On-line vs. written homework assignments

[luriol]

Recently most of the physics textbook companies have been pushing on-line homework systems to accompany their books. (e.g. Wiley plus). I've tried assigning these to my class, and while I thought the on-line problems were good, the students absolutely hated them, and furthermore I found zero correlation between homework scores and exam scores. I was curious if other people have had experiences with this which they would like to share.

 

[D H]

It is often very obvious when a student who comes to this site for help has been given assignments on an on-line homework system.

The reason it is obvious is because those students ofttimes are clueless. They don't know how to solve a problem. They aren't learning. Those on-line systems encourage rote memorization and discourage deep understanding.

 

[Andy Resnick]

Recently most of the physics textbook companies have been pushing on-line homework systems to accompany their books. (e.g. Wiley plus). I've tried assigning these to my class, and while I thought the on-line problems were good, the students absolutely hated them, and furthermore I found zero correlation between homework scores and exam scores. I was curious if other people have had experiences with this which they would like to share.

Some of the other profs at my department use WileyPlus or WebAssign for the intro classes as a way to manage grading the homework. I made a conscious decision not to use online homework problems when I started teaching.

So far, I've been able to have a grad student handle the homework grading. If that help goes away, I'm not sure what I am going to do- I simply don't have the time to grade 600 homework problems per week. In my class, the homework is an essential way for the student to gain mastery of some course content, but I agree that I don't see a clear correlation between homework and test grades.

Why are you using WileyPlus? How many students are you dealing with?

 

[luriol]

I wasn't using Wiley plus to save time on grading. The class size is only 30 (although grading 30 papers is still a bit of a pain) I was using Wiley plus mainly because a lot of the students were submitting copies of each others homework, and with on-line homework they at least all got different numbers for their problems. Also, I had liked the feature that they could get instant feadback on whether their answer was right or wrong, so that they could try again if they got in wrong.

 

[D H]

The ultimate point of our education system is to educate. If students help you as a teacher accomplish that ultimate goal by teaching one another (e.g., working together on homework), what exactly is the problem? The better students will still show themselves as such on the exams.

From what I have seen helping students here at PhysicsForums, these on-line tools do not teach. While they may (see below) know when the student gets the right answer, they do not have any AI that can help students get past their mistakes. Without such aids these tools are worse than useless. They do not promote learning. By promoting numeric plug-and-chug over symbolic problem solving, I see these tools as an impediment to learning.

One final problem with these tools: The canned answer is not always being correct. Sometimes the canned answer is the answer to a question other than what was asked, sometimes the canned answer is just wrong, period. You can recognize when the teacher's edition of some text has a wrong answer. When you don't recognize the textbook's error and erroneously mark a correct answer as incorrect, students will argue that they are right. They do occasionally win such arguments. When those errors occur in the teacher's edition of a book the publishers will eventually publish an errata sheet. When those errors occur in these on-line tools the tool just marks the student as wrong.

 

[Andy Resnick]

I wasn't using Wiley plus to save time on grading. The class size is only 30 (although grading 30 papers is still a bit of a pain) I was using Wiley plus mainly because a lot of the students were submitting copies of each others homework, and with on-line homework they at least all got different numbers for their problems. Also, I had liked the feature that they could get instant feadback on whether their answer was right or wrong, so that they could try again if they got in wrong.

I understand- and the insta-feedback feature *is* nice. Personally, my decision not to use on-line homework problems was based on my dislike of multiple-choice questions, my dislike of the user interface, and my concerns regarding additional costs and IT/system compatibilities for the student.

Something to think about- why are you assigning homework? I ask this seriously, because if you have an educational goal associated with homework, you are then able to tune the homework to better match your goal.

For example: My goal for homework is to ensure the student gains mastery of problem-solving techniques: mathematical techniques, reasoning, simplification, and exposure to a variety of applications. Every homework assignment I give has a few "group problems", that the students *have* to work on together in a small group (I assign small groups at the beginning of the semester, they are simply lab partners). The group problems are harder problems from the chapter, and group work is designed to provide 'peer instruction', similar to class discussions.

If your students are *already* working in groups, why not take advantage of that?

 

[TaylorRatliff]

I'm a student and undergraduate TA at UT Austin, and have some strong but mixed feelings about the use of Quest in lower division courses. It certainly helps the course run smoothly from the instructors view, but I would rather see more written problems emphasizing problem solving and thought process, even if that would mean more grading. In some of the classes every homework and every test is through Quest, which is a shame. I think we need to find a balance.

The iClicker in class exercises however are a great tool to keep students engaged, especially when they have a few minutes to work with students around them. It also gives the instructor instantaneous feedback on how well they explained something, so they know if they need to revisit that subject next time, or move on if the class has mastered it.

 

[Fightfish]

Perhaps one key feature of written work that is not shared by online assignments is the need to present one's answer clearly (including steps, reasoning etc.) in the former.
Through students' presentation of the process that led them to the answer, misconceptions or reasoning errors can be picked out and corrected, and students can still be awarded partial credit for correct analysis or reasoning even if their final answer turns out wrong.
My teacher held the belief that it is this presentation of the steps leading to the final answer that illustrates whether the student has truly understood the concept. It is also a key skill to be able to present one's work clearly to others, something that doing written assignments would help cultivate.

 

[Andy Resnick]

The iClicker in class exercises however are a great tool to keep students engaged, especially when they have a few minutes to work with students around them. It also gives the instructor instantaneous feedback on how well they explained something, so they know if they need to revisit that subject next time, or move on if the class has mastered it.

I agree- the clickers can be a very effective tool in the classroom. One does not actually need the clicker- I simply poll the students and have them raise their hands.

In any case, the goal is the same- keep the students engaged, and provide feedback to the instructor.

 

[Andy Resnick]

Perhaps one key feature of written work that is not shared by online assignments is the need to present one's answer clearly (including steps, reasoning etc.) in the former.
Through students' presentation of the process that led them to the answer, misconceptions or reasoning errors can be picked out and corrected, and students can still be awarded partial credit for correct analysis or reasoning even if their final answer turns out wrong.
My teacher held the belief that it is this presentation of the steps leading to the final answer that illustrates whether the student has truly understood the concept. It is also a key skill to be able to present one's work clearly to others, something that doing written assignments would help cultivate.

This is my opinion as well- that's why I don't use multiple-choice questions.

 

[Ryker]

The iClicker in class exercises however are a great tool to keep students engaged, especially when they have a few minutes to work with students around them. It also gives the instructor instantaneous feedback on how well they explained something, so they know if they need to revisit that subject next time, or move on if the class has mastered it.

I agree fully, and this is exactly how it's used in my Physics class. If the great majority gets the question right, we just move on, otherwise we first have a quick debate with the person sitting besides us, then vote again, and then the professor explains things in a more thorough manner. It really is a great way to engage students, although I've noticed some people didn't actually buy the iClicker to save on costs. Still, I've found that, since our class is at 9am in the morning, these questions kind of wake me up, and it's much more interesting being in the lecture where you somewhat actively participate rather than just listen to the professor explaining things.

I agree- the clickers can be a very effective tool in the classroom. One does not actually need the clicker- I simply poll the students and have them raise their hands.

In any case, the goal is the same- keep the students engaged, and provide feedback to the instructor.

As for hand raising, I don't think it's as efficient as the iClicker. I think less people are then willing to participate (aka raise their hands), be it due to being shy or unsure of their answers. And it's a lot easier to just get a person to anonimously press a button and see the feedback, without having to go through the process of exposing your choice to everyone. Granted, this shouldn't (but I think it does) affect anyone, as people should not have to worry about expressing their honest opinions, but still. Also, when you're unsure of your answer, sometimes you see other people raising hands for a particular answer, and you automatically do the same thing, even though you don't really get it. With an anonimous voting device I think you're less likely to do that, especially if in the first run you don't discuss the problem with other students.

 

[Andy Resnick]

As for hand raising, I don't think it's as efficient as the iClicker. I think less people are then willing to participate (aka raise their hands), be it due to being shy or unsure of their answers. And it's a lot easier to just get a person to anonimously press a button and see the feedback, without having to go through the process of exposing your choice to everyone. Granted, this shouldn't (but I think it does) affect anyone, as people should not have to worry about expressing their honest opinions, but still. Also, when you're unsure of your answer, sometimes you see other people raising hands for a particular answer, and you automatically do the same thing, even though you don't really get it. With an anonimous voting device I think you're less likely to do that, especially if in the first run you don't discuss the problem with other students.

It's all about trade-offs, right? Do I add another 'thing' that the students either have to pay for or I have to keep track of, or are there alternatives that accomplish the same goal?

To be sure, for the first week the students are timid. But that wears off as soon as they see I don't penalize incorrect answers.

 

[Ryker]

It's all about trade-offs, right? Do I add another 'thing' that the students either have to pay for or I have to keep track of, or are there alternatives that accomplish the same goal?

To be sure, for the first week the students are timid. But that wears off as soon as they see I don't penalize incorrect answers.

Yeah, I agree, and I guess an iClicker has more of an advantage in bigger, first year classes. But then if you buy one in your first year, well, it doesn't just stop working after that

 

[physics girl phd]

I found that in a summer course I teach (intro-level, calculus-based EM for engineering students, with an enrollment of about 40), that the submission rate was miserable for online homework, but MUCH higher for turned in written assignments, so I reverted back to written assignments. Motivation during a summer course can wane (especially in a harder course) and something about the fact that I would SEE them turning in the assignment motivated them more.

To ease my grading of submitted work, each assignment is worth 5% of their final grade (there are 6 assignments and I drop the lowest), and I generally grade some problems for credit (8-12 problems at 0.25% each, which makes grading "easy") and some for completion only (for the remaining 2-3% of the assignment grade, which generally means about 0.05-0.1% per problem). Also, if they don't follow my submission guidelines (which include stapling the assignment in the right order and limiting the number of problems per page to 2) they automatically lose 0.5-1%. (Although my syllabus says I'll deduct a full percent, I find if I give them the smaller deduction the first assignment, it won't happen again and they view me as a bit nice.)

For an easier summer course (which is an engineering course but is more algebra-based), I use an online system, and it works "fine" though they almost certainly treat the problems as more "plug-and-chug". That course is not lecture-based, but is taught in a problems-based/activity-based way, which generally gets them thinking beyond plug-and-chug with equations.

 

[physics girl phd]

With regards to iClickers and such, most students I know view them as a glorified way of incorporating an attendance grade. They generally feel they don't think they learn much if the professor only uses clicker-response questions, without much lecturing, so it's probably all about effective use (which although there are some established baseline via research, can vary by professor, institution, etc.).

I personally don't use them because (except for the above-mentioned EM course, which is time-limited as a summer course) I teach with an activity-based style (which I'll admit does, via the activities, incorporate an attendance grade of sorts, as well as peer instruction). In my "mini-lectures" (the day after the activity and preceding a new activity) I generally refer a few times directly to the activity just completed... otherwise the student can feel like the activity doesn't fit in.

 

[Andy Resnick]

I found that in a summer course I teach (intro-level, calculus-based EM for engineering students, with an enrollment of about 40), that the submission rate was miserable for online homework, but MUCH higher for turned in written assignments, so I reverted back to written assignments.

That's a good observation- My students have online 'reading quizzes', and the submission rate is around 50% (vs. 80% for the P2P homework submission).

 

[clope023]

Some of the other profs at my department use WileyPlus or WebAssign for the intro classes as a way to manage grading the homework. I made a conscious decision not to use online homework problems when I started teaching.

So far, I've been able to have a grad student handle the homework grading. If that help goes away, I'm not sure what I am going to do- I simply don't have the time to grade 600 homework problems per week. In my class, the homework is an essential way for the student to gain mastery of some course content, but I agree that I don't see a clear correlation between homework and test grades.

Why are you using WileyPlus? How many students are you dealing with?

I am not an educator, still a student myself, but why not simply assign homework and not grade it? Only grade the exams? I find I learn more when I don't have to turn in homework; when doing assigned homework I find I'm only trying to get the right answer to get full credit. When it isn't assigned to turn in I find I'll do the problems to see the concept behind them and I learn more that way, my scores go up on exams as a result.

 

[vela]

A few years ago, I tutored a student in physics where the class had online homework. One reason he hated it was simply because the site often didn't work. He'd enter the right answer, and it would tell him he was wrong. Or the answer he had was technically correct, but the website didn't like it because it wanted the answer in a different form. Plus it was just tedious having to enter the answers.

I don't care for online homework simply because I think you lose too much flexibility. If students just has to type in a final answer, you don't see how they reached it, and you can't assign certain types of problems that might be more pedagogically useful. Also, with graded written homework, you get better feedback on where students are having trouble and you can provide better feedback to students, pinpointing where they are making mistakes.

Also, I think there's value in having students write stuff down on paper. It encourages them (I hope) to think more clearly when they have to express an idea on paper. (I told my last class that if I had to work too hard at trying to figure out what they did, they weren't going to get credit even if they did solve the problem correctly.)

 

[Andy Resnick]

I am not an educator, still a student myself, but why not simply assign homework and not grade it?

I could do that, sure. But then what is the point of assigning homework in the first place? Students are *always* free to do more than what is assigned. Indeed, many graduate classes are only graded based on a final exam, or perhaps a few exams. But I am not teaching a graduate class.

And many students like/want/need/enjoy getting feedback regarding their level of subject mastery, under the assumption that it is related to their final grade.

 

[Andy Resnick]

Or the answer he had was technically correct, but the website didn't like it because it wanted the answer in a different form.

That's an excellent point- for example, if a student has the correct answer but expressed in non-MKS units, the student should get full credit.

 

[Ryker]

That's an excellent point- for example, if a student has the correct answer but expressed in non-MKS units, the student should get full credit.

You can easily take stuff like that into account, though, either by expressly stating which units you want the answer in, or by making the solution given in another set of units correct, as well. I know that's how our online assignments in Physics work, and so far that hasn't been a problem. Each question namely explicitly states, say, "give answer in grams or kilograms", or "give answer without units", as the units are already specified.

I do agree online quizzes don't allow for much room, though, and it is all or nothing.