What are your opinions on collecting homework in college?

[MetalManuel]

Not projects, but just regular homework. It always baffled me. If the homework counts as 10% of the grade, then a student could get 100% on all the exams and get a B, and a student could get B's on exams and get an A.

 

[jtbell]

So? Are you implying that the only thing that should count are exams?

 

[streeters]

It acts like a buffer in case good students screw up in the exam.

 

[Mmm_Pasta]

Homework can be good for practice and it also is a good way to ask questions that wouldn't otherwise be put on an exam.

 

[MetalManuel]

So? Are you implying that the only thing that should count are exams?

Yes. If the final exam isn't cumulative then all the exams should be equal percentages. If the final exam is cumulative then I think it should count as a lot more than the other exams. That's just me though.

Homework can be good for practice and it also is a good way to ask questions that wouldn't otherwise be put on an exam.

I agree, but I believe that homework should be optional and professor should still assign recommended homework. Some students need more homework than other students, some might not need any. I'm not against homework, I'm just against homework counting as part of a grade. It just seems somewhat childish in my opinion.

 

[fss]

Homework **is** optional. Making homework count as part of the final grade rewards students that do their work and penalizes those that don't. Seems reasonable to me.

 

[MetalManuel]

Homework **is** optional. Making homework count as part of the final grade rewards students that do their work and penalizes those that don't. Seems reasonable to me.

So do you believe a grade is based on how hard a student works and not on how well they learned the material?

I also would agree with you if a class was the same difficulty for all students in a class, but in real life, for some it's easy while for others it's difficult.

 

[Jack21222]

So do you believe a grade is based on how hard a student works and not on how well they learned the material?

I also would agree with you if a class was the same difficulty for all students in a class, but in real life, for some it's easy while for others it's difficult.

Do you know how I know you've never taken an upper division physics class?

(I'll give you a hint: they don't allot 12 hours of class time for an exam)

 

[Nabeshin]

I'm of the opinion that homework should count for ~50%+ of the total grade for physics classes (at least beyond the introduction ones). The kind of physics questions which are really worth asking take hours to solve, and often times have additional parts like writing computer programs. This is simply not something you can replicate in a 1.5hr exam period.

For the same reason, I'm also rather a fan of take home exams (where you have, for instance, several days complete the exam, something like a super problem set). The time constraints imposed by conventional exams have never seemed to me like a legitimate way to test a student's knowledge of physics.

 

[MetalManuel]

Do you know how I know you've never taken an upper division physics class?

(I'll give you a hint: they don't allot 12 hours of class time for an exam)

Sure I've never done that, but I will in the future. I don't see how 12 hours is relevant to exams. Yes homework could take that long to solve if you've never done homework like that before, but what does that have to do with exams?

I'm of the opinion that homework should count for ~50%+ of the total grade for physics classes (at least beyond the introduction ones). The kind of physics questions which are really worth asking take hours to solve, and often times have additional parts like writing computer programs. This is simply not something you can replicate in a 1.5hr exam period.

For the same reason, I'm also rather a fan of take home exams (where you have, for instance, several days complete the exam, something like a super problem set). The time constraints imposed by conventional exams have never seemed to me like a legitimate way to test a student's knowledge of physics.

I guess what I am saying doesn't apply to upper division courses. In that case, it will make them easier to pass. I'm looking at the syllabi for a lot of classes and it's similar to what you're describing. I guess the homework assignments take so long that they basically become projects, and usually group projects. I doubt most students work alone in these classes.

 

[cjl]

Sure I've never done that, but I will in the future. I don't see how 12 hours is relevant to exams. Yes homework could take that long to solve if you've never done homework like that before, but what does that have to do with exams?

An exam that properly tests on upper division or especially graduate level material (in the engineering or science fields) in a comprehensive fashion could easily take 6-8 hours (or more), even if you know what you are doing. Honestly, this is why many of my graduate courses use take home exams and projects more than in-class exams.

 

[Choppy]

There are a number of problems with basing all of the course mark on a single final exam. These include:
1. Student stress.
2. Not everyone performs optimally on any given day. People get sick or distracted by things that while they wouldn't qualify for an exemption, don't allow them to fully concentrate. I once broke up with someone not too shortly before an exam.
3. As you progess in your studies you will learn that it's virtually impossible to cram all the concepts that need to be covered in a class over brief exam period (as Nabeshin and cjl have pointed out).
4. How would you grade lab courses?
5. As a professor, you have a certain responsibility to do what you can to ensure that people will learn the material - across a heterogeneous distribution of learners. If the homework is optional, few people will do it. Then at the end of term you will be faced with a pile of complaints and suggestions that you make the homework manditory. In my experience, homework weighting tends to be about making the assignments heavy enough that most people will do them, but light enough that the copiers will still fail out in the end.

Here's what I don't get. If you don't need the homework to learn, doing it should be pretty easy - taking only the time to actually write out the answers off the top of your head. So why complain?

 

[fluidistic]

At my university the final exam is worth 100% of the grade.
If you fail 2 of the 3 tests before a final, the final has either more exercises (and no more time: 4 hours) or you get a more severe grading. I.e. they remove full credit of an exercise for a little algebra mistake.
I don't really like this system: it's very stressful and sometimes I can't sleep more than 2 hours the night before the final.

 

[Jack21222]

Yes homework could take that long to solve if you've never done homework like that before...

Hahahaha! Oh, wow, that's the funniest thing I've read all day.

Only the best students in my E&M (Griffiths level) class were able to complete the homework sets in only 12 hours. Many of my classmates reported taking 16 hours or so. For 8 problems. So, even if you only had one "full" problem from each chapter on exams, and you cover 8 chapters, that's 16 hours worth of exam time. In a 2 hour class, that's 8 classes. In your system, you'd have to eliminate a month's worth of class time to take exams.

Instead, it makes sense to put more "simple" problems on the exam, straightforward ones that can be completed in 20-30 minutes, and leave the "real" problems for homework sets.

I don't see how 12 hours is relevant to exams.

I hope you see now.

Homework counted for something like 35% of the grade in that class, if I recall correctly. Intermediate Mechanics (out of Thornton/Marion) had homework count for 50%. In the Quantum Mechanics class (Griffiths) that I'm taking now, homework is 43%. The homework sets for this class are only taking about 6-8 hours or so (so far, anyway), so we're lucky.

Anyway, I think you should take some real classes before you do any more complaining about homework grades.

 

[deluks917]

I would prefer if the total course grade was based on a single 6 hour exam. I imagine this would not be popular among most students.

 

[johng23]

The problem with homework counting for a lot is that some students will work in big groups and basically get all the correct answers because of it. I never liked working on homework with others. There are also some professors who assign tedious work which takes many hours to do even once you feel that you understand the concepts reasonably well (this may happen more in certain engineering courses).

I think homework should be graded such that it's important that you do it and spend a reasonable amount of time (I feel I spent anywhere from 5-15 hours depending on the class), but you don't need to grub for every last point, and so long as you do well on exams you will get a good grade even if you didn't obsess over every last point on the homework. Maybe I've been lucky, but that's usually how homework has been graded in courses I've taken.

 

[symbolipoint]

Not projects, but just regular homework. It always baffled me. If the homework counts as 10% of the grade, then a student could get 100% on all the exams and get a B, and a student could get B's on exams and get an A.

?

Most typically, students who do not do their homework (assigned or not, collected or not) do not learn adequately.

 

[DeadOriginal]

Not projects, but just regular homework. It always baffled me. If the homework counts as 10% of the grade, then a student could get 100% on all the exams and get a B, and a student could get B's on exams and get an A.

Assuming homework counts as 10% of the grade and exams count as the leftover 90%, a student that got all B's on his exams wouldn't be able to get an A. Assuming B=80%, 80% of 90 is 72. If the student did all of his homework, his final grade would be 82 which is pretty much also a B to me... All in all, the student got what he deserved.

 

[Jack21222]

Assuming homework counts as 10% of the grade and exams count as the leftover 90%, a student that got all B's on his exams wouldn't be able to get an A. Assuming B=80%, 80% of 90 is 72. If the student did all of his homework, his final grade would be 82 which is pretty much also a B to me... All in all, the student got what he deserved.

A B can also equal 89%. 89*.9= 80.1. The full 10 points from homework would give a 90.1.

We're talking about one percentage point difference between the final grade and the exam average. I really don't know what the OP is so upset about.

 

[fss]

So do you believe a grade is based on how hard a student works and not on how well they learned the material?

Not sure how you reached that assumption. I'm of the opinion that if two students get equal grades on a test, the one that does their homework will get a higher grade than the other who doesn't.

I also would agree with you if a class was the same difficulty for all students in a class, but in real life, for some it's easy while for others it's difficult.

If a class was the same difficulty for all students grades would be meaningless. University is an "all people are equal" environment. The top students earn the top grades. Those who struggle will probably get lower grades.

 

[Andy Resnick]

So do you believe a grade is based on how hard a student works and not on how well they learned the material?

Do you think there is no correlation between the two?

 

[clope023]

Do you think there is no correlation between the two?

Sometimes there is, sometimes there isn't.

 

[fss]

Sometimes there is, sometimes there isn't.

For the fat part of the bell curve there is a very direct correlation.