Colleges Stepping Up Anti-Cheating Technology 439
Bruce Schneier's blog highlights a New York Times piece on high-tech methods for detecting student cheating. Schneier notes, "The measures used to prevent cheating during tests remind me of casino security measures." "No gum is allowed during an exam: chewing could disguise a student's speaking into a hands-free cellphone to an accomplice outside. The 228 computers that students use are recessed into desk tops so that anyone trying to photograph the screen — using, say, a pen with a hidden camera, in order to help a friend who will take the test later — is easy to spot. Scratch paper is allowed — but it is stamped with the date and must be turned in later. When a proctor sees something suspicious, he records the student's real-time work at the computer and directs an overhead camera to zoom in, and both sets of images are burned onto a CD for evidence." The Times article quotes from research published a few months back suggesting that the more you copy homework, the lower your grades.
It's better to have students that don't cheat (Score:3, Interesting)
At my university, in scenic New Jersey, we had an Honor Code that we had to sign after every exam; saying that I didn't cheat. I felt proud signing that, and believe that most of the other students felt the same.
If some folks want to cheat, they will find a way: Chewing gum or no chewing gum. With such measures, you will only force the cheaters to be more creative. Try to teach them values so that they will know that it is wrong instead.
Do they really think it's cheating? (Score:5, Interesting)
My brother in law, an economics professor, recently had to grade a paper from the freshman class he was teaching. He found that virtually every paper had the same ideas in the same sequence, and frequently the exact same wording (I.E. cut-and-paste). Even more interesting, and disturbing, he found that by comparing the texts they could be roughly grouped by the race of the student.
His theory is that the current generation is so used to forwarding, re-tweeting, re-blogging, and re-posting that they literally don't see it as cheating.
cultural differences (Score:5, Interesting)
I suspect there are serious cultural differences regarding cheating. For example: at my university, the Indian comp-sci students all knew each other and held regular "study sessions." I was once invited to one. I was amazed to observe that it was simply a highly-organized cheating exercise. These guys had graded homework assignments and exams from all classes, and they passed them around, casually copying solutions verbatim to their homework assignments and recording exam answers. They begged me for all of my exams and homework assignments from current and previous tests so that they could add them to their collection. And they didn't see anything wrong with this.
What I found particularly amusing was how amazed they were at my abilities at coming up with solutions when we had non-trivial group projects. "How did you know that would work?" they would ask. I had to try hard to avoid saying "I don't cheat so I have to actually understand the material to pass the classes."
When I was a T/A (Score:3, Interesting)
I taught a circuits lab class when I was in grad school. I eliminated cheating quite easily. I generated an individual test for each student with the exact same problems but different values for the components. I also randomized the order of the questions and used different color paper to create more confusion. For example, I'd hand out 1/4 of the test each in 4 different colors, with no two adjacent students having the same color - to discourage the thought of cheating in the first place.
I'll never forget, though, the time that two students in different sections turned in lab writeups with the exact same measurement data - out to 5 decimal places (because that's what the Keithley meters were set to display).
Re:Hmmm ... (Score:4, Interesting)
I went to a Polytechnic, so it was really difficult to cheat, in that most of our grade was made up of group projects and VERY-hands-on timed exams. First year, finals consisted of being given 1 regular Linksys Router, 2 desktops (1 with a fully configured running Windows machine), cabling tools and supplies, a port to the internet, and 2 discs, one of WinServ2K3 and the other was Fedora Core 4, and an HP Printer.
The end result was to have the Windows computer host a virtual server through VMware running the Windows 2003 Server, which had to host active Directoy and a print server, and act as a router for traffic on a specified Class C subdomain. The Linksys router had to act as a router for a Class B subdomain, which the Fedora Desktop had to be on. The end result was that both the host Windows machine and the fedora machine had to be able to print a document. The internet port was for general debugging purposes, though they had blocked every site except the Polytechnic Campuses website (so no Googling!).
You had the whole day (8 hours), If you got it done in the first 4 hours, you guaranteed passed, any longer and the teacher would gauge your progress and how you have things set up. It was the most intimidating test I've ever taken, though I passed - the only way you could cheat really is if you watched someone else and managed to follow them step by step, but then you might run the risk of making the same mistakes they did (and there were mistakes made by just about everyone. I remember getting stuck on having my Fedora Box access the webs properly, linux was not my strong suit*.)
And really, you can't cheat in making a cable, plain and simple - either you know how to do it or you don't. You can't exactly pick up another students cable and look at their colouring scheme without getting noticed. All in all, I think all tests should be done in a hands-on way. We had the benefit of a small class size (maybe 20 students) so I can understand it being impractical for those big 100 people lectures at the university, but really its the best way to cut out cheating. Also, for an arts degree, I wouldn't even know where to start. All they ever do is write endless essays.
Anyways - I got a little side tracked there.
Our Prof - whom was nicknamed Lord of the Strings because he was a bit short and chubby like a hobbit, was commissioned by the Dean of the local university (why they didn't use their own IT/IS/CS department I don't know) to write an application that went through the internet and compared papers to help catch plagiarizing. He even showed us the code, which was quite impressive - almost overwhelming when you are first starting in programming.
You enter in the topic of the paper.
It went through the top lists of essay sites (which you could add or remove sites), and the first 2 pages worth of Google results for the words involved in the topic.
Then it went through a statistical analysis on how similar some papers were. It could easily detect word for word copying, but he also had it set up to detect whether 1 or 2 words in a sentence were changed, and/or if the structure was simply reworked a little. At the end, it would give you a percentage on how much of the paper looked like it was just taken from online. It was then up to the prof to determine if that percentage was high enough to warrant further investigation. It also generated a report based on what sites it found the correlation.
I guess what I'm eventually trying to say is...
Who cheats anymore? You're almost guaranteed to get caught.
Re:Hmmm ... (Score:5, Interesting)
It's okay for colleges to copy so long as they pay the money upfront. ;-)
"the more you copy homework, the lower your grades." - I disagree. If the homework is worth 10% of your grade, then it's better to copy rather than run out of time and never turn it in (a zero). I recall in my engineering classes most of us copied, not because we wanted to, but because the profs so overloaded us with homework that it was impossible to get it all finished.
It was ridiculous - about 10 hours of homework/week for a 3 or 4 credit class. Typical 18 credit load is 50 hours just on homework. Plus 18 hours for the lectures. Plus 10-20 hours on labwork. == 80-90 hours per week!
Hmmm... on second thought maybe they were trying to prepare us for the Real World Suck of 60-70 hour weeks.
Re:cultural differences (Score:3, Interesting)
I had to try hard to avoid saying "I don't cheat so I have to actually understand the material to pass the classes."
You should have said it. I had a friend who was taking a CS test, and the indian sitting next to her leaned over and said, "do you want to share answers?" She said, "no, I prefer to do it myself so I can learn." The indian looked at her, amazed, and said, "that's so impressive." It was as if the idea of doing that had never occurred to him. So if you had said something, you might have helped to change someone's life.
Re:Hmmm ... (Score:5, Interesting)
on second thought maybe they were trying to prepare us for the Real World Suck of 60-70 hour weeks.
That was actually part of our orientation. They drew a graph for us. It went like this:
This line here represents how hard the average worker works in the field.
The first year, you'll be working about this hard: Half of what the average worked works at. Just introducing you to all the concepts, learning, seeing if you enjoy it, that kind of stuff. We don't want to scare anyone off in the first year basically.
Second year, you'll be right on par with the working force. Expect your classes to be a little less than working a job but homework will make up for it.
Third year, you'll be at about 1.5 times the average work load. This is so that when you get out of school, you are more than equipped to work in a variety of fields, and be an expert in each one. We pride ourselves on our graduate employment rate, we have to keep this high.
Fourth year, You'll be at 2 times the average work load. This is so that when your boss comes down Friday at 4:55 pm and says "Holy Gosh Darn Crap, the server room has smoke coming out of it" you can go "No problem chief, I'll have it up before anyone is in on Monday, I better get that raise I asked for".
I've noticed something related to that (Score:5, Interesting)
We have a lot of foreign grad students, Indian and Chinese in particular. Well something I notice especially with the Chinese students but the Indian ones to an extent as well is the idea that all knowledge is something that someone already has. If you do not know the answer to a problem, the correct course of action is to seek out the person or book that does. Everything is already known, you just have to find who knows it. The idea of problem solving is one they don't grasp.
So their computer will break (that's what we do, we are the systems and network support) and they'll come and ask us about it. They get vexed when we say "I don't know what is wrong," they often look at you like you are an idiot, and why don't you go find the person who does?
I remember one time when a lab lost network connection so I was heading down there and he says "Why is the network down?" I said "I'm not sure," that got me a very quizzical look. So we got there and I said "Where's the switch, let's reboot that first," he said "Will that fix the problem?" I said "I don't know." He didn't seem to want to do it, since why bother if it wouldn't fix the problem? I found the switch, rebooted it, and the problem was solved. This was a totally mysterious process to the guy. How the hell could someone who didn't know what the problem was solve it without asking someone who did?
There does seem to be a cultural difference with this, and I think it comes down to the education system. My mom went to teach English in China for about half a year (she used to be a teacher in the US) and said that their version of teaching English was route memorization. Students were presented with a couple hundred phrases per night they were expected to memorize. That was it. Needless to say, that works for shit. The Chinese government realizes it doesn't work very well, which is why they bring in US English teachers, but it is fighting against a cultural attitude of eduction through memorization. Mom said the teachers were very skeptical of her methods (which did not include memorization).
My favorite cheating story.. (Score:3, Interesting)
Was actually an English class. (Yes, I'm an engineer, but I love to read, and I wanted to be a better communicator, so it seemed like a good elective.)
We have a major essay on "The Scarlet Letter". After we hand it in, the prof announces that she did her masters degree on the book. She says she has read everything ever written on the book. And she mentions that she has detected plagiarism. She says "If the cheaters drop this class immediately, I will not pursue charges. Otherwise, expect this to be brought up with the University." Now, I hadn't even looked at other texts. Everything I had written was straight out of my head. I don't cheat normally, but in this case, I knew I couldn't even accidentally cheat.
Next class I show up... 66% had dropped the class. We literally had one third of the students still in the class. It really opened my eyes with regards to how common cheating is.
Oh, and for the record, for those who know the book, my essay had argued that colour vs. black/white was what defined what was acceptable in Hester's world. And thus the 'Black man' was not an outsider, but instead a necessary part. (Kinda along the lines of "There would be no God without Satan, so Satan is actually a positive Christian force, a good guy.") I still remember the response which was "This is entirely original... And wrong. But you did a wonderful job trying to make it work." and I received an A on the paper. So the incident also gave me insight into profs that have seen it all... If you can bring them something original, even if it's wrong, they're just happy to see someone breaking new ground, so they'll give you marks for trying.
Re:I guess I'm old fashioned (Score:3, Interesting)
Yup. As I say, cheating is just admitting to yourself that you're not good enough to win on your own.
Re:Best way to stop cheat sheets... (Score:3, Interesting)
Indeed, the hardest test I ever took in the Navy was one where I could use any resource I wanted *other* than asking another individual for help or advise. But in the interests of full disclosure, that was a practical exam where they took us into a room and presented us with a completely disassembled (old style/washing machine) hard drive and expected us to put it back together and align and calibrate it. It was pass fail too, either the machine worked or it did not.
But not every topic or even every test within a topic is suitable for that style of testing. Different types of tests test for different things in different ways. Consider the difference between an equation and a word problem on a math test for example.
Re:Hmmm ... (Score:3, Interesting)
Here's a way it could help:
You're unemployed for more than 16 months- happens to the best of us, especially in an economy where it is assumed that anybody with 2+ years of experience is completely interchangeable (a wrong assumption, but corporations are full of wrong assumptions). Your unemployment runs out because gasp, the Sentate decided to vote it down just before leaving on a long 4th of July Vacation.
Without these skills you're up a creek. But WITH these skills, you start canvasing your friends and family for people who either just got broadband, are setting up a small business network, or are dissatisfied with the subjective speed of their computer. And with the money you earn from that, you get to eat for another day.
Re:Hmmm ... (Score:5, Interesting)
Sounds like your college is run by people with no significant grasp of reality. Yes, bosses occasionally ask you to pull all-nighters or all-weekenders, but that happens occasionally. If it's happening continuously, then that's a sign that this is a job for first-year college grads that they don't expect you to actually stay in for very long before moving on to a real job. Either that or the company is in a death spiral. Either way, you don't want to stay there.
For a school to pull that on a continuous basis isn't preparing you for the real world. It's preparing you to go postal and shoot up the campus. Just saying. An 80 hour week is simply unsustainable for more than about two or three weeks at a time, whether you're talking about a workplace or a college. You're going to spend, at minimum:
That's 12 hours and 45 minutes, which leaves 11 hours, 15 minutes MAXIMUM that you can usefully use. An 80 hour week requires 11 hours, 26 minutes on average per day. So it simply can't be done without cutting into something that's actually critical for your health and well being, and that's if you don't take a single minute to just relax and enjoy life, don't attend any sort of religious institution, aren't in any fraternity or sorority, don't get any actual exercise beyond walking to/from class, etc. In short it's very unhealthy.
Over long periods of time, such insane levels of work lead to serious mental health problems. If your school is truly working you 80 hours per week, you should contact a mental health professional and have them do a study on your school's population. I suspect you'll find higher than normal rates of anger management problems, severe inability to concentrate due to sleep deprivation, and a significantly elevated rate of depression and suicidal thoughts. It simply is not healthy to work people that hard over an extended period of time. It's so unhealthy that nearly every civilized country in the world (except the U.S.) has laws limiting work hours to significantly less than that.
And it's not just unhealthy in the short term. College is a critical time in people's lives for creating new social relationships. For most people, their first thirteen years of education was spent with mostly the same people. College is the first time that they break away from that, and it is critical that they have sufficient time during all four years to get used to making new friends quickly. It's going to happen when they move from job to job later in life, and if they aren't prepared to handle that, it can be socially damaging, even devastating. When you lose a job, your whole social network goes away. Preparing students for that is at least as important as any academic information that they can impart---maybe even more important. The social learning will still be useful in twenty years, long after any detailed technical knowledge has become dated and stale.
Also a substantial percentage of marriages occur because of people meeting in college. With an 80-hour instruction week, such socialization becomes nonexistent, leading to even further elevated rates of depression in graduates down the road. The lack of adequate time to socialize also results in those graduates having a hard time with social interaction after they graduate simply due to not having socialized much for four years. This further exacerbates the problem.
Finally, such
Re:I guess I'm old fashioned (Score:3, Interesting)
I'd rather hire a dumb hard worker than a lazy genius.
tech solutions to societal problems don't work (Score:1, Interesting)
"We will not lie, steal, or cheat, nor tolerate among us anyone who does" is the US Air Force Academy honor code. Cheating still occurs occasionally, but when it's discovered, the consequences are immediate and severe.
One of the benefits is the fact the honesty is expected and assumed to the extent that no special 'anti-cheating' measures are needed. I still remember the occasional 'take home' mid-term exam and, to the best of my knowledge, NOBODY ever cheated.
Re:Best way to stop cheat sheets... (Score:3, Interesting)
Agreed. I actually thought take-home tests were among the best. Yes, there was potential for cheating, but aside from directly askign another student for the solution, we were *expected* to utilize the toosl we had. Reference books, websites, even (near-exact quote) "this is a tricky operation, so you might want to implement your solution on your circuit board to test it. You can start from the project 3 code if you want".
The reason I loved this type of "test" is that it is very much how the real world works. The professor knew we had access to tools that would allow us to answer the questions, but also knew that we didn't have the answers to most of them directly (the exact register settings needed to set up the clock in that way hadn't actually been discussed in class, and since he'd written the coding assignment the week before it was unlikely there'd be a solution online...). We had to combine what we'd learned in class about the general principles of an embedded system, plus the tools and references for our specific board, with the ability to use those tools and references (and posibly find more online).
We had a week to do probably about 4-5 hours of actual work, but only somebody who was already an expert on that system (and on the class material, since some of it was also textbook-type stuff) could have done in in that short a time. Instead, what we were tested on was whether we'd picked up the skills needd to become (enough of) an expert in a week to solve a problem that was essentially another 3-4 hour lab assignment, without the instructions containing a basic framework for solving the problem and a list fo the relevant references that had the details we'd need.