Monday, May 19, 2008

Public service announcement re: cheating

I want to alert faculty colleagues to a website of which they need to be aware if they teach, particularly undergraduates. I won't link to them since I don't want to drive up their revenue, but it's called cramster.com, and while they bill themselves as a "24/7 study community", what they do is provide links to scanned solution manuals for many many textbooks. What this means is, if you teach a course from a reasonably popular book, you need to be aware that students can and often do buy the homework solutions online. As far as physics goes, they have a rather eclectic assortment. Lots of intro books, and a few major upper level ones (Griffiths; Goldstein; Jackson). If you make up a final exam using problems from the textbook, you're opening yourself up to this problem. If your problem sets contribute a lot to the final grade in a course and you use verbatim problems from the book, again you are almost certainly going to see this on some level. The more you know....





9 comments:

Anonymous said...

People have been trading scanned textbooks and solutions manuals on P2P services for years. It's not at all a new problem, unfortunately. The good thing is that exams usually expose the idiot cheaters.

Anonymous said...

One advantage of sites like this is that students can use the solution manuals to study and check their answers. It can be a good thing if the manuals are able to show students how to do the work in order to solve the problem. If the manuals don't show the work but just the answers and students are just memorizing the answers, this -is- completely unfair. I have never heard of this site in my undergrad though.

Silas said...

I would bet that the ability of students to recognize a solved problem from the textbook and successfully solve it on the test is indistinguishable from their ability to solve it without having had the solutions manual. *Telling* them which problems will be on the exam and solving them in class doesn't affect their grades.

For problem sets, though, a solutions book is a huge advantage.

Douglas Natelson said...

I know that solution manuals have been out there forever; this particular site is just a well-organized "one-stop-shopping" place for them.

Unfortunately (or fortunately) the best way to really learn a lot of areas of physics is to do problems. If the first impulse is to google looking for the answer rather than work it out, then that may achieve short-term success (a decent score on a problem set) but it's not at all clear that it teaches any physics.

Wallflower Physicist said...
This comment has been removed by the author.
CarlBrannen said...

This is part of the reason why I used to grade homework with a score of "1" if it is turned in, and "0" if it is not. The students had no reason to cheat on it as they'd get full credit for their homeworkd just by turning in a blank sheet with their name on it (though no one ever actually did that, hmmm I wonder why).

And it was easier on me, in grading homework I would just add comments on the work. And I didn't have to fight with the "pre med" students about whether the fair grade on their homework assignment was 95 or 94.

Angry said...

Carlbrannen: But what do you do if someone just scribbles some junk for some problems? You have to draw the line somewhere.

Sometimes I get a bit paranoid about such solutions manuals, but I generally just pick up a similar but different text book for good test problems.

For the homework, I don't worry because, as cliche as it sounds, "they're just cheating themselves." And I give close to full credit when someone shows that they've struggled through much of the problem so there's not a lot of incentive just to copy the answer from somewhere. Plus, for those homeworks just copied from a key, they sure are easy to grade!

Anonymous said...

This semester I evolved to giving a point for getting the problem right (with each subproblem of a multi-stage problem counted as a separate point) and then gave them the chance to earn part of the missed points back by redoing the incorrect problems. Later on one of the students pointed out that everything was available online, which solved the riddle about why some people were getting answers so clearly written by someone else and nearly word for word the same.

If I hadn't been swamped (not to mention not paid nearly enough as a mere adjunct to put even more time into this class), I'd have tracked it down. But without the time and energy I didn't report anyone for cheating, I just let it hit them on the final, to show how much they were missing, and then gave grades according to the syllabus.

Douglas Natelson said...

Let me add one more ingredient here.... At Rice and other places with honor code systems, it's not uncommon to give take-home exams. There's a lot to be said for take-homes (provided people don't cheat). They can be self-scheduled to an extent, and I've definitely seen people who perform better in terms of test taking psychology in a take-home setting than in a classroom. Clearly if students have decided to cheat on the test, then giving them book problems on a take-home exam is making it easy for them.

I wonder if cheating in these environments is more prevalent now (because of the easy access to information that just didn't exist when I was a student), or if I was particularly naive back in the day.