- EphBlog - http://ephblog.com -

How Much Cheating?

How much cheating is there at Williams? A student writes:

For those who have never read the honor code committee reports, especially current students, they’re a very worthwhile read. They alert you to the specific kinds of behaviors that actually get you the black-mark of academic dishonesty on your transcript. Some notes about them:

1.) Why are there so many typos in the honor committee reports? Even a cursory reading of these 4-6 page documents would correct for these rather glaring errors. If you’re publishing something that will have your committee’s name on it, and your committee is essential to the academic integrity of the college, you’d think the document would be a little more polished.

2.) There doesn’t seem to be a strong trend in what class years are accused/found guilty of plagiarism. If, as Shevchenko asserts above, academic dishonesty stems from different high school backgrounds, we’d expect for the influence of those differences in secondary education to diminish over the course of students’ time at Williams, leading to an overrepresentation of freshmen in honor committee hearings. There’s many other reasons we’d expect for freshmen to be overrepresented (e.g., students get better at cheating). I haven’t run the data, but there seems to be a pretty even mix of freshmen, sophomores, juniors, and seniors being tried for honor code violations.

3.) Professors have tremendous difficulties catching students who are cheating on take-home exams. During my time at Williams, take-home exams were incredibly common, especially in DIII courses, and it was common knowledge that students would cheat on these assignments. When you’re alone in your room taking these exams, there’s not a lot to stop you from opening your textbook or phone to look for answers on a surprisingly difficult question, and resisting this urge is difficult with up to 30% of your grade is on the line. I believe this is probably the most common source of cheating at Williams, and the most pernicious, since take-home exams are frequently major assignments and professors will be hard-pressed to catch students.
– Only 3 students in the 2016-2017 school year were accused of cheating on take-home exams (I would guess that over one-thousand take-home exams are administered each year and the incidence of cheating is much, much higher than 0.3%).
– These two students were caught due to incredibly flagrant violations of the honor code: one had verbatim copy/pasted material off of Wikipedia (laugh, then expel this student immediately for their sheer stupidity); the other two had identical portions of their assignments, obviously indicating collaboration. All failed the courses, no additional sanctions.
– The previous year also had two violations, one with obviously identical material between two students and the other with a student who turned herself in.
– Conclusion: Professors are not detecting/reporting who is using textbook or online sources during take-home exams. This should be a huge concern to professors and the college.

4.) Similar to #3, only one student in the past two years has been found guilty for cheating with the use of a smartphone in general. Once again, among students, it’s common knowledge that you can have your phone in your pocket and then go to the bathroom to use your phone to look up answers during a self-scheduled or even an in-class exam. One student being found guilty of this behavior is surely the result of a very low detection rate rather than a low prevalence rate among students. As with cheating on take-home exams, this should be a huge concern of the college.

5.) Only incredibly sloppy and obvious instances of cheating are being detected. Take a scan of any of these documents; a large majority of cases involve verbatim similarities between two students’ work or between a students’ work and the internet. Virtually none of the students who are cheating in more careful ways are being caught; it’s all the low-hanging fruit of lazy or stupid students who make the egregious error of copying text verbatim.

So, if you’re planning to cheat at Williams, don’t verbatim copy text from an internet source or a friend. This is essentially the only reliable way you will be put in front of the honor committee; such violations constitute a large majority of honor committee hearings. With a little bit of cunning, you can *easily* use technology to get away with cheating. Until the college finds a better way to catch students who are cheating, possibly by banning take-home exams, it’s almost guaranteed some of your peers will be engaging in this behavior and will get away with it.

How much cheating is there on take-home exams?

Comments Disabled (Open | Close)

Comments Disabled To "How Much Cheating?"

#1 Comment By anonymous On October 11, 2018 @ 10:58 am

I’m still in a state of disbelief over div 3 take home exams. Thinking about my own experience in Organic Chemistry – crazy hard class for me. I can’t imagine the temptation to cheat when your alternatives are between failing the class versus probably NOT getting caught looking up some compounds/formulas in a text book.

I remember the last time Ephblog discussed this topic, a professor weighed in and stated that take home exams in Div 3 classes are fantastic because you can get much deeper into the topic than you can in a 1 hour proctored exam. So, why not give a 3 or 4 hour proctored exam? Seems to me that professors have ample time to supervise such an exam.

Also, cell phones in the bathroom? Easy solution there. When you leave the room for the bathroom, you must leave your cellphone behind.

#2 Comment By ambrosius aurelianus On October 11, 2018 @ 11:48 am

>why not give a 3 or 4 hour proctored exam?

Exams at Williams are centrally scheduled and uniform across the entire college in terms of format: Students get 30 minutes to read and then two hours for the exam itself. If you want to administer an exam that requires more than 2.5 hours of work, take-home it is.

>When you leave the room for the bathroom, you must leave your cellphone behind.

You can’t search the students; they could conceivably claim not to have brought their cell phone and take it to the stalls to look for answers. Other workarounds are also conceivable.

#3 Comment By PTC On October 11, 2018 @ 2:52 pm

Ambros- Why not make it open book and timed pressured to get the results that properly measure the student?

You can do this by allowing a take home with a time limit. You can create an exam, that even with the use of the internet, getting a better result relying on the medium is impossible given the material and time pressure.

Time constraints can be achieved through a wide variety of mediums, such as CANVAS. Also, you can curve the results… make it so that the best grade is not an “A” without the curve.

That makes it so students will not share answers…

#4 Comment By ambrosius aurelianus On October 11, 2018 @ 3:03 pm

PTC: These and other measures are indeed workable. Personally I don’t favor enormously comprehensive multiple day take-homes and prefer shorter exams that randomly sample knowledge. That said, I’m not in Div 3 where the concerns presented in the OP are clustered.