Tuesday, November 23, 2010

Oh rly?

http://idle.slashdot.org/story/10/11/18/152256/200-Students-Admit-Cheating-After-Professors-Online-Rant?from=rss

Over 200 University of Central Florida students admitted to cheating on a midterm exam after their professor figured out at least a third of his class had cheated. In a lecture posted on YouTube, Professor Richard Quinn told the students that he had done a statistical analysis of the grades and was using other methods to identify the cheats, but instead of turning the list over to the university authorities he offered the following deal: "I don't want to have to explain to your parents why you didn't graduate, so I went to the Dean and I made a deal. The deal is you can either wait it out and hope that we don't identify you, or you can identify yourself to your lab instructor and you can complete the rest of the course and the grade you get in the course is the grade you earned in the course."

In principle, I like this.  Cheating is bad.  Cheats prosper, but at the expense of honest people, which is fairly shit.  It’s best if nobody cheats, but degree courses are exactly the sorts of thing that can be infected by cheating and once it’s happened, the pressure for everyone else to cheat increases.  I suspect we can reduce opportunities for cheating in university exams, but I’m not sure these approaches are compatible with aims of sending as many people as possible to university.  For instance, my lecturers knew I wasn’t cheating because they set extra, unmarked assignments.  I did them all and I tried to find unexpected answers to them and discuss why I’d done it that way.  Of course, I’m an unrelenting geek and – I’ve just realised – a bit of a swot.  And of course it isn’t clear how this kind of thing can scale.  I’m not describing a solution so much as suggesting that something like a solution might be doable if we were to have realistic expectations of the education system.

But back to the link.  The professor makes claims that aren’t substantial.  He begins, bewilderingly, by comparing two different types of distribution from two different events.  The things he says are clear from his graphs are not clear to me.  He claims he worked out from this analysis that people must be cheating and that later an anonymous student revealed that the answers to the exam were in the wild and that the students might have used them.  This, he says, confirmed his suspicions.

What do you do if you find out a substantial contingent of your class has been cheating?  Well, the only thing you can really do is set another exam.  Shame for those who didn’t cheat, but what choice do you have?

And that’s what this guy did too.  However, he is a passive, aggressive, and passive aggressive wanker, so he couldn’t leave it at that.  He had to pretend he knew all along that cheating was being done and that it was just confirmed by the anonymous confession.

He had to pretend that he had statistical reasons to suspect it.  Those reasons make no sense.  He compared two different distributions from two different times.  Nothing about this can be used to demonstrate that cheating happened.  It’s just a demonstration of authority – and an incompetent one at that.  Personally, I’d fail all students who failed to see through such transparent crap, whether they cheated or not.

But he doesn’t stop there.  He goes on to pretend that his advanced statistical analysis of ‘the data’ plus unspecified analysis of (presumably internet) ‘traffic’ has narrowed down the cheaters to 1/3 of the class.  He can’t (yet) guarantee that he has all the perps bang to rights and he might have fingered some non-perps.  He has a mystifying 95% confidence that his list includes all the cheaters, but some non-cheaters might be on the list too.  It’s an astonishing claim and you’d imagine he’d back it up, wouldn’t you?  Well, it requires ‘further forensic analysis’ apparently, which is expected by the end of the week.  Right.

But he’s declaring an amnesty: if the cheaters come clean, they can retake the exam (if they also complete a 4 hour ethics course, hilariously).  He doesn’t actually say what will happen if they don’t, which rather takes the teeth out of the pretend threat.  Some sinister-sounding department will ‘get involved’.  That’s even less scary than his original toothless threat.

Don’t get me wrong.  I don’t want people to cheat.  But I don’t want self-obsessed passive, aggressive, and passive-aggressive wankers to vent their failings on students.

His final flourish is just amazing.  He’s close to tears.  He claims to know who the cheaters are.  He clearly doesn’t.  He talks about sending ‘information’ to people who couldn’t possibly have anything to do with the situation.  He breaks down, wondering what his last 20 years teaching were ‘for’.  He washes his hands of the cheaters, even though he doesn’t know who they are.  While pretending he does.

Then he says that pregnant women who have to give birth at the time of the make-up exam will have to give birth in the exam.

Oh you worthless prick.  You don’t have any weight to throw around, so you’ll throw around hatred and stupidity instead.  Someone cheated so you’ll pretend you know who did and who didn’t and you’ll punish everyone anyway, regardless.  You don’t actually have any power to really punish people so you’ll pretend you’re punishing them BY ROBBING THEM OF AN ACTUAL LECTURE, YOU FUCKING DICK. 

Supposedly some people confessed to cheating.  Me? I’d have failed those idiots.

No comments:

Post a Comment