catching Chegg cheaters

[evil_physics_witchcraft]

It would be pleasant to plant some little red flags in there to retrieve when proving the point to administration. Like "Charles Humphries exerts g gravity, currently heading east at ten E = R S." (Only, you know, one that made sense.) Spells out CHEGG CHEATERS.

I purposefully write questions that are identifiable as products of my mind. For example, I use uncommon names, write out prefixes and units sometimes instead of abbreviating. These students were foolish enough to even copy my instructions and include them in their posts on chegg (Some did screenshots)! Really disheartening.

[Caracal]

Unfortunately, I also found some cheaters who used chegg. Our institution would not allow us to use Respondus or a webcam (no comment), so I knew something like this would happen. I managed to nail down the days that the questions were asked on chegg and matched them up with the days that students took exams. Unfortunately, I cannot match up times (no obvious time stamp on chegg). One poor soul is the only one of my students to take my exam on the same day a chegg question was asked for that test. This particular question is very unique- I drew the diagram myself and reworded the question.

My admins have been alerted and I'm awaiting further instructions.

Is that conclusive though? Couldn't another student have taken the test before, and then sent the questions to another student who asked the question on Chegg? That seems totally plausible to me. How fast is the turnaround time on Chegg anyway? Would you really be able to get an answer to your question in time for the test?

I looked at the source code to find the date the questions were posted. All were posted this semester during the times my exams were open. The questions are also very unique and easily identifiable as items that I created. I have spoken with my Chair and it's moving up to the Dean.

Right, but you seem to assuming that only a student taking the exam that day could have access to your exam. But that's pretty clearly not true. Someone could easily have taken it earlier and sent questions to a friend, who posted them on Chegg and then took it on another day. For all you know the guy taking the test that day had nothing to do with it.

[evil_physics_witchcraft]

Unfortunately, I also found some cheaters who used chegg. Our institution would not allow us to use Respondus or a webcam (no comment), so I knew something like this would happen. I managed to nail down the days that the questions were asked on chegg and matched them up with the days that students took exams. Unfortunately, I cannot match up times (no obvious time stamp on chegg). One poor soul is the only one of my students to take my exam on the same day a chegg question was asked for that test. This particular question is very unique- I drew the diagram myself and reworded the question.

My admins have been alerted and I'm awaiting further instructions.

Is that conclusive though? Couldn't another student have taken the test before, and then sent the questions to another student who asked the question on Chegg? That seems totally plausible to me. How fast is the turnaround time on Chegg anyway? Would you really be able to get an answer to your question in time for the test?

I looked at the source code to find the date the questions were posted. All were posted this semester during the times my exams were open. The questions are also very unique and easily identifiable as items that I created. I have spoken with my Chair and it's moving up to the Dean.

Right, but you seem to assuming that only a student taking the exam that day could have access to your exam. But that's pretty clearly not true. Someone could easily have taken it earlier and sent questions to a friend, who posted them on Chegg and then took it on another day. For all you know the guy taking the test that day had nothing to do with it.

Right. I probably should have mentioned that in one class, only one student took the test on the day a problem was posted (1st instance of the problem being posted). So, in that case, this student was the first person to see the test and the problem showed up on the day he/she took it. Yes, the student could have passed on this information to someone else, but there is a correlation between when the student took the test and when the problem showed up on the site. Again, these problems are VERY different from test bank stuff, so it is highly unlikely that someone else came up with the exact same problem, wording and image. Seems pretty damning to me.

I am *hoping* to get some kind of timestamped data and also user data (within reason) to match up at least the student who posted the problems. Yes, I may not get everyone. Yes, info may have been passed along. We shall see what happens.

[clean]

What is the processing for seeing if your problems are posted on Chegg?

[evil_physics_witchcraft]

What is the processing for seeing if your problems are posted on Chegg?

I googled the problems (for the hell of it and just to see what's out there). 1st hit was on Chegg. On one of my tests, 75% was up there. I then went to the site and did more investigating.

[clean]

Well, I googled mine, but Im not finding anything.  Well, actually, ... as part of the problem I included the names of Maryanne and Ginger (and the rest in one way or another (except Gilligan)... that made THEM him!)

So I searched the first 2 lines of my question and ALL I get are Gilligan's Island posts!  So let that be a lesson to you!  Never name anyone Ginger Maryanne (or the other way), or they will be lost on The Google!

Maybe that is what we should all do to our tests.  Include 'famous' or even Infamous names so that IF students search for the question online they get some history  Lesson.  (Start a problem with Sherman Lee Grant, and let the rest fall into place!
Or maybe "Jethro Bodine was ciphering the distance betwixt Bugtussle and Hooterville a fortnight ago. "   
If line 2 includes "Andy and Barney" or Wally and The Beaver"  then the Chegg notations may no longer be the top of the search results!

[Caracal]

Unfortunately, I also found some cheaters who used chegg. Our institution would not allow us to use Respondus or a webcam (no comment), so I knew something like this would happen. I managed to nail down the days that the questions were asked on chegg and matched them up with the days that students took exams. Unfortunately, I cannot match up times (no obvious time stamp on chegg). One poor soul is the only one of my students to take my exam on the same day a chegg question was asked for that test. This particular question is very unique- I drew the diagram myself and reworded the question.

My admins have been alerted and I'm awaiting further instructions.

Is that conclusive though? Couldn't another student have taken the test before, and then sent the questions to another student who asked the question on Chegg? That seems totally plausible to me. How fast is the turnaround time on Chegg anyway? Would you really be able to get an answer to your question in time for the test?

I looked at the source code to find the date the questions were posted. All were posted this semester during the times my exams were open. The questions are also very unique and easily identifiable as items that I created. I have spoken with my Chair and it's moving up to the Dean.

Right, but you seem to assuming that only a student taking the exam that day could have access to your exam. But that's pretty clearly not true. Someone could easily have taken it earlier and sent questions to a friend, who posted them on Chegg and then took it on another day. For all you know the guy taking the test that day had nothing to do with it.

Right. I probably should have mentioned that in one class, only one student took the test on the day a problem was posted (1st instance of the problem being posted). So, in that case, this student was the first person to see the test and the problem showed up on the day he/she took it. Yes, the student could have passed on this information to someone else, but there is a correlation between when the student took the test and when the problem showed up on the site. Again, these problems are VERY different from test bank stuff, so it is highly unlikely that someone else came up with the exact same problem, wording and image. Seems pretty damning to me.

I am *hoping* to get some kind of timestamped data and also user data (within reason) to match up at least the student who posted the problems. Yes, I may not get everyone. Yes, info may have been passed along. We shall see what happens.

Oh ok. He was the first person to take the test? If so, that makes more sense.

[clean]

I have reached the BIG LEAGUE!  My final exam is on Chegg!

Unfortunately, Every one of them got a different version, so I have been able to figure out exactly who posted the question.

When I emailed the students the take home portion (what is on chegg now) I told them that they were not to talk to anyone about the exam and all work must be their own. 

[marshwiggle]

I have reached the BIG LEAGUE!  My final exam is on Chegg!

Unfortunately, Every one of them got a different version, so I have been able to figure out exactly who posted the question.

We'll all be looking forward to updates on this........

[Caracal]

I have reached the BIG LEAGUE!  My final exam is on Chegg!

Unfortunately, Every one of them got a different version, so I have been able to figure out exactly who posted the question.

When I emailed the students the take home portion (what is on chegg now) I told them that they were not to talk to anyone about the exam and all work must be their own.

Oops, now that's good evidence...

[evil_physics_witchcraft]

I have reached the BIG LEAGUE!  My final exam is on Chegg!

Unfortunately, Every one of them got a different version, so I have been able to figure out exactly who posted the question.

When I emailed the students the take home portion (what is on chegg now) I told them that they were not to talk to anyone about the exam and all work must be their own.

Oops, now that's good evidence...

Dang clean. I'm interested to find out what happens with this. Was it just one version of the test?

[clean]

I teach finance.  I copied a test I wrote 20 years ago into Excel (line by line as Word and Excel are not all that cozy for what I wanted to do. Then I created an answer key from the Master Copy of the test.  I then copied from the master to a separate worksheet and named the tab with the student's last name. Several of the numbers were then changed throughout the test.
As I changed the cell locations in the questions, the answer portion generated the separate answers for each student. 
After the tests were done I copied the first several columns that housed the questions, stripped out the color coding I used to mark the numbers to change, copied to a separate file and then emailed each student their personal excel file. 
When I found the Chegg page I figured which main subgroups the exam fit (was it a 3 year or a 7 or a 10 year lived asset;  Did it use MACRS or Straigtline depreciation, etc), and then I looked to see which particular tax rate, unit sales and prices I was searching for.  Only ONE exam matched (as it should as I was pretty sure that I had plenty of numbers and changed several of each). 

On exam day, they will upload their documents to Blackboard and then take a fill in the blank test.  I already know the answers so they will answer specific questions about the work they did.  What is the Revenue in year 3?  What is the Depreciation Expense in Year 3?  What is the Initial Outlay in Time Zero? 

Every one gets the same questions, but the answers are all unique.  I tried, but it is hard and I know that this didnt work well, but I tried to have it so that a mistake in one part does not cost them points from several questions.  So IF they  miss the Initial Outlay, then the Net Present Value has to be wrong as well, so I tried to ask the questions so that they can make mistakes but get partial credit.  That is the hard part.

Worse, I have more instances of inappropriate behavior! They have all had their exams more than a week.  The exam is Monday. I told them that if they sent me their first year's cash flows, that I would compare their answer to mine and let them know if we agreed, but they had to submit by 8pm (as a reward for those that are working on it now, they can get some feedback if they are doing it correctly).  Two submitted, not the cash flow, but their spreadsheets. They were submitted at almost the same instant.  The spreadsheets for their work is similar, but that is not too unexpected because i showed them all a template to use to answer such questions.  What I did not expect was that each had their first year cash flows in Red!  Usually that means negative, so I checked the work.  It was just red.  Not negative.  ANd only the first column value.  Why would 2 "independently created" spreadsheets both have a non negative number highlighted in red, I wonder?   

So I emailed them that they had better be sure that they were working on the test alone as instructed!  I emailed the class the several instructions that indicated that they were to work alone!  I highlighted the "Work alone" part in yellow.
I have not heard from the Students in Red   objecting to the warning to work alone!  I will certainly have to go through those 2 spreadsheets with a fine tooth comb to find similar Divergence from the Instructions!


So, no there are 45 versions of the test in total.  One for everyone! 
If we were not CV19 separated, I would not have given a take home test! BUT WHEN I DO give a take home test, it better be only the work of the test taker!! 

[evil_physics_witchcraft]

Very nice clean! I also make multiple versions of physics tests and I use spreadsheets to generate numbers for lab exams, etc. This is why I know that more than one person cheated in my class. There were two versions (different numbers) of the same problem online.

[dr_codex]

Not Chegg, but I often catch students cheating on essays because they have done one of three things:

1) Left obvious traces of cutting and pasting text-- changes to font, and even hypertext links. My students rarely know how to "scrub" text, and my plagiarists are usually too lazy / stressed / clueless to hand-type the original.
2) Copied the same thing as somebody else in the class. Sometimes I wouldn't know to look for a common original if I didn't have the replicants in front of me.
3) Written something that's too good. (One of my Engineering colleagues caught somebody this year for this reason.) I have a friend who was offended that somebody bought the "C" paper rather than shelling out another dollar for the "A", but Stu Dent knew he'd never be able to pass off the A. He didn't even get away with the C.

Hold the line, everybody. Hold the line.

[hamburger]

How about posting a bunch of incorrect answers using different accounts?