Wednesday, November 18, 2009

Grade Eachother?

I started doing daily homework assignments for my calculus class when we got to series convergence tests. The idea is that they will read the section in the book and work some problems to turn in, before we talk about it in class. It isn't going well.

Mostly students are "getting the right answer," but the write-ups are fairly bad (mostly making me feel like understanding is pretty low). I spent about 20-30 minutes in class Monday telling them about the things they were doing that are driving me crazy.

We've got three more of these daily assignments due over the next week. I'm thinking about doing the following: During class, have the students gather themselves up in pairs and have each student write their name on their partner's paper. The students will spend time looking for mistakes on their partner's paper. And I'll tell them that whatever points I take off of a paper will also be taken off of the partner's paper.

Surely somebody out there has done something like this before. How does it go? Does it help? What do I need to watch out for? I'm a little worried that "correcting" won't happen as much as "copy down what is hopefully a better answer, without understanding it".

6 comments:

Unknown said...

Doesn't sound like something I'd be a big fan of. I wouldn't expect them to understand everything perfectly the first time through (otherwise, what do they need you for if they have a book?), but rather all the mistakes they make on the homework are opportunities to see what the class needs to spend the most time on.

sumidiot said...

Thanks, @Meekohi, for the feedback. Decided not to go for it today, probably won't the other days as well.

Of course, part of my goal is to convince them that they mostly don't need me, as you hinted at.

Chris said...

When you say the writing is bad... is it bad in the sense that you're talking about baseball and they're writing about underwater basket weaving OR is it that if you dig really deep and kinda squint you can see that they might have an idea what they're talking about but they just don't know how to write math?

Because I've known a lot of undergraduates in calculus who mostly know what they're doing but really don't know how to express that knowledge. That might not be the case here, but it wouldn't surprise me if it is for at least some of the people.

To answer your question, I think your students are going to feel it's an unfair practice because even though they're helping to correct the paper, it's still not their own work, ultimately.

sumidiot said...

@Chris, you are probably right that students wouldn't find it fair. One response to that is that mainly I don't take off more than 1/2 point for the sorts of writing issues I'm currently worked up about. Another response is... what if I only _told_ them I was going to take off points from both papers? I know this would only work once, but at least for that day they would actually focus on reading another paper...

To address your first question... they are on track, and I don't even generally have to squint that hard. Which is why I think we should be spending time on the writing aspect. They are showing that they basically have some understanding about the mechanics, so it is time to move up the understanding ladder. ... was my thought, I guess.

MTK said...

Peer feedback is really valuable, but it's got to be structured right for it to work. I'm not sure that penalizing the partner for a mistake of the other will have the results you want. I also think that you have to give them lots of practice at giving peer feedback and explain why you're doing it early on so that they buy in. My students have had sessions 10 times over the course of the semester where they gave each other feedback on problems. These are problems where their solutions are required to have a complete sentence explaining every step of their work. I've seen them progress from near gibberish that was getting high marks during the peer feedback phase to really, really, really high-quality work. Honestly, I would be happy if calculus students solved and explained their solutions to word problems half as well as my precalculus students are doing right now.

sumidiot said...

Thanks, @Mitch. I think "providing more structure" has been a lot of what is lacking from my course this semester. I've sort of adopted a "let them learn to swim" attitude. I like the idea of doing peer feedback regularly through the semester, as you suggest, so perhaps I'll try that in the future.