[Bioteaching] advice for addressing widespread cheating on exam

Fri Mar 19 09:50:37 PDT 2021

Hi Michelle,

Ditto on the upper-division grading approach. Basically, I tried to translate that down to lower division courses.

Mike

---
Michael J. Harms Ph.D.
Associate Professor
Chemistry & Biochemistry
Institute of Molecular Biology
University of Oregon
harmslab.uoregon.edu<http://harmslab.uoregon.edu>

541-346-9002 (work)
541-968-0035 (cell)

On Mar 19, 2021, at 09:47, Anne 'Michelle' Wood <miche at uoregon.edu<mailto:miche at uoregon.edu>> wrote:

HI MIke, I use that approach on a lot of things, and almost entirely in my 400/500 classes. I'm not a big fan of using grades to create 'rankings' anyway and, if everyone can get it and do good work, all "A"s is fine with me.
For the purposes of grad schools and medical schools, I think there would still be enough difference among gpa's of applicants with this approach to give them the information about students that they actually need.

 Michelle
________________________________
From: bioteaching-bounces at lists.uoregon.edu<mailto:bioteaching-bounces at lists.uoregon.edu> <bioteaching-bounces at lists.uoregon.edu<mailto:bioteaching-bounces at lists.uoregon.edu>> on behalf of Mike Harms <harms at uoregon.edu<mailto:harms at uoregon.edu>>
Sent: Friday, March 19, 2021 9:42 AM
Cc: bioteaching at lists.uoregon.edu<mailto:bioteaching at lists.uoregon.edu> <bioteaching at lists.uoregon.edu<mailto:bioteaching at lists.uoregon.edu>>
Subject: Re: [Bioteaching] advice for addressing widespread cheating on exam

Hi All,

@Kelly: This probably doesn’t help your current situation, but I thought I’d toss out how I thought about this when I taught two relatively large remote courses this Fall (BI281H and CH360).

I took the radical step of saying up-front that everyone could work together and could use any resource on every assignment, including exams.

In my view, anything technically feasible will be done. I can’t keep students from working together or accessing the Internet, just like legacy record companies could not prevent people from sharing music files back in the day. I felt that if I made working together against the rules, the students at the greatest disadvantage would be those that followed my rules.  This disincentivizes rule-following and hurts the most honest students.

I therefore gave out challenging, reasoning-style exams with only one rule: each person had to draw their own graphs and write their own text.

So what happened when I did this?  Everyone did really, really well.  The exam quality was top-notch, across the board.   I also observed very little evidence of copy-paste style cheating.  There were clearly groups that worked together and had similar lines of reasoning, but students generally followed the “own text” rule.  In talking to some of the students, it basically turned the exams into high-stakes homework sets. The stronger students had to explain their reasoning to the weaker students, and the weaker students had to pose questions to the stronger students and rearticulate what they understood. My gut says the students learned more in this way than they would have with a conventional exam.

The downside, of course, is that I don’t really know what the students learned because I my exams had no power as assessment instruments.  Many students received As that would have certainly not gotten As under a conventional exam/grading scheme. If the goal of grading is to sort students into categories of success, my scheme failed miserably. It is also probably not sustainable long term. But as a stop-gap measure in this extraordinary situation, I think it worked pretty well.

Mike

---
Michael J. Harms Ph.D.
Associate Professor
Chemistry & Biochemistry
Institute of Molecular Biology
University of Oregon
harmslab.uoregon.edu<http://harmslab.uoregon.edu/>

541-346-9002 (work)
541-968-0035 (cell)

On Mar 19, 2021, at 09:14, Mark Carrier <mcarrier at uoregon.edu<mailto:mcarrier at uoregon.edu>> wrote:

Hi Kelly,

I'm so sorry to hear this, and not in the least surprised.

We used a combination of proctoring, the fabricated story that we were making use of a new Canvas academic integrity tool set (fooled my TAs even), lots of Canvas formula questions, lots of similar looking but different questions between versions of the tests, puzzles using data from recent papers but with the gene or protein names changed, and a relatively time-constrained open-notes exam format. I'm pretty sure we thwarted most cheaters from cheating productively on exams.

But on all the other assignments (like homework sets and lectures embedded within quizzes) they cheated with gusto. We had access to a social media site a large number of students were using, using Discord software, and even though I was very specific about what sort of group work was encouraged and what constitutes cheating, they cheated.  They shared confirmed answers, for instance. My putting the homework problems on an exam and pointing out that by cheating they had shot themselves in the foot only pissed them off. By "them" I mean a small loud set who behaved disgracefully all term. It was as depressing to witness as national politics, with conspiracy theories and vaguely threatening posts directed at those who might consider "narc-ing" about what was being posted. Social media really brings out the worst in people.

To counter this we are setting up the Discord site ourselves next term. Students can go form a different site of course, but by creating the default place to go chat about the class we hope to keep those less motivated to cheat and bully from participating in the worst behavior.

I just keep reminding myself that most of them don't cheat. In a class of 360 the number of bad actors feels large, but is still a relatively small fraction of people in the class. At least on assignments on which cheating is inconvenient.

Mark

On Mar 18, 2021, at 4:31 PM, Kelly Sutherland <ksuth at uoregon.edu<mailto:ksuth at uoregon.edu>> wrote:

Hi all,

I am grading the Canvas online final exam for BI357 (Marine Biology) and there’s evidence of widespread cheating.  I don’t think it’s realistic to pursue each student that I suspect of cheating and it can also be hard to prove.  For example, some students had answers that were technically correct but included information that we never discussed in class.

I am not shocked but definitely disappointed and am wondering what other folks have done to address instances of widespread academic misconduct, especially this year.

Would people be willing to share their experiences/approaches?
Many thanks,
Kelly

________
Kelly R Sutherland, PhD (she, her)
Associate Professor of Biology
Oregon Institute of Marine Biology
University of Oregon

Office: 473B Onyx Bridge
Phone: 541-346-8783

sutherlandlab.org<https://urldefense.com/v3/__http://sutherlandlab.org/__;!!C5qS4YX3!RHMCol_4JVjY1fytqwbekjqbIyR7Vln6K2yU358QBypJpoEBG2_4X-3Vw2MS2ys$>
@SutherlandLab

_______________________________________________
Bioteaching mailing list
Bioteaching at lists.uoregon.edu<mailto:Bioteaching at lists.uoregon.edu>
https://lists.uoregon.edu/mailman/listinfo/bioteaching

_______________________________________________
Bioteaching mailing list
Bioteaching at lists.uoregon.edu<mailto:Bioteaching at lists.uoregon.edu>
https://lists.uoregon.edu/mailman/listinfo/bioteaching

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.uoregon.edu/pipermail/bioteaching/attachments/20210319/1dd15f02/attachment-0001.html>