Elen sila lumenn omentielvo!

Tuesday, November 23, 2010

"You know who you are!"

I found this video absolutely fascinating:

This has apparently been going around the Interweb a bit, so you may have seen it, but for those who haven't, the Readers Digest Condensed Version is this: a professor of business at the University of Central Florida gave a mid-term exam, graded it, plugged the grades into a spreadsheet, looked at the results, and noticed that the grades on this particular mid-term were ticking noticeably higher than for midterm exams from previous semesters. Then someone anonymously provides him the obvious reason why: some students managed to get a copy of the "test bank" for the course and circulated it amongst a bunch of other students, who used it to prep themselves for the test. In other words: they cheated.

Now, I've been outside the realm of academia since I graduated college in 1993, so I had no idea what a "test bank" was. Basically, as I understand it, the companies that publish textbooks also provide, as a service to instructors, things called "test banks" which are nothing more than giant collections of test questions, from which instructors can then pull questions for their exams. My understanding is that in this case, there was a "test bank" of 700 questions from which 50 were drawn, so students using the test bank to prepare themselves had a pretty big job ahead of them. It seems to me that if those numbers are accurate, then maybe their time might have been better served just studying the course material. In fact, it seems to me that maybe a case can be made that when you prep for a test drawn from that many questions, maybe in a way you are studying the course material.


I've always found that there's something of a grey area when it comes to cheating. There are obvious cases of cheating, clearly: a kid who hacks a professor's computer to steal the test behorehand, or another kid who decides to plagiarize a paper for a course. But then there are things like this, where from certain perspectives, it certainly seems like cheating, while from others it just seems like...well, not really, I guess. It's cheating, but it's cheating that seems to me partially aided and abetted by "the system". It's not unlike the use of steroids in baseball in the 1990s: cheating, yes, but the system was set up at the time to make it very easy to do, with very little penalty.

The whole process of the way this course seems to operate seems a bit...lazy? Maybe that's too strong a word; certainly there seems to be an "assembly line" aspect of the teaching enterprise going on here. That prof gives lectures, and the students go off to "labs" to do...more coursework, I guess. There's not a whole lot of context here to go on, but the structure seems to put a lot of the work of running the course on the heads of the lab instructors, from grading the exams and homework to spending long hours creating a new mid-term exam once the old one has been compromised. But when you have 600 students taking a course, I suppose there really isn't a great alternative.

Some other things strike me from the video. First, the professor here continually refers to "forensic analysis of the data", which will presumably reveal all of the cheaters. I am no statistician, so I'd be interested to know how the data can reveal the difference between a student who prepared for the exam by use of the test bank and a student who prepared for the exam by traditional studying of the course material. He implies, a number of times, that the cheaters will soon stand revealed and that at that point, the folks from Academic Affairs -- whom he really seems to be going out of his way to make sound like a bunch of relentless Imperial Stormtroopers -- will decide their fates, particularly with the ominous statement, "I don't want to have to explain to your parents why you're not graduating." He also says something like "The net is closing around you even now!"

But then he also offers the cheaters a deal: all they have to do is admit that they cheated, and take a four-hour ethics course, and...all is forgotten. They'll take the same re-written midterm as anyone else and go on from there. Well, that certainly does not seem to me the type of offer one makes to someone when "the net is closing around them". It doesn't make sense to offer a deal that carries with it minimal effort and zero long-term consequences for the guilty if you're already certain of their guilt and are near to proving it. This whole part of the lecture strikes me as the kind of veiled threats of legal action that collection agencies use; for all their harsh language and legal-sounding threatening, it turns out that collection agencies are really pretty toothless.

And frankly, I as a student was present for more than a few instances where the teacher had to dress down the entire class for lack of specific knowledge as to which student did something naughty, and in my experience, the teacher or principal or whomever always pretty much lost the room when he or she uttered the phrase "You know who you are." (Actually, the prof in the video has an even more laughable line than that: "The days of students figuring out new ways to beat the system are over." Yeah, dude. Sure they are. As of this day, all future forms of cheating have been rendered impossible! Huzzah!!!)

I didn't really witness much outright cheating when I was a student. I myself never cheated, although the thought did occur to me a few times, here and there. I was always deterred more by the thought of dolorous consequences than by any good and noble commitment to my own learning, although I did generally think that I'd rather have a bad grade be my own doing than a good grade be artificially achieved. Generally, though, I just didn't have the confidence that I could pull it off and I certainly didn't want to get caught. I'm like the guy in Dial M For Murder who says that he only believes in the "perfect murder" on paper: "My murder would be like my bridge-playing -- I'd make some stupid mistake and only realize it when I noticed everybody else staring at me." Plus, since I was a Philosophy major, there really wasn't much you could do in terms of cheating beyond plagiarism -- and on plagiarism, I do have a pretty rigid moral code. But when your exams are all-essays all-the-time, well, you either know your shit or you don't.

There was one cheating incident that I remember fairly clearly, though, from my grade school years. In seventh grade, several classmates of mine somehow managed to get access to our math teacher's grade book, and they took that opportunity to take a pencil and adjust their own grades upward. I don't recall how they managed to score a few minutes of unsupervised time in her room, but they did. And then, afterwards, they bragged about it, so everybody knew. Someone blew them in, and that was that. One fine day we're gathered for class -- it's toward the end of the year -- and in storms the teacher, grade book in hand; she announces that she's just discovered that the grades in her grade book don't match the grades she routinely turns in to the Guidance office (a story she made up to cover the fact that someone had reported them, which everybody figured out anyway).

This particular teacher was...well, let's just say she was not one of my favorite teachers. At all. Never liked her in the slightest, although she was inexplicably beloved by most of the rest of the school. (It's the same teacher in Item #7 in this old post of mine.) After making her announcement, she proceeded to take malicious glee in tormenting the kids who had cheated, saying things like "Hey, you're looking a little pale there!" and stuff like that. I always thought that was crappy on her part -- you've got the kids by the hair, just take 'em down to the principal and be done with it. But that teacher sure enjoyed dealing out public humiliation to her chosen few. I don't recall what the consequences were for the cheaters in that instance, but it certainly made for an interesting few days in a small school.

Anyway, moral of the story: don't get the test bank.


Kal said...

When I was a teacher I would hand out a sheet of questions before every test, midterm and final. They contained all the possible questions that I could ask on the test - in effect it was a test bank. I did it for two reasons. One was to show the students that therew ould be no trick questions and the second was to reduce their workload when it came to studying for the final. If a student had used each of my test questions to study for the unit tests then it would be all they would need to do well on the final. I was also teaching them how to study which was a much larger and more important skill than whatever the subject was that I was teaching. I gave them the questions and if they worked at it, they would have the answers. I think that is where this prof missed the point. You defeat cheating by taking away the insentive to cheat. No one could do well on my tests and complain later because I had already given them all the questions. My system exposed the cheaters because they were the ones with the lowest scores, not the highest.

Roger Owen Green said...

I know nothing about test banks.

But you do make an interesdting observation about steroids. I'm a lot less critical about Bonds, McGwire, et al. because, until 2004, it was allowable behavior. You can talk about Bonds' boorishness, or McGwire's bad answer to Congress, but if I were voting, I'd put them both in the Hall of Fame (tho not on the first ballot).

Call me Paul said...

Kal, you were a math teacher, right?

Kelly, I agree with your very first point. If these kids had to familiarize themselves with as many as three hundred questions in order to do well on a possible 50 to be on the test, it seems to me they probably know the course material better now than when they started - which is the goal, is it not? Oddly enough, they probably studied harder than they planned to - because they got the "test bank." Those test results are probably valid.

Derek J. Punaro said...

I find it hard to consider this cheating. First, most professional certification exams that I've seen/taken work precisely this way. The company provides sample exams with questions taken right from the test. "Study guides" are sold containing many of the exact questions asked on the exams. The point is they show the format and style of the questions that are going to be asked, so you can adequately prepare for the exam. And if are studying based off sample questions, you're certainly learning the material in the process. The answers of multiple choice questions are generally randomized, so you can't just remember the answer is "B". You have to know what the answer actually is. There were things I learned from studying sample questions for a certification exam that I would have never learned otherwise as they covered some obscure technical function that I would have never used in real life.

If the professor doesn't want the students to study the questions, easily solved by writing his own.