Most exams I grade are not multiple choice.

Yes, be nice to the one who provides your T&A...

Oh, that wasn't what you meant? Sorry.

@carpenter

I'm not complaining. I'm just saying, now that I know what TAs do, I wish I had been nicer to mine.

I didn't make the exam. The first time I saw it was when I was handed them to grade. The solutions provided were poorly done and I needed to redo them.

I did have a time restraint. The professor wanted them back this morning.

The exams are the same, but the answers weren't. Because I actually care about my students, I go through, line-by-line, checking their math to see if I can give them partial credit. That takes time.

These exams sure as fuck aren't multiple choice. Their short answer and calculations, every one of which needs to be carefully read.

Again, I'm not complaining. I'm just saying I didn't know grading was so hard.

Who provides the TAs with T&A?

@abg: I meant, there's usually a structure in a question and since most students at least go through some of the slides or to the lectures you could assume that to a certain extent questions are approached in the same way. And based on this common approach you could make a correction sheet. This leaves 3 to 4 possible ways to get to the answer. And you're right, I had no clue about additional constraints on your part.
@Putin: I'm still wondering what people are testing with a multiple-choice questions.

It still takes forever to grade 40 papers sheet or no sheet, especially I imagine for math where you have to check all the work and monitor their "p's and q's" so to speak. So I don't get why you're giving Abgemacht such a hard way to go.

Some math depts, or at least the one here, has their TAs all grade the very same set of questions for each set of exams for the entire department, to monitor against bias. Hard as hell to coordinate though, have to turn in the exams much much sooner.

@carpenter

The problem is, not all wrong answers are equally wrong. So, for every question on every exam, I need to think about how wrong the person was.

For instance, depending on the parameters, different equations can be used. If a person uses the wrong equation but does everything right, I need to decide how wrong that is.

I would assume you also give credit for later work that is correct on a problem if they just screw up something in the early part of the solution (like 2+2=5) whihc means you have to solve it from their screup forward and look to see if they have another screw up.

I give them partial credit if their error propagates through. Being right is important as an engineer.

If they get an answer that makes no sense and don't mention it, I don't give them any credit, because it shows a complete lack of understanding.

@ Draug - "Who provides the TAs with T&A?" Whether or not someone is, it doesn't sound like they have time to enjoy it.

That's what I meant by "give credit". I didn't mean full credit as if they had gotten it right. I meant partial credit for understanding the problem and having the right conceptual solution, but points taken off for the error.

I feel your pain. Three of us went through about a 120 papers of Advanced-Computing a couple of months ago, and it was horror. I've had to check papers for two courses so far, one math and one CS. I've found math a bit easier because there is rigor, and most answers have a finite number of steps, so you can sort of deterministic-ally assign marks. Checking code is terrible. Variables don't get declared, and when code is written on paper, type conversion is free-for-all, so I have to make several passes before I understand what the guy's trying to do before I can conclude he's wrong.

Can't believe I missed this thread!

Want to take it even further? Try grading ~60 CS paper exams. Pages and pages of handwritten code, with partial credit given. The word 'nightmare' doesn't even begin to describe the pain experienced as a TA in the CS dept.

My favourite is no one showing up at hours until the day something is due. And they just expect you to look at their code and be able to read their mind.

Oh come on! You guys have "objective answers" at the end of the line. I've TAed in philosophy for just about 8 or 9 years before teaching my own classes! Trust me: going through impossible arguments with flawed structures and poor syntax while trying to be as generous as you can is no small feat. Doing it 80 times over is even worse!

It's not as objective as one would think. The equations we use cannot be analytically solved, and so a lot of the class deals with approximation techniques. It's not always trivial to determine if a particular approximation is valid. And even if you use 2 equally valid approximations, you could get very different answers in the end.

Right.
Same thing with code.
Some of these students come up with answers that even I didn't know would work. So it's as much a learning experience for me as it is for them. Which I love, but I definitely agree that students just don't know what the TAs do for them.

The most annoying of the lot are the ones where the question tells them where to start from and where to get. Some solutions follow the pattern

1. Start from knowing starting point
2. Some logical, but irrelevant steps
<-------Jump logical chasm here----->
3. Some logical, but irrelevant steps from the other side
4. Arrive at known ending point.

Sometimes it's not clear this is what is done, and I doubt it's
always accidental. On the other side of the spectrum are the
delightful chaps who don't attempt a question they can't answer,
I have nothing but good things to say about such people, and
being able to instantly move this paper to the checked pile is a
wonderful feeling.

Ok. I did say "at the end of the line"!!!

Though I would not pretend for one second that I know what you guys are talking about, I have a few friends who have studied long and hard in the field of mathematics. I also have read a fair bit on the relation between creativity and solving mathematical problems, and I've come to understand that proofs can sometimes have an "artsy" side to them (though I disagree that they are "creative" for all that, but that's an other debate).
You will grant me, however, that you generally know whether someone's got it right or not once you look at the "final" line of their answer. The rest is also something you need to evaluate, but it's really hard for a student who's got the wrong final line to argue long and hard about his getting it "somewhat" right. I would assume that the TA can present a fairly obvious argument in these cases.

Not so much in philosophy, where some issues can only be solved interpretatively and where arguments abound to support competing views on one and the same issue. Granted, undergrads rarely face such issues, but good ones (or intuitive ones) sometimes poke at them. That's the shit that makes being a TA so hard.

That, and the prof presenting the shit wrong... !

So, all you TAs, with your years of doing assignments as well as marking them, is it worth it to just skip questions you don't know how to do? I realize that it's easier for the TAs but I sometimes feel that I should at least try the question. Which is better overall?

Always attempt the problem. I'm never happy to see a blank page.

I TA a course on Semiconductors.

"I've come to understand that proofs can sometimes have an "artsy" side to them"

Proofs can indeed be beautiful. Either because they are so simple, or they use a trick rarely thought of, or they enhance the understanding.

"You will grant me, however, that you generally know whether someone's got it right or not once you look at the "final" line of their answer."

Not always. As I said, because these are approximations, they can get rather different answers. Also, for the questions that are short answer and not math, they may say something that I don't immediately know if it is right or not, and have to go look it up or think about it for a while.

@gman, you should definitely try.
@abge, +1. Math here. CS sounds terrible, Geofram.

@gman

If the page is blank, it's very easy for me to grade, but the chance of you getting credit is 0. If you put anything that shows any comprehension, I'm likely to give you partial credit.

(@gman, more generally, you should try to write down everything you know. Doing so, you might actually figure out how to proceed).

I usually advise the students to first move to answer all the questions they KNOW they can answer. Once that's done, spend the rest of your time on the stuff you weren't so sure of. And if you still can't figure it out, try with your most intelligent guess.

Like Geo then, I would tell you to never leave a blank.

Oh, and @cachimbo, I'd love to hear your argument on why proofs don't involve creativity. :-)

Also, when doing HW, it's easy to fall into the mindset of "oh, this is too hard, I'll leave it blank." If you force yourself to put an answer for every question, you may be surprised how many you can plow through.

@Abge: I said GENERALLY!!!! Give me a break why don't ya?!?!?!?

@Semck:
That will be in my next book. Right now, I'll just get by to writing the groundwork to that argument by finishing my stupid thesis!

lol

I'm not trying to give you a hard time; I'm just answering your question. People not in the hard sciences seem to think that every answer is black and white and that's really not the case.

@gman Yes, of course. Novel solutions, or genuine attempts at solutions are always good to see. But don't try to make your solution look like one by sneaking an incorrect step in the middle if you know it isn't correct. Sure, this could happen by error, but it happens way too many times on harder problems to be error all the time.

I know it's not. I was trying to say that without presuming too much, but yeah, I know it's not the case.

But you'll grant me this: most every student thinks that grading papers in my field is an arbitrary process (there's nothing objective) while marking papers in your field is an absolutely and rigorously objective process. You and I both know that's bullshit, but I would still argue that there is *some* (if just a very little) truth to that prejudice. That is to say that there are objective criteria that apply when I mark papers, but it's more often left to my interpretation of how the particular case fits under that criteria. Conversely, your grading is far more objective and, I believe, GENERALLY devoid of a need for sustained interpretation. It's not to say, however, that it is always thus. And I would even readily admit that in some particular fields (fundamental math, for example) it tends to look more like grading philosophy papers.

"But don't try to make your solution look like one by sneaking an incorrect step in the middle if you know it isn't correct"

As usual, there's an XKCD on this, but I can't seem to find it.

@Cach

Yes, I would agree with all of that.

I will now post a closing comment:

Teach your own classes. Get your own TAs. There is no other way.

I first loved being a TA, probably because I worked for some really cool profs who would let me teach a class here and there and those were great learning experiences. Once I moved to a different Uni for my PhD, however, I rapidly found out that my teaching style clashed with that of most old farts there. I rapidly started to hate TAing. But all that was quickly forgotten the second I started teaching my own classes. And when I was finally awarded TAs, I almost cried in joy!

Those that can, do... Those that can't, teach...

<* ducks and runs for cover *>

Just kidding of course. :-)

lol

well, yeah, of course. But, I'm a Master's student, so I have neither the time nor the experience to actually teach a college course. (That's not true; I could teach this course, but I wouldn't expect to get hired for it).

@Draug

There is some truth to what you say. If you're a bad teacher, it's often hard to tell if you are bad at teaching, or if you don't understand the material yourself.

Because we often have low thresholds for teaching skill, it's very possible that there are people who actually can't do what they claim, and pass it off as being just bad at teaching, which, for some reason, our society accepts as less bad.

@Abge: When I was doing my masters, I asked the prof I was working for if he'd be up to letting me teach a portion of the class. That was my first teaching experience as a "prof" and I got fiercely hooked! If you want to get exp, and you feel you can ask your prof, that's something you might want to look into.

@Draug: here, in Québec, we have a different schooling system where undergrads have to go through CEGEP, which is between High School and Uni and lasts 2 years. Then it's generally 3 years for your bachelor. All that to say:
Those who can, do "teach at the uni level"... Those who can't, teach at the CEGEP level! (you only need a masters to teach at that level and philosophy is mandatory in CEGEP so you need lots of profs).

@Cach,

Yes, my professor has been very supportive of me teaching. I've already taught 2 lectures and I wrote 2 questions for the exam.