what Canadian students will see as) an accent. You might be perfectly fluent in English, but if Canadian students have never heard someone speak with an accent like yours they will probably find it very difficult to understand you. It's your job to compensate by speaking more slowly and clearly than normal, and to use the board as backup too. [I have a strong British accent, but I normally get by OK because Canadian students have heard British accents on TV. But I got badly caught out just recently. I said "So the elasticity is nought point five". The students all gave me a puzzled look. I checked my math, which was right for once, and gave them a puzzled look back. After a very confused discussion, we figured out they had heard me as saying "So the elasticity is not point five". Canadians say "zero point five".]

7. Office hours. For the first few weeks, almost no students will show up at all. So it's very tempting to not bother showing up yourself. Resist temptation. Just bring some of your own work to do if they don't show up.

Your main job here is to try to explain difficult concepts clearly and simply so the weaker students can understand them. This is hard. Try to use concrete examples. Do not say: "Country A has a comparative advantage over country B in producing good X if the opportunity cost of producing good X in terms of good Y is lower in A than in B." Instead say: "Suppose the opportunity cost of one apple in Canada is half a banana. Suppose the opportunity cost of one apple in the US is one banana. Then Canada has a comparative advantage over the US in producing apples, because Canada has a lower opportunity cost of apples than the US." Use different examples, and try to explain it as many different ways as possible: words, pictures, numerical examples, whatever, because different students learn in different ways. Go from the specific to the general; from the concrete to the abstract; not vice versa.

So, for those who have worked as TAs, what other advice did you wish someone had given you when you first started?

Comments

[Chris J]

You must treat students with respect and dignity. ==> And they must do the same to you.<=== If students harass, insult, belittle, threaten you, go to the department chair immediately.

[Nick Rowe]

Chris: in your experience, how frequent a problem is this?

[Jacques René Giguère]

I once had a drunk student. It was during the student carnival. And he was a gentle drunk. First time he had ever drinked...Took him to student services who got him home to sleep it off. He was so sorry the next time...

[Livio Di Matteo]

I'll be sending this link to my grad coordinator. Nice post.

[Felipe]

If you are grading with other TAs, you want to try to ensure you are all grading consistently.

At my university what we did was split questions across TAs, not tests. That way, some questions might be graded more harshly than others, but all students are graded by the same hand on each question, it should be fair.

For the first few weeks, almost no students will show up at all. So it's very tempting to not bother showing up yourself. Resist temptation.

This happened to me once, the day after one test (we don't have midterms, but 3 tests and one final exam). Usually nobody bothers going the day after the test. This time some students did. Very, very embarrassing (and lots of getting yelled at).

It's pretty obvious that you want to prepare, but specially for the first times standing in front of the class, I found it very useful to have my class printed in my hand (or the desk). And by "the class" I mean the entire needed material, written in a readable prose that would be fit for both reading out loud and publishing on the course's web site. This helped me build confidence that if I forgot something or got lost I could go check it and continue without much hassle.

[JeremiahStanghini]

I'm a TA at a school in the States, (but I grew up in Canada, so I'm familiar with the Canadian grading system). Beware: it's not the same in the US. Most of what you've said would apply to TAs in the US save for the grading.

[CJ]

How much do econ TAs get paid?

[Chris J]

@Nick - not often, but for a TA it is likely to be their first experience with workplace issues such as these so they need to be beacked up in advance. You see a few hundred students a year and one or two will be jerks.

Besides, not training your students in workplace safety and harassment policies is *against the law*.

[Nick Rowe]

Felipe: "At my university what we did was split questions across TAs, not tests."

That's a good way to do it, if possible, but it does take extra time and coordination. But even then I think it's good to have all TAs get together at the beginning so they are all grading equally. Because there is usually choice on exams, and you don't want some students being unlucky in choosing questions that get graded by the TA who grades the toughest.

[Donald A. Coffin]

Being a TA at West Virginia University in 1970 was somewhat different. We were asigned to *teach* two classes, not to assist a faculty member. What was hardest for me was to write good exam questions and to remember, when grading them, what an "average" entering college student was (I attended a small, private, highly selective,liberal arts college). My expectations needed changing. I remember talking at length with the faculty member who supervised the TAs about that, and asking him to review and critique my tests both before and after I gave them. That was about the only guidance I got as a TA, and I had to seek it out...things, in that respect have changed dramatically for the better.

[Nick Rowe]

Donald: Yep, teaching your own course is very different. IIRC, I taught my own course at Western in 1978 after one year of having been a regular TA/RA. I think that nowadays in Canada some TAs will go on to teach their own course after a couple of years of being a regular TA, but it's unlikely a new TA would be given a course of their own. (Am I right?)

New teachers/profs get a lot more orientation now. Yes, in the past we got nothing. I wonder if it does any good? (I once did an internet search for empirical studies on whether teacher training for schoolteachers helped their students learn more. I couldn't find much good evidence.)

[Determinant]

I was an undergrad TA in a first year engineering course for two years, it was a communications course, not very theoretical. But you had to do a group report. Ok, here's the issue. We do not expect you, young frosh, to be brilliant, much less original. But please, please cite your sources. My first group didn't, I saw the change of voice plain as day in the report and it was terrible. So the next four groups I did I gave them a "cheating is wrong" speech and gave them a web page on the citation style they were to use. It is formulaic, cookie-cutter, but it prevents nasty accusations flying around. I led them to water and they drank abundantly.

An ounce of prevention is really worth a pound of cure, especially as you paid real money to be here.

Second, on public speaking, please be enthusiastic. Show some interest. My parents are church ministers and "pulpit presence", as it is known in the trade, matters. Ministers were on to "the medium is the message" long, long before Marshall McLuhan ever was.

Third, on ESL. My brother TA'ed an accounting course for MBA students. My brother's English is perfect, clear as day, no slang, ordinary educated Canadian English as it is his first language. His supervising prof received a petition from the MBA class that his English was "too perfect", they couldn't understand him. They my brother to use more choppy, heavily accented English.

The prof laughed in the petitioner's face and told them that my brother was exactly what they would hear in the business world in Canada and they should get used to it.

[Frances Woolley]

Nick, great advice.

On your point #2 - it's not just that "what goes around comes around". What goes around comes around far more quickly than you could ever possibly imagine. This is especially true in Canada, where the degree of separation between any two economists is seldom more than two (they have a mutual friend/acquaintance) and is frequently one (they know each other directly).

My own two cents: what is taught in grad school is very different from what is taught in first, second, and third year undergrad courses. Consequently, TAs not infrequently find themselves in the embarassing position of not being able to answer the questions on the assignment or midterm exam. Expect to put in time to get up to speed on the course material. If you aren't sure of how to answer a question, ask the prof to explain it to you. It is far far better to confess your ignorance than to mark 200 exams incorrectly. As for the prof's opinion of you - well, the prof has seen lots of TAs before. Odds are you probably just an average TA like all of the others, and there is little you can do to faze him (or her).

[Nick Rowe]

CJ: It varies a bit across universities. At Carleton, a graduate TA gets about $10,000 per year, for 10 hours per week over two terms.

Frances: Yep, as I have said before, sometimes to howls of outrage(!), there's a lot to be said for reading a first year textbook. And that goes for grad students too.

Yep, it's a very small world. Your little first year student may be your future colleague or boss. One of our colleagues was once one of our first year students, and one day may be our boss.

[Brett Reynolds]

6. Leading Discussion groups/tutorials. Try bringing a set of discussion questions each printed on its own paper. Hand out one to each student (print two or three sets if you don't have a lot of questions or if you have a lot of students). Give the students some time to work out their own thoughts and then get them to find a partner. A asks B the question and supports B in answering (call over the TA if there's confusion or complete disagreement). Switch roles: B now asks A B's question. Finally A & B exchange questions and find a new partner. After this has gone on for some time, reconvene the class and debrief.

[tomslee]

It's a long time since I was a TA, but this would have been really helpful to hear at the time i started. It's a great list.

Number 1 is particularly important, because most grad students aren't there to be TAs of course. I know I wasn't. But in addition it helps to correct an impression that some parts of some universities do give. We had to push back as a TA union in the '80s against fourth year students being offered TA-ships in return for course credits, for example.

If I could add one thing it would be on leading discussion groups and tutorials. "However well you think know your material now, you'll know it far better once you've taught it. There's no better way to learn."

[Moebius_strip]

Plenty of this is good for new instructors as well.

I teach math, where there is less leeway in grading than there is in essay-based courses (though still plenty), and I would add this: "You are an authority figure, even if you don't feel like one yet. Stand your ground. Listen to your students and be kind, but don't be afraid to say no. If you accidentally added up a student's test mark incorrectly, or if you accidentally didn't give full marks to a correct solution, apologize and correct your mistake. But if you gave a mark of 2/5 on a question and the student thinks they should have gotten a 4/5, don't be afraid to tell them that it's not up to them to decide how much credit a wrong answer deserves."

[Simon Halliday]

Something else which is worth considering is the number of people in Econ grad school who themselves did not do economics during their undergraduate degrees. I've encountered, for example, someone who did engineering as an undergrad and is now a PhD econ student and being asked to TA UG econ and, say, mark an essay. Much chaos can ensue unless support is offered by the prof. Yet more support for your comment about reading first year texts & guides.

In a similar vein, I've found it important to encourage TAs to ask me questions about the content and not to be embarrassed/worried about stuff that they haven't covered themselves (even if they've done an econ undergrad). This is especially true of TAs who happen to be international students and who find themselves in a very different cultural context to the one they're accustomed to; students aren't always aware of their own beliefs about the 'right' way to teach, or solve a problem, etc, so the TA needs to be made aware in advance and it sometimes needs to be handled carefully. Some TAs seem to forget that the professor wants them to succeed & the prof will help them if they can.

[Nick Rowe]

Brett: that sounds like a good thing to try.

tomslee: "However well you think know your material now, you'll know it far better once you've taught it. There's no better way to learn."

I very strongly agree with that. And it doesn't stop after the first time either. I must have taught intro economics a couple of dozen times in my life, but I still learn something new every time I teach it. (Though I expect I forget stuff too.)

Moebius: I solve the problem by not allowing my TAs to change grades. Any student who doesn't like the grade can bring it to me for a second look.

And yes, if a student comes to my office querying something I have graded, the first thing I do is tell the student to sit down and stay quiet while I re-read the exam. If I think my first grade was too low I change it. If not, I *then* ask the student why he thinks it should be graded higher.

Simon: this is getting to be a bigger and bigger problem over time. And I think it is related to the increasing mathematisation of economics. They think if they can solve the equations for graduate economics courses they understand economics. It's not just that they can't explain it to undergrads; they don't really understand it. And they make stupid mistakes.

[Frances Woolley]

Nick: "And yes, if a student comes to my office querying something I have graded"

Like you, I don't allow TAs to change grades. For all sorts of reasons. I say "write on the front of the assignment/exam why you think you should get a higher grade, and then I'll reread the entire assignment. Your grade may increase or decrease." I find it very difficult to do what you do, that is, sit and read the exam and make an objective decision about it while an anxious student is watching.

[Felipe]

Nick: I solve the problem by not allowing my TAs to change grades. Any student who doesn't like the grade can bring it to me for a second look.

We had a similar system. One thing we did is that, after receiving their graded tests, students can resubmit them for grading, but this time not to the TA but to the prof, who would grade them from scratch. So the grade could go up or down. And another thing we (usually, but not always) required is to resubmit the whole test, not just the one answer with issues, so that grades could change in other questions/sections, again in either direction.

[Felipe]

Also, just as Frances says, the re-grading didn't (usually) happen with the students present.

[Mandos]

One piece of advice I'd give mostly reflects #4. In three parts:

1. What is being meted out to you, you once meted out.

2. What you are meting out, was once meted out to you.

3. You know the annoying keeners and grade-mongers who hang out around office hours browbeating you into giving them a few more points so that they can get that A/A+ on the course? *cough* There is a reasonable chance that you were one of them...

[Frances Woolley]

More on why native language speakers with "perfect" English can be difficult to understand.

One problem is the indirectness of English, particularly British English. See, for example, this image which provides a helpful guide. When an English person says "I'll bear it in mind" they mean "I've forgotten it already", rather than "They will probably do it."

Another issue is that English is a very idiomatic language. This morning over coffee I found myself saying, "It's an ill wind et cetera." Which means "It's an ill wind that blows nobody any good". Which means "That was a bad thing that happened, but at least someone benefited from it." It easy to be completely unaware one is using some idiomatic phrase - like nought point five - until one is met with blank stares and confused looks.

[Nick Rowe]

Frances: yep. Like when we were all teaching in Havana, and the Cuban students said they could understand the English better of those Carleton profs who had learned English as a second language. Fewer idioms.

[Joel W]

@Chris J"If students harass, insult, belittle, threaten you, go to the department chair immediately."

Not just students, it is important to let TA's know (especially international student ones) that professors cannot "harass, insult, belittle, threaten" them either. Most TA's in Canada are probably unionized, and it is important for them to know what is acceptable behaviour from their boss, i.e., the prof.

@Nick: As someone who has TA'd a ton of classes (graduate and undergraduate), I think your advice is very good. The one thing I would add is that it is very important to have a thorough discussion with the professor you are TA'ing for about what their expectations are for you as I have found expectations differ a lot between professors. I have had some professors that expected me to work 10 hours a week and to meet with them everyweek to ensure I have enough work, and I've had professors that expected me to do virtually nothing at all other than proctor their midterm exams.

[Moebius_strip]

@Nick - good policy on changing grades, and one that I implement now as an instructor.

Now that I'm teaching, I am realizing how inadequate my own supervision was when I was a TA. Despite your comment that no one hired profs for their managerial skills, I would have benefited from half the guidance you offer in this post. I was a grad student at UBC, where math TA's actually run their own classes (usually sections of first year calculus). I took a one hour "how to teach" seminar, was handed a textbook and a list of chapters to cover, and was let loose on 40 students. I chose what homework to assign, I designed my own tests, and did all my grading. At the end of the term all dozen or so sections of the course wrote a common exam that we instructor/TAs didn't get to see in advance. As you can imagine, a student's success in the course was heavily dependent on whose section they ended up in. (One fellow grad student decided to teach an overly theoretical course that had almost no connection to the introductory curriculum he was being paid to deliver; TA oversight was so nonexistent that the course coordinator didn't find out about this until a second grad student TA gave him an anonymous tip, two thirds of the way through the term.)

@Mandos: Yes to your #3, and it embarrasses me anew every semester.

[Ian H]

Cheating and collaboration: If you suspect students have cheated in some way, perhaps by collaborating on assignments, your job is to retain all relevant evidence and refer the matter to the professor. There are official procedures for such things but the official procedures will flow from the evidence that you retain. It is not your job to act as judge or impose some punishment. DO NOT ever accuse a student of cheating, especially in the presence of other students. Such incidents can arise when you are not expecting them and it is important to have thought about what your priorities will be in advance. Retaining control over the evidence can mean in some cases taking photocopies or returning photocopies to the student and retaining originals. Understand that the bar in terms of evidence of cheating can be set quite high. If the professor decides that the evidence you have retained is insufficient to take action then you must accept this decision completely even if you yourself are in no doubt that the student has cheated, but it would be reasonable to keep a quiet close eye on those students in future.

Appeal of Grades: Students may ask you to reconsider a grade or suggest that you have made a mistake in the marking or adding up of marks. And in fact because you are human it is possible that you may have done so. Make it a rule never to decide on such matters immediately. Get the student to note the question and nature of the claimed marking problem by writing on the front of the document and hand it back to you so that you can take it away and consider it. This lets you take time to think about the matter without the student in your face. Also be awake to the possibility that the student may have altered the document.

[Nick Rowe]

Ian H: That is good advice on what to do if the TA detects cheating. Policies may vary from university to university, so I was wary of including it in this post. Our policy is: TA takes evidence to prof, who takes it to the Chair, who takes it to the Associate Dean, who deals with it.

[Moebius_strip]

Re cheating, I think the takeaway point from these comments is that the TAs' supervisor should make a point of going over the school's policy with the TAs, and TAs should make sure they are well acquainted with said policy. I remember having to improvise the first time I caught cheaters, which leaves way too much room for error: my experience is that post-secondaries will not take action against cheating unless the case is airtight, which means that it's easy for an instructor unfamiliar with protocol to have their cheating accusation dismissed on a technicality.

[Jacques René Giguère]

And how could you prove cheating? Because they all have the right answer? or more probably the wrong one?

[Moebius_strip]

Jacques, here were the last two times I accused a student of cheating:

* In one class, I gave individualized assignments: everyone gets the same basic questions, but with the numbers changed. This way, students can collaborate but not copy off one another. This didn't stop a group of three students whom I knew to be close friends, who all handed in identical answers....quite a feat, as they didn't even have identical *questions*. All three students were given zeroes on the assignment in question, and protested but did not formally appeal.

* At the beginning of the final exam, I told students to put all of their notes away, and verified that they had done so. Then I came around with the tests and formula sheets. Half an hour later, I lifted a student's formula sheet and found, underneath, several pages of his own notes. This time the dean got involved and issued a warning to the student but no punishment.

Pretty clear-cut, I think, but neither got the severe punishment that I had been told as an undergrad was given to cheaters (expulsion, black mark on record).

[Ian H]

I suggested the cheating advice because exactly this got me into trouble back in my days as a TA. A student came up at the end of class when I'd just finished going over the test and said that I hadn't marked some of his questions. I looked at it and it happened to be a test that I specifically remembered marking (I have very good spacial memory) - and there were a bunch of correct answers in places where I distinctly remembered blank spaces used to be. Not having given any thought to these matters in advance I immediately blurted out the first thing that came into my head, namely "Those answers weren't there before!" - after which things went rapidly downhill. WORDS were exchanged. And of course I didn't retain the test.

Next thing I get a letter from the student's lawyer threatening me with a lawsuit for defamation. The Dean basically told me the faculty would not support me in any way and I was on my own. Eventually a meeting was organised between myself, the student, his lawyer and the Dean at which I was coerced into writing a letter of apology to the student for accusing him of cheating in front of other students. In light of the fact that there was a possibility that the test COULD have been altered however the grade was left unchanged. The whole experience was very unpleasant.

[Jacques René Giguère]

Moebius Strip: I also did that years ago. Much easier in an introductory course where you want them to compute a simple multipler with varoius propensity to tax, consume, import and so on. And a hassle to correct. My next week exam has 6 different versions...the basic one scrambled. My secretary hates me..or se she days with a smile ( I kind of garantee her job after all.

Long ago ( we were soldiers once, and young,). we had an assignment in Intermediate Macro. It was a reality-based problem , not a theoretical one. Our team, the B+, was stumbling around some solution but it didn't seem quite right. I suddenly remembered that it looked like something I had seen in a monthly bulletin from CityBank. ( It was before the web when you subscribed by mail so anyone under 30 can ask me what is that concept). Went home and there it was. So we used it and got an A while all the other teams, including the A and the A+ got a B. Commotion as the social order had been broken. Explanation required by the other students, as the prof knew what had happened ( he had askec to confirm his hunch). The A and A+ said we had been cheating. The prof answered that we had done our researc which they should have as responsible pro-in-training and didn't do...Our future bosses would expect result more than pure process (up to a point) Grades stayed.
Of course, on the next assignment , we got our usual B+.
Draw your moral where you think it should be.
BTW, the young at the time but already wise teacher went on for a fruitful career , including president of the CEA. The veterans of this blog will probably know who I am talking about.

[Moebius_strip]

Jacques, re scrambled exams, funny story about that: a former officemate of mine once gave 2 exam versions: same multiple choice questions but with the choices scrambled. Student X copied the answers off Student Y, who had a different version of the test. They handed in identical answers. Y got close to a perfect score; X got zero.

A day later, X went to my officemate to plead for mercy. He'd been having a hard week, he explained; lost his job, broke up with his girlfriend, got sick - the whole nine yards. He hadn't had time to study. Please could my officemate allow for a retest?

X spent fifteen minutes begging my officemate, but she wouldn't budge. She told him that his zero stood. At no point during this conversation did she accuse him of cheating, and at no time did he confess.

Incidentally, this was the best- and most successfully handled case of cheating I'm aware of.

[kutschke]

Excellent advice.

The best advice I received is that it was not my mission to convert first year students in a physics for chem/biochem service course into physicists. They had their own interests and career plans and we must respect them.

And a grading anecdote.

At Mac in the late 1970's, the first year chemistry midterms and exams were given on pre-printed pages with room for the answer following each question. After the midterm/exam had been graded and returned, surplus exams were made available outside the department office. Students were encouraged to pick up an extra copy and to use it as a practice test before the final. The chemistry department had one secretary whose main job was to be the master recorder of all first year grades; the secretary was empowered to correct an addition mistake made by the grader. The flaw in the system was that there was no feedback to the TA or the prof that a grade had been changed.

One of the students in the lab group that I TA'ed badly failed every test. He picked up one of the surplus test and, using the posted solution set, wrote a B+ or A- exam. He carefully arranged right and wrong answers so that it appeared as if the grader had simply missed a few of the
pages when adding up the final total grade. He then graded the exam himself, making the addition mistake and brought it to the secretary for correction.

He apparently ran this forgery scam in several other departments too. In the other departments grades were computerized but there was no change log. In my department the grades were kept in a large hard bound ledger book; the change log was visible in the string of erasures and whiteouts.

Someone in one of the other departments became suspicious but could not prove anything and raised the alarm. One day, just before finals, the department chair called me to his office and handed me the forged exam. He asked if I could tell if the grading had been forged. The student had made two mistakes that sealed his fate. First, he had used the wrong type of pen to do the grading. Second, we had not been given a scheme for partial credit and each TA made up his own. A few weeks prior to the exam I had made a big deal in class about about citing sources in a particular circumstance and a relevant question happened to be on the exam; I deducted 0.5% from the few students who failed to cite the source. The net result is that I would have given the prof's solution set only 99.5% ! The student had not known this and gave himself a perfect score on that question.

Several years later I learned that the student was not permitted to return the following year but would be permitted to reapply after some number of years had passed.