Panos

The student admitted that the whole email was total BS. I took excerpts of the email and posted them here, not the full message.

Come on, could you really believe that this email contained a single element of truth?

Yes, indeed, the post was supposed to be "a story with the twist." Had I known that I was going to have hundreds of thousands of people reading the post, I would have followed the standard journalistic practice of writing a summary at the very first section. (See http://t.co/2kJEkJW for a copy of the post. Note: I asked the post to be taken down until I repost the original article but the journalist is really playing childish games.)

I kind of felt this a few hours after the post went out, so I added two clarification points early on in the article: 1. I am not giving up the fight, I will just fight differently, please see the conclusions (link), 2. This was not about NYU and people cheating in business schools; people cheat everywhere: the story gives an explanation why they remain undetected.

Oh well, people could not even read these two points.

But I want to write stories in my blog, not papers with an abstract, executive summary and table of contents.

Thank you for a constructive disagreement. Perhaps car theft and bicycle theft would have been a better analogy, when talking about the offence (but I could not find a good analogy for the underlying problem).

"In practice, there were apparently edge cases (students with ambiguous evidence..."

These where exactly the cases that I focused on. I honestly thought I was playing in a "Numb3rs" episode. Constructing models for detecting the probability that an individual student would structure an Excel spreadsheet in a specific way, measuring likelihood of sentences using Google N-grams, and other things that are not the obvious things that the honor council would immediately understand. At the end, the students were admitting guilt but at a great personal cost for me.

When nobody else is fighting like that what is the incentive for me to continue? Not all students go through my classes, so that all have equal chance of being caught. In a (very) perverse way, it is a matter of fairness to the students for me to be as vigilant as everyone else.

Well, there was a point in that analogy. In the past, I was catching cheaters who put no effort in the assignment.

This time, I had a zero-tolerance policy. No matter if you plagiarized the whole assignment or a paragraph, no matter if you copied the spreadsheet, or if you just had the solution as a backup to ensure correctness, it was the same penalty: A -30 for an assignment that had a +30 maximum points and report to the Dean's office and the honor council.

I did this lecture in class. I had to do a quick Google search that day and I found two cases of plagiarism in the news.

* ArsTechnica, Oct 8, 2010, Copied pleadings show there's no honor among antipiracy lawyers http://arstechnica.com/tech-policy/news/2010/10/copied-plead...

* Associated Press, Sep 30, 2010, Arizona US Senate candidate denies plagiarismhttp://www.google.com/hostednews/ap/article/ALeqM5jCoMCrHFbR...

Well, some students got the (subtle) point that cheating is often undetected but when detected is lethal for your career. But I believe these students would not cheat in any case.

If you attribute the 20% rate to NYU, and not to my increased vigilance, then you may want to read again the latest blog post.

[and the reply that I also posted on my blog :-)]

What you describe was my policy in all prior years. I just stupidly felt last year that I need to hunt further.

Nope. Big mistake.

Just FYI, I could always locate 1-2 cases of cheating in every semester. But I treated treated as isolated cases.

:-)

Admittedly, I got the impression that most people do not bother to read anything longer than 140 characters.

Mea culpa. I just wanted to close the loop and identify a point that people would relate to. (I was not sure that people would relate to the fact that the student evaluations dropped. Apparently most people do not even know that student evaluations are used to determine salary increases.)

In retrospect? One of my biggest mistakes to even mention it.

No, I could not really think of that. I do not filter my blog posts through any PR department, and I have written plenty of stories that received wide attention. And going through a PR department typically means that the resulting document lacks any punch and originality. (Been there, done that, for other types of communication.)

If I could think that people could leverage the post to post their negative preconceptions about business schools, NYU, Turnitin, academic, US, or whatever else they hate, I would have not posted the blog post to start with.

In most of my blog posts I try to propose and describe solutions, even when I "whine". (See, for example, http://behind-the-enemy-lines.blogspot.com/2010/10/plea-to-a... and http://behind-the-enemy-lines.blogspot.com/2010/12/mechanica...). The blog post about cheating was not an exception.

The reason that it was in my draft folder was because I did not want to put it out just to talk about cheating. The post felt incomplete. Yes, I could complain. But after that, what?

It took me a while to realize that people did not cheat in some of the assignments, but tried in others, was because some assignments were designed to withstand cheating. And that was a point worth talking about.

Yes, I had to remove the article from my own blog.

Absolutely correct. The compensation part was just one of the many data points, but a point that many people can relate to. A couple of thousand dollars per year would not make a difference in my life. It is mainly a signal of appreciation for the work.

The essay was about doing a Porter's Five Forces analysis of the wireless industry, as the new 4G technologies enter the market. The WiMax question was just a "warm up" easy question, so that students that do not know what WiMax is can start learning about the concept, before doing the 5-forces analysis.

This was the reason that, initially, I did not bother penalizing the copy & paste behavior for these questions. But, unfortunately, this was just a signal of a deeper problem with the student. After going through unharmed in the first assignment, the student cheated again in the second, Excel-based assignment.

Well, for non-office hours I can always say to the student that they cannot come in, as I have other things to do. There is a legitimate excuse not to even start the conversation.

However, during office hours, any student can come in and discuss class-related topics. I cannot prevent students from entering my office, just because I believe that they will want to discuss their cheating penalty.

From the syllabus and the assignment description: "Students may not work together on graded assignment unless the instructor gives express permission. Collaborating on graded assignments with students from other sections, is a violation of the Stern Honor Code. Similarly, getting help to complete the graded assignments from students that attended the class in the past is also a violation of the Stern Honor Code."

I explicitly prohibited students from working together on individual assignments. There were plenty of group projects for doing so, but for the individual assignments I wanted the students to work alone.

The student clearly violated the Stern Honor Code, and he is openly admitting that.

"A clever teacher would figure out a way for students to police each other. I've seen it work to great effect. No one grades harder than a peer."

I could not agree with you more. This assignment, in which students get to teach "emerging technologies" to each other, and grade each other, is the highlight of my class. The whole point that I wanted to make with this post is that cheating is something that needs to be structurally avoided, not something to be detected and penalized.

Btw, touché on the libertarian point.

Wait, as I just noticed this: You collaborated with another student? In an individual assignment? In which you had very clear directions not to collaborate with other students?

I am really sorry that I did not catch you. You would have received the penalty that all other students who "collaborated" received, which I guess would have brought your grade to a C-.

When I detected 15+ cheating cases, I just asked permission from the Dean's office to notify students over email, and then direct them to the Dean's office for appeals.

So I did that, notified the cheating students they got a -30 (negative the points of the homework) and directed them to the Dean's office for appeals.

Well, it was very difficult to enforce the "talk to the Dean's office if you have a problem with my judgment." When the student comes in my office during office hours, I cannot say "I do not want to talk to you, get out of my office". Yes, I could do it, in theory. Hard to do in practice. Would I physically push the student out of my office?

Ah correct. I should have mentioned that.

I was detecting some cheating in Excel assignments, but I was only pursuing the very obvious cases. At most a couple of cheating cases per semester.

For essays, it was pretty much impossible without the use of Turnitin, which I used for the first time in the Fall 2010 semester.

Again: read the article until the end. I do not blame the students.

Students respond to incentives. If I give them opportunities to cheat and remain undetected, they will cheat. I do the same as a human. If I were getting a 50% increase in my salary for detecting cheating so aggressively, next year I would have tried to detect cheating even when there is none.

What I truly blame in the article is the structure of the homeworks and the overall student evaluation approach. Which is something that I use. And I criticize my approach of trying to detect cheating post factum, instead of building projects and homeworks in which cheating is just meaningless.

Just tell me: Could anyone cheat and copy the presentation on augmented reality by your fellow classmates? It was simply stunning and they fully deserved the 10 points extra that you and your fellow classmates assigned to them. That is an evaluation strategy, where nobody could cheat because nobody could cheat your own instincts for detecting and recognizing quality.

I see your point. But consider the following: these numbers are for the students that admitted cheating. There was a significant number of other assignments for which I suspected cheating but I did not have enough evidence. So, I erred on the side of having false negatives in the accusations, rather than false positives.

Now, if you are teaching a class where you have some suspicion that half of the students have cheated, one way or another, the attitude changes. I wish I could be a cold, rational, emotionless professional, and ignore this feeling altogether. Unfortunately, I am a human. And I felt that I am teaching a class full of kids that come just for the grade, and not for learning.

And I am kind of disappointed that nyustern still demonstrates this grade-seeking attitude: He learned in the class, he appreciated my attitude in the lectures but his main complaint was that I was not approachable outside the class, and outside office hours? And that, out of all things, did not return his smiles? Hint: at 5pm, after 11 hours of work, 3 hours of lecturing, I am tired, mentally exhausted and I want to go and get some rest.

By the administration, which separates the credits for the courses into "recitations" and "lectures". The TA's get paid for the recitations/activities-based hours, the professors for the lectures.

I have the proposed solution at the end of the article. Please read that part as it is at the very heart of the point that I want to make.

Let me start by indicating some logical flaws in the timing of the events: The office hours for the class were from 12.30 to 1.30, Monday and Wednesday. The class was starting at 2pm. According to your account, I showed up at 1pm, being 30 mins "late". Then the first student came in at 1pm and came out at 1.20pm? Then it took me another 50 minutes while I was running around? So I came to talk to you at 2.10pm, when I was supposed to be in the class? What time did I show up in class? 2.30, and half an hour late?

I do remember the incident but you are greatly exaggerating the timing. The first student was waiting, I showed up at 12.30 and she was out at 12.50. Then I did run down to the cafeteria to grab something to eat. Sorry, my blood sugar was running low, the line of students was getting longer, and I could not go and teach at 2pm for 3 hours straight without food. (Usually I was getting lunch at 1.30pm but with this line of students, things were not looking good.) At 1pm, I was back, we discussed the case and I continued the discussion with the remaining of the students in line.

You were waiting for two hours mainly because you were so worried that you showed up outside my office at 11am, a full hour and a half before the beginning of the office hours. Expecting me to be always in my office, and at your disposal, is understandable from the point of view of an undergraduate. But you may need to learn that professors have other tasks they need to accomplish during their workday.

Btw, I thought that the you got the joke about the "emergency" (that I had to grab something to eat), by obviously seeing me walking in front of you with a sandwich in my mouth. You can blame it to my bad attempt at humor.

Now that the timing issue is resolved, let me get back to the core of the argument, which I actually addressed in the post. It is the fact that I felt mostly sorry about: The fact that the cheating cases really changed my mood and attitude towards teaching this class, and this also affected the class dynamics. What you felt as unfriendliness was the direct causation of the cheating cases. It is very difficult to feel like hanging out with students when feeling that a very significant fraction of the class is actively trying to cheat. This the part that I said that I hated the most: the very different dynamics in class.

This is the part for which I should apologize: For now offering you the same experience as I offered to the students in prior semesters. Go and ask students that took my class in prior semesters. Or even check my ratings in prior semesters. I did not suddenly become an arrogant bastard.

PS: About the issue of a "true capitalist," yes I am a libertarian at heart. And I do believe that people respond to incentives. And I would strongly encourage you to read the article until the end.

I thought of that as well. With this approach you hit another structural problem: activities such as solving problems etc are considered appropriate only for "recitations" (i.e., done by TA's) and not for "lectures" (the part where the professor teaches).

Kind of stupid, I know. But it is really impossible to violate the mandate that my teaching should not be "wasted" on doing assignments in class etc.

And yes, when I tried it, I loved it. The teaching time becomes so much more interesting and stimulating. But I can only practice this a couple of times over the semester, as otherwise I will be seen as "wasting" my time in class.

It was less than inflation, I am just confirming that :-)

I do regret letting the "copy & paste" students go relatively unharmed. I was naive to believe that they will not do it again. They were even more naive to try to cheat again in my own class.

Btw, all students were reported to the administration and have a mark in their transcripts that they cheated. This is communicated to all future employers and is an automatic block to entrance in law schools.

But again: 22 out of 108 is not a problem to be solved by force. It is a deep structural problem that got me thinking about evaluation strategies that are independent of cheating.

Btw, the girl was not crying to manipulate me. She really had a serious nervous breakdown, bringing out all sorts of issues that she had with her life. Apparently her parents kept reprimanding her for not "performing" at the university. There was a long term stress issue and the realization that she was going to fail the class was just the straw that broke the camel's back. But it was a very uncomfortable spectacle: her classmates outside my office looking at her, and other faculty members stopping by and making gestures, trying to figure out what is going on.

Derrida, most of the professors teach because of students like you. Nobody that I know enjoys teaching disinterested or bored students.

Of course, there are professors that fail to teach well, the students get disinterested, then the professor gets even more disinterested and teaches worse etc etc.

You are correct. I did not want to sound racist. Perhaps "cultural ties" would be a better term. Btw, most of this happens because students with common cultural ties tend to socialize together in student clubs, etc.

(I am the blog post author).

No the suggestion was not to create entirely new assignments every year. This is problematic, not because it takes time, but because is it hard to "debug" the assignment, and make it to be not too hard, not too easy, and not ambiguous. You cannot know this before actually giving the assignment out to students.

The senior professor was just suggesting to change the numbers, or small elements of the assignment. A thing that I was doing already.