Take-home tests: Why?

September 29, 2005 10:56 AM

What's the point of a take-home exam?

A regular exam, completed in class, tests a student's ability to use knowlege acquired during classes and through studying. But a take-home exam tests a student's ability to look up the answers. Why does my prof want to know if I am capable of looking stuff up?

posted by scratch to education (40 comments total)

Possible answers (others may have more, and these may not hold in every case):

(a) your prof would rather use class time to teach more things instead of administering an exam

(b) take-home exams tend to be written in coherent typed English as opposed to handwritten scrawls, so they're somewhat more pleasant to grade.

This assumes the class is in the humanities, though.
posted by Prospero at 11:00 AM on September 29, 2005

I don't see it that way, however, it may depend on the subject. When I was in college (88-92), the take home exams I generally got were in Philosophy or Literature courses. In those instances, the point to my mind was to essentially make your final exam like a term paper. You were able to put more effort into the polish of what you were trying to say as you weren't rushed by the time limit of a class, however, you still had to make a case and could obviously not just regurgitate anything from your books.

In the case of science or math related courses, perhaps it allows the prof to give more in depth and sophisticated questions than would be fair in a class-based test environment?
posted by spicynuts at 11:01 AM on September 29, 2005

Probably the point is that a take home exam has no time limit. The prof is guessing it will take you longer than 3 hours to complete it. And that the questions wouldn't be so simple as to be solvable by looking up the answers.
posted by CrazyJoel at 11:03 AM on September 29, 2005

Wow, loaded questing. It depends on the format of the take-home exam. My college has a tradition of offering take-home exams whenever possible, with the understanding that the professor can limit both the duration of the exam and the resources available to the exam-taker. So, a three-hour take-home exam that is "open-notes, closed-book, closed internet" still tests the student's ability to use knowledge aquired during classes and through studying, since the problems will probably be much much harder than the questions that could be given during a 75-minute in-class closed exam.

Another benefit of take-home exams is that the students can take them whenever and wherever they see fit. Many students experience test-taking anxiety that negatively affects their performance. This can be mitigated by taking the exam in an environment they are comfortable in. Also, people are more mentally active at different portions of the day - what we call "morning people" and "night people". If you have a scheduled morning exam, isn't it unfair to people who aren't so smart in the morning, when it is just as easy to let them take it in the afternoon or at night?

Perhaps you could be more clear about the format of the exam, the professor's limitations.
posted by muddgirl at 11:05 AM on September 29, 2005

Far more time for students to work on the problems presented, allows more complete coverage of material and harder questions.

I've never had to write one, although I did have a four hour exam once...
posted by Chuckles at 11:05 AM on September 29, 2005

In the sciences,all my take-homes were for things that couldn't be looked up. They did not test facts, but concepts, and problem-solving. Alternatively they were of the "derive this equation..." or "prove this relation" type.

For me, there was little difference between a 'take home exam' and a 'problem set', except that the exam maybe had an earlier due-date.
posted by NucleophilicAttack at 11:06 AM on September 29, 2005

I think there is something to be said about guessing on an answer during an in-class exam or actually finding the correct answer in a take home.

I think sometimes the goal is retention of correct information and learning.
posted by Gooney at 11:11 AM on September 29, 2005

All the take home exams I had to write were fucking hard; you wouldn't simply be able to look up the answer to a question somewhere. They tested whether you had a solid grasp of the meterial or not.
posted by chunking express at 11:14 AM on September 29, 2005

My CMOS analog design course was a take home exam. It took 40 hours to complete and it actually tested (to failure) my knowledge.

I'd have been screwed if I was taking more than one course at a time with similar exams though.
posted by substrate at 11:25 AM on September 29, 2005

Yeah, what NucleophilicAttack said. Many exams tend to be more about "do you understand these central concepts, and can you successfully apply them?" as opposed to "do you remember all these facts?" The latter type of exam wouldn't work as a take-home, but the former certainly does.

Almost all of my upper-level physics exams were take-homes.
posted by Johnny Assay at 11:33 AM on September 29, 2005

I think, at least in the science/engineering classes I took, that they allowed the professor to test your ability to actually perform in the field. Consider that a working scientist has access to all the books, web sites, etc. that he wants. So, a professor makes a very realistic (i.e. difficult) test requiring not that you have memorized a bunch of stuff, but that you know the subject matter and how to apply it (even if you can't spew out the equations verbatim). That's my take on it, at least, and I think it's a godo way to test college-level students who are serious about entering the field.
posted by abingham at 11:34 AM on September 29, 2005

Does the exam intend to measure fact retention? Then an in-class exam may be useful.

A test intended to reflect higher-level understanding and application of concepts is much better suited to a take-home exam.

My favorite college courses had take-home exams, which were actually challenging and interesting. These tests often had questions which did not immediately seem related to the course material. You had to understand how the formulas you had learned could apply to different contexts.

I can look up facts on the job. I can't fake deep thinking.
posted by sonofsamiam at 11:37 AM on September 29, 2005

Cramming worked for me for in-class exams. I usually forgot everything within 2 weeks after the exam. For take-home exams, I actually had to understand the subject and know how to apply it, and that understanding didn't disappear so easily later on.
posted by fuzz at 11:49 AM on September 29, 2005

Exams should be a learning opportunity as well as a test of knowledge. Normally this is achieved by the review of notes, reading etc, prior to the exam and achievement of some sort of mastery of these, then a mind-dump at the exam -- point being - the studying actually reinforces the learning. A take-home exam fulfills the same function of combining teaching AND assessment, only more thoroughly.

I assign a complex take-home question at the beginning of each semester and the students can then slowly work on it as we cover the relevant topics in class -- I think it enables them to synthesize material with some sort of focus/goal in mind. however, unlike some take-homes, they have the question well in advance but they come into the exam room and answer it there, without notes, etc -- this helps police plagiarism etc. to some degree. And, it is only part of the final exam -- 25%
posted by Rumple at 11:51 AM on September 29, 2005

I think most of the respondents in this thread are right on. The main reasons I prefer to administer take-home exams are:

1) They test ability to synthesize information and marshall evidence in support of a clearly articulated argument, rather than rote memorization and the ability to regurgitate facts under time pressure.

2) They test students on real-world skills (which includes knowing where and how to look up information and what to do with it once you have it) rather than mere test-taking ability, which helps mitigate the fact that some students are simply poor test-takers.

3) They elevate the quality of the work, which makes grading easier and fairer (its difficult to fairly assess a student whose handwriting is so poor that you cannot understand what they have written).
posted by googly at 11:58 AM on September 29, 2005

What exactly is wrong with testing your ability to look up answers to complete an exam? In the real world, are you normally going to be in an environment where your only resources are a #2 pencil and a piece of scratch paper? Of course not.
posted by odinsdream at 12:03 PM on September 29, 2005

Interesting answers, all. This exam is for a core class in cataloging (MLIS grad program). I think Abingham and Sonofsam are onto something, though. The exam's not until December and I know none of the details yet. I was just surprised to find myself confronted by such a thing in a grad program, having associated take-home tests in high school with awful, lazy teachers and obtuse, equally lazy students. (Yes, I am arrogant and have a bad attitude. But boy can I ace those take-homes!)

Thanks to all who responded.
posted by scratch at 12:05 PM on September 29, 2005

I took a course in Applied Electromagnetics once where the professor not only gave us take-homes, he encouraged collaboration among the students. His test problems (there was usually only one per test) weren't carefully bounded to produce a definitive answer, but were generalized questions to force us to research the subject and brainstorm approaches to the solution. Best course I ever took.
posted by forrest at 12:09 PM on September 29, 2005

Odinsdream, personally I am quite confident in my ability to look stuff up. I had been assuming all take-homes were like the ones I had in high school--easy and pointless and based on readily available information. Apparently it just ain't so, especially in the sciences.
posted by scratch at 12:13 PM on September 29, 2005

I find that adult life is a lot more like a take-home exam than a test of your ability to memorize and regurgitate. Maybe your professor just recognizes that reality.
posted by pomegranate at 12:13 PM on September 29, 2005

having associated take-home tests in high school with awful, lazy teachers and obtuse, equally lazy students

Rather like how my feelings were about multiple guess tests, at least the way they are usually set -- any easy out for profs who prefer not to grade exams, easy on the students who can use guessing strategies on the ones they're not sure of.

Then I took a class with an economics professor who set really hard multiple choice tests: all the answers were plausible (no easy dumping of obviously wrong answers), none of the above was the correct choice more than 10% of the time, and so on. Fiendishly hard. It really made you pay attention.

I can't say I came to like the tests, but I did come to the realization that easy/hard, long retention/quickly forgotten, and so on have less to do with the format of the tests themselves and more with how they are set, how it integrates with the material, teaching style and so on. The same with open book -- yes it can be a cop out, but there's no reason why it needs to be, at any level.
posted by Quinbus Flestrin at 12:31 PM on September 29, 2005

As a little warning from a T.A. who has had to mark more than a few take-home exams:

The marking standard you'll be held to in a take-home is likely much higher than a regular format exam. Take home exams generally mean that profs/TA's expect you to do some serious critical thinking on whatever your subject is, and give very detailed answers.

We use them (in philosophy) as a test of critical thinking skills. Anyone can regurgitate from the lectures or texts. Perfect regurgitation earns you a C+, tops. What they want on a take-home is real engagement with the material.

I hope that helps answer your question.
posted by generichuman at 12:36 PM on September 29, 2005

IIRC, my core cataloging final was take-home as well. I think, as others have said, to simulate the actual environment of cataloging material without fighting over limited resources. Actually, come to think of it, almost all my graduate exams, in both Library Science and English, were take-home. I think the philosophy was either that the answers would all be idiosyncratic enough, like an essay, that you couldn't cheat and copy, or that you were more motivated to learn the material than your average el-hi or college students. If you did copy catalog records from a friend, then you wouldn't know how to catalog, and wouldn't last very long in a cataloging job. Ultimately, it's your ass, not theirs, so why should they babysit you to make sure you learn anything.
posted by bibliowench at 12:44 PM on September 29, 2005

I find that adult life is a lot more like a take-home exam than a test of your ability to memorize and regurgitate.

Pom, yer makin' me chuckle. My adult life involves more memorization and regurgitation than a bulimic chorus girl could handle.

If you did copy catalog records from a friend, then you wouldn't know how to catalog, and wouldn't last very long in a cataloging job.

Well, I don't particularly want to be a cataloger, so that's not a concern.

Ultimately, it's your ass, not theirs, so why should they babysit you to make sure you learn anything

'Cuz I'm paying them $675 a credit, that's why.
posted by scratch at 12:53 PM on September 29, 2005

An essay question is for finding out your ability to think through and analyze the issues. The ability to look up the facts doesn't necessarily help you with this.
posted by winston at 12:54 PM on September 29, 2005

scratch, what is your job?
posted by odinsdream at 12:56 PM on September 29, 2005

'Cuz I'm paying them $675 a credit, that's why.
No, you're not paying them to make sure you learn the material. You're paying them for the exposure to the material and an evaluation of your progress. If you can't put forth the actual effort to do any work without a teacher looking over your shoulder, you'll have more problems in your field beyond not knowing your way around LC classification rules .
posted by bibliowench at 1:02 PM on September 29, 2005

This exam is for a core class in cataloging (MLIS grad program).

Ah, all the more reason then. You'll need to find DDC, LCC, and LCSH for these, and your school probably doesn't have enough copies of these for everyone to use at once. Even if it could be an in-class exam, I'd think it would still have to at least be open-book, given the many many rules in AACR2 and all the LC interpretations. No one could possibly expect someone who's only been cataloging a few months to know all of those without looking them up. And once you're giving an open-book exam anyway, it's not that much of a step to a take-home exam.
posted by DevilsAdvocate at 1:35 PM on September 29, 2005

Odinsdream, I'm wary of answering, because I feel I'll be ambushed a la Bibliowench (and I'll get to you in a minute, hon). I'm a copy editor. I look stuff up all the time, in fact.

Biblio: I disagree. That was the point of my original question: it was unclear to me how a take-home exam can measure my progress. I love being exposed to the material, and I enjoy going to class, and I enjoy studying (and learning). Kindly refrain from making assumptions about my "actual effort."
posted by scratch at 1:44 PM on September 29, 2005

It's kind of hard not to make that assumption with the attitude of "I'm paying them, so they need to do things my way." If you do the take-home as instructed, the teacher can measure your progress and correct your errors. The only benefits to an in-class test would be the time constraints and restricted access to material (and DevilsAdvocate is right on with the limited materials). If you cheat (and I'm using the universal "you" here), then you get as much out of the class as you put into it, and in a professional degree program, you're shooting yourself in the proverbial foot.
posted by bibliowench at 1:51 PM on September 29, 2005

But a take-home exam tests a student's ability to look up the answers. Why does my prof want to know if I am capable of looking stuff up?

A good (or even perhaps a mediocre) take-home exam does not test a student's ability to look stuff up. It tests students' critical think and problem solving abilities, in a way that can actually be difficult to do in an in-class exam. It also, at least in much of my field (linguistics), tests understanding of the material in a way that I think goes much deeper than what an in-class exam can do. In fact, most proponents of take-home exams would probably argue that in many cases it is in-class exams which are pointless - in not all subjects is memorization ability a useful indicator of understanding, learning, or ability. There is a lot more to knowledge than facts (at least, in a good class, in my opinion).

I respectfully suggest that if you approach your exam in the way you assume is correct (looking stuff up and regurgitating), you likely won't do as well as you think you will. Of course this may vary by topic, etc. etc.
posted by advil at 1:52 PM on September 29, 2005

There are many ways to catalog a given book. Cataloging has the "If it's a computer book put it in the 000's" part of it, but it also has the "Is this more about travel, or more about Burma? Is it more about plants or more about farming?" part of it. The former types of questions you can look up. The latter types of questions you'll have to really understand the gestalt of cataloging to be able to answer effectively, no matter how much time and how many resources you're given. The idea of "aboutness" in cataloging means that there is more to a cataloging test [a good cataloging test] than just copying down DDC/LCSH. I'm so happy there are so many librarians on MetaFilter!
posted by jessamyn at 1:58 PM on September 29, 2005

It's kind of hard not to make that assumption with the attitude of "I'm paying them, so they need to do things my way."

I never said they had to do anything my way or your way or Bozo the Clown's way. I was simply asking about the take-home because I found it surprising.

Advil--I suppose I was assuming that the correct way to take this exam would be to look stuff up, but it turns out I have previously been exposed to only the worst kind of take-home test. Now I'm armed with more knowledge and when I get an A, I'll be sure to come back to the green and tell everyone.

Don't let's fight, folks. I asked a question and received a lot of interesting answers. Anyway, the whistle's blowing and it's Miller time. See you all tomorrow.
posted by scratch at 1:59 PM on September 29, 2005

I don't mean any disrespect, scratch, but your question suggest you have the typical bias about exams (which is unsurprising, because most schools encourage this bias). You think exams are about PROVING you've mastered the subject so that they teacher can provide you with a fair GRADE. Granted, this IS how exams are used in most schools. In my opinion, this whole approach is misguided.

The purpose of school SHOULD BE to help you learn. Everything that goes on in a school should serve this purpose, including exams. So, how could an exam help you learn?

a) by presenting you with difficult problems to solve. We learn when we try to solve problems.

b) by giving your teacher feedback about what your current strengths and weaknesses so that he can can adjust his teaching to serve you better.

(Note: in the ideal school, (b) wouldn't be needed. Your professor would know how to teach you, because he'd be working closely with you throughout the course.)

Unless performance-under-stress in integral to the subject, it's actually BETTER to give a take-home exam then to make you complete an exam in class. In general, we solve problems better (and learn better) in a place where we're comfortable and at our own pace.

You might object that take-home exams open a wide door to cheating. I agree, but I don't care. I would care if I agreed with the (unfortunately popular) FORCING and POLICING method of teaching. I deeply disagree with it. I think it fosters about the worst possible environment for learning.

If you take home an exam and cheat on it, then you won't learn for the exam. It's as simple as that. It's your choice. It shouldn't be all tied up with a parent-child, policing scenario. A teacher should provide opportunities to learn which you should be able to avail yourself of (or not).

In SOME situations, people DO need to be assessed in order to determine whether they have mastered a skill -- i.e. in hiring situations. Exams are, perhaps, one way of making these assessments. But assessing (for purposes other then teaching) should be totally divorced from the teaching environment.

Final exams are completely insane. What's the point of giving an exam if there's no time left to utilize the results for further teaching and learning?
posted by grumblebee at 2:18 PM on September 29, 2005

Don't let's fight, folks.
Agreed. Looking back over my initial answers, I can understand why you'd take offensive to them - "your ass" was not your specific ass, but the hypothetical student ass, and I didn't assume that you would cheat. But the logic behind in-class essays seem to be based, in large part, on the assumption that students need to be under surveillance in order to perform honestly on a test, and I was arguing that, in a professional skills class like cataloging, why waste a class period on people who don't care enough to learn a skill that will cause them trouble later in their careers if they don't know it?

I should have used "some students" as my subject, not "you."
posted by bibliowench at 2:32 PM on September 29, 2005

I think the idea is that it's more like what you do in the real world - you do have resources on hand, but you'll be evaluated on how you use those resources.

Plus, being good at looking things up is always important in an MLIS program!

Good luck on the exam!
posted by srah at 3:05 PM on September 29, 2005

I don't give tests anymore, just fiendishly difficult homework. And I tell the little darlings I expect them to work together. They learn a lot more this way, and it really lowers their stress level.

When I was giving tests, I would usually try to do something amusing:

- Multiple choice tests where the correct answers (A - G) would sound out a tune, like Dies Irae or the university alumni song. One time I set it up so that the answers produced the Serbian national anthem, for the benefit of the one Serbian student in the class. I later learned that I used the older commie anthem. Whoops!

- Multiple choice test where every answer was "C". That one was a final exam. It really blew their minds - if you're not expecting it, you have to have a lot of confidence to put down the same choice for every question.

- An exam where every answer was the same as the problem number. (The answer to #1 was 1 volt; #2 was 2 ohms, etc.) Twelve people out of 90 figured it out - the two smartest guys in the class, and ten near the bottom. Hey, pattern recognition is an important skill too!

- My favorite: every answer was "666". One guy was just about in tears, said he thought his calculator was possessed. If I ever write another textbook I'd like to set up the end-of-chapter problems so all the answers work out to be 666.
posted by Wet Spot at 5:51 PM on September 29, 2005 [1 favorite]

Advil--I suppose I was assuming that the correct way to take this exam would be to look stuff up, but it turns out I have previously been exposed to only the worst kind of take-home test. Now I'm armed with more knowledge and when I get an A, I'll be sure to come back to the green and tell everyone.

I didn't mean to attack whatever take-home tests you might previously have taken, all I was trying to point out was that your assumptions about what the normal purpose of take home tests is may be wrong. I've taken many take home tests when I was an undergrad (it is the norm for my field; perhaps I am generalizing too much outside of it), and exactly none of them involved simply looking things up. They typically involve problem solving of a complexity that goes somewhat beyond what could be easily given in an in-class exam (and reasonably expect anyone to finish in time). Anyone who tried to do these tests based simply on looking up and extracting text from their notes/previous hwks/textbooks/etc. would not have done well. I apologize for extending qualitative judgements about this kind of exam outside what I really know - let me rephrase my claim: a linguistics take-home exam that just involved looking stuff up would probably be a bad exam, and someone who took this strategy to complete the exam wouldn't do well. Let me add that some of the take home tests that I took were by far the most rewarding (for me as a student) examination experiences that I had.
posted by advil at 6:50 PM on September 29, 2005

Take Home Exams/Tests are held to whatever standard your prof sets. We were told (1998-2002 here) when notes, books, or other resources were okay to use and when they were not. Typically the prof expected you to take your time and write a good exam -- typewritten, coherent, no grammar boo-boos, etc. You get to choose your environment, so you're writing your exam on your bed, in your study carrel, or in front of the TV --- and that is nice TO YOU. If you work well at 3 am or with Law and Order going 24/7, here's your chance to use caffeine and Lenny Briscoe to fire up your creativity. The profs at my school LOVED take home tests.
posted by Medieval Maven at 8:09 PM on September 29, 2005

Multiple choice test where every answer was "C".

Wet Spot, I have forsaken my god in favour of you. Please let me know where I can find an appropriate shrine to make burnt offerings.
posted by Johnny Assay at 8:03 AM on September 30, 2005