We had an education talk yesterday afternoon, because today's colloquium speaker, Ann Martin from Cornell, has strong interests in that and wanted to talk to people about it. A lot of the discussion had to do with teaching students to write, and getting them to accept feedback. Martin spoke very positively of a writing-intensive introductory course she did about cosmology, and said she saw significant improvement in both students' understanding of key concepts and the quality of their written work over the semester.
Those of us who teach a lot of introductory physics classes with labs all said "How did you do that?" One of our most frustrating problems is that many of our intro students don't seem to pay an attention to the comments we write on their reports-- when we do two or more lab reports over the course of a term, the second and third are just as bad as the first, in exactly the same ways. Even when we do see improvement, in many cases it doesn't stick-- I had a student a few years ago whose second lab report was so bad that I sent it back to him ungraded, saying "If I grade this, I'll have to give you an 'F' in the course, because it is utterly unacceptable." that got him in to my office, and he produced an acceptable report. The next term, he was in my section of the second intro class, and his first lab report there was every bit as bad as the one I had sent back to him the previous term, as if I might've changed my standards from one class to the next.
In kicking around ideas about what to do, somebody suggested sending students the comments on the report without the grade, which would force them to read the comments to see what they were getting. I said that I already do this, and what it mostly accomplishes is getting me sent a bunch of emails saying "I got my lab report back, but where's the grade?" In return somebody suggested, "Just say 'Well, what do you think you're getting, after reading the comments?'"
And, you know, that's just crazy enough that I might try it.
Not in the intro classes, but in our upper-level writing-intensive course next term, I may give that a shot-- send them back their marked-up lab, along with a copy of the grading rubric I'm using now, and tell them that they can have their grade after they return the rubric to me filled out with what they think they're getting based on my comments. It'd at least force them to read and think about the comments a little bit, rather than just setting the marked-up lab aside and waiting for me to post the final grade, which I'm pretty sure is what happens now. And it would let me know which students don't care if they ever see their lab grades, which is probably correlated with not caring about the grades at all, and will let me feel less bad about giving them awful scores.
This probably has its own failure modes, but seems to have some potential, at least for classes with a clear emphasis on writing (which I think is the real difference between Martin's class and our intro courses-- most of the shoddy writing we see results from the fact that the labs just aren't that big a piece of the grade, so students rationally decide not to put that much effort into them). It can hardly work less well than what I'm doing now, so it's probably worth a shot.
- Log in to post comments
I think that falls in the category of things that might (probably) work, at the cost of completely enraging half your students.
I grin evilly at this idea.
oc...is there any room for them to stretch it?) I think there's room here for you to show a level of trust that would reward their being honest with themselves and may even let them appreciate re process.
Could you post a good report and a bad report so that we can get an idea of what you are expecting? (I am a high school physics teacher)
This method worked very well for me as a student at Reed College. The grades were regularly tallied and recorded, but not displayed. The focus became the work itself, and the comments.
This method worked very well for me as a student at Reed College. The grades were regularly tallied and recorded, but not displayed. The focus became the work itself, and the comments.
Those of us who teach a lot of introductory physics classes with labs all said "How did you do that?"
I'm curious how Dr. Martin replied to this question. I suppose you omitted that because it's hard to summarize, but still, is there anything you can say about what she said worked for her?
You may already be doing this, but I found a dramatic improvement in the quality of lab reports when I handed out an example lab at the beginning of the quarter (I just took a high scoring lab from the previous quarter and cut out all the calculations, etc.). A lot of the labs I get now follow that template, but at least I don't have as many completely awful ones.
You may already be doing this, but I found a dramatic improvement in the quality of lab reports when I handed out an example lab at the beginning of the quarter (I just took a high scoring lab from the previous quarter and cut out all the calculations, etc.). A lot of the labs I get now follow that template, but at least I don't have as many completely awful ones.
I've had mixed results. There hasn't been a particularly good correlation between whether I gave them a sample lab and the quality of the results. Similarly, the lengthy written (and recently re-written) guide to lab writing doesn't seem to make much impact.
Just this term, I wrote a new sample lab report, and annotated it heavily to point out the important features (numerical values always paired with uncertainties, figures with labels and captions, etc.), so we'll see if that makes a difference.
I suspect, based on conversations last year, that we're just running up against the fact that the labs don't count that heavily, and students are making a rational decision to not expend much effort on them. And there's not going to be much we can do about that, ever.
I am a huge advocate of peer review for lab reports. You would be amazed how much better the reports become if the students know their friends will be reading it. I find that students are pretty hard on each other, sometimes even harder than I would be. You have to be careful with this, however. You need to have a rubric that everyone follows, and you need to review the peer reviews to make sure that they are reasonable. I think there is simply no better way to teach the students how to write: show them lots and lots of examples of things that do and don't work!
Long experience says they ignore comments that do not carry a penalty. They will even ignore penalties that don't result in an unacceptable grade. You gave a perfect example of the latter in your original comments. Maybe the solution is to "fail" any report that is below a B, but allow them to rewrite it (based on the comments) and turn it in for an opportunity to raise their grade to a C or C+. An automatic fail would be generated by violating any one of three rules on the "unacceptable" list I suggest below.
If doing it right the first time results in less work, they might be more likely to do it right the first time.
I will also add a vote for peer review, particularly if you have a long cycle with only three reports a quarter. Not only are they writing for other students, they get to see the good reports that other students write.
Concerning (emphasis added):
LENGTHY written (and recently re-written) guide to lab writing
Make it shorter. Easier said than done (it kills me to try to say less), but necessary. Are there three or four things -- you listed two good ones @9 before you got to etc -- that would make you happy? Give them just those as mandatory for an acceptable report. [Hmmm, maybe I should now proceed to take my own advice.]
We had a somewhat eccentric teacher in introductory numerical analysis. Half the grade on the course was for an essay on a topic of your choice (I wrote about root finding with Newton-Raphsons method), four pages minimum. Half the grade in turn was for the content, and half for the writing.
Not everybody liked it, but I have to say that essay is what made me realize I needed to improve (well, acquire, really) my technical writing skills. It lead to a couple of small writing courses at the language department, and, eventually, to blogs and other incidental writing. All of which I'm really grateful for now.
Could you do a group lab report? Couldn't it be more productive for all?
Rather than spend the time you would take to read and grade each report spend the time working with a 'lab" of students as they as a group write the report. Or, couldn't an advanced undergraduate do this?
If the lab reports count for a small portion of the final grade, why grade them? Couldn't they just be evaluated on a pass/no credit basis?
If the students are evaluated more formally on the material they learned from the lab, e.g. through some type of exam, then that's the point where they are actually graded.
Mike --My course was actually offered through our Writing in the Disciplines program, so it was explicitly a (small, seminar-style) writing course and not an astro/physics course. I found a balance between teaching the cosmology content and working with my students to improve their writing, but I was free to focus almost exclusively on the latter. I think that is the biggest difference between the constraints in an intro course with lab and the constraints I was dealing with. Some of the techniques I used were similar to what other commenters are suggesting. Like Joanna, I used (guided) peer review, with both a written component and a discussion component. Part of my students' participation grade hinged on them taking peer review seriously and offering substantive advice to their peers. My students were very good at helping each other identify weaknesses and problems, and that might be especially true in the case of a lab report where students have done a lot of their writing without understanding what's going on. I think this might work well for labs, given that some homogeneity and iteration towards a standard isn't an undesired side effect. In my own grading, I avoided line-editing, and tried to be very blunt about what a given problem was (for example, "I have no idea what you are trying to say here" instead of the tried and true "Awk."). One last bit of advice: I focused a lot on encouraging my students to think about the audience for whom they were writing, which I varied from assignment to assignment, and how that would/should influence the choices they make. I think it helped them to have a specific goal in mind, rather than some nebulous sense of trying to please the instructor.
Thanks for the shout-out, Chad!