College Guide


July 14, 2010 2:46 PM Telling the Truth about Cheating

By Daniel Luzer


Schools and the media have lately gotten all worked up about the dreadful, epic academic dishonestly going on in the Republic’s hallowed institutions of higher learning. The worry, of course, is about students cheating, not professors. Let’s find a way to stop to those bad students, perhaps by using expensive private companies.

But as one writer explains in the New York Times’s Room for Debate section, maybe colleges are worried about the wrong thing. As Alfie Kohn writes:

Rather than counting the number of students who cheat, or figuring out how to catch (or deter) them, I’d prefer to ask two questions that rarely figure in these discussions: What kinds of teaching elicit cheating? And what assumptions and values lead us to define some acts as cheating in the first place?

As Kohn explains, students are more likely to cheat if teachers don’t have much valid interaction with students and in situations where the grade appears to matter more than the learning. If students see actual work as pointless and the grade as the only reason for work, of course they’ll be more likely to cheat and find their way around actually doing the work. “Taken seriously, these results invite us to stop playing “gotcha” or focusing on the mechanics of cheating and trying to stay a step ahead of the students,” said Kohn.

Kohn offers one of the most interesting perspectives on academic dishonestly available right now. What he invites us to ask is, essentially, do you want to just catch academic dishonesty or do you want to curtail it?

The truth is if instructors assign projects on which students can’t cheat, or can’t very easily cheat (essay exams, problem sets, term papers requiring outlines, bibliographies, and multiple drafts) they won’t. More importantly, they’ll learn the information better.

This obviously isn’t possible with all information and all subjects but it’s important to understand why people feel compelled to cut corners.

Cheating doesn’t happen because students are particularly bad or lazy. It doesn’t even really happen because of the ease of gathering electronic information; it happens because the way students are expected to demonstrate understanding makes cheating worth it. [Image via]

Daniel Luzer is the news editor at Governing Magazine and former web editor of the Washington Monthly. Find him on Twitter: @Daniel_Luzer


  • Crissa on July 14, 2010 6:56 PM:

    Multiple drafts suck. With modern typesetting software, they hardly exist like they did in the past, and even so, they require a single type of workflow. They completely ruin alternate methods of writing and require changes for the sake of changing, and extra typing and work, for the sake of work: The exact set of conditions set forth in your premise.

    However, it's not a bad idea to look for methodological conditions of cheating.

  • Anonymous on July 14, 2010 7:31 PM:

    Coming from the Humanities, I can say that no matter how creative the assignment and how open the teacher is to exploring ideas, cheating still happens, often in the form of plagiarism.
    I feel that this needs to be confronted at the institutional and cultural levels. Students stress over their GPA even when they have no idea what they want to do after college. Some students would cheat at yoga if it affected their GPA. I agree that students need to be able to see some point to what they are learning, but this needs to start earlier in the education process and should be reflected in the structure of an institution's curriculum design.
    It is not the professor's job to entertain students, so as to trick them into thinking learning is easy and fun all the time. The pursuit of knowledge is hard work and students need to understand that the knowledge is the point, not the grade. Aside from grad school applications, no one cares what your GPA was after you graduate, they just want to see what you can do, what knowledge you bring to the table.

  • Tom Dibble on July 14, 2010 7:52 PM:

    There are a few things which can be done.

    Cultivate a culture of honesty. You don't have epidemics of cheating at the Air Force Academy, for instance, because it has a very strong culture which forbids such. Generally speaking, smaller schools, or to a lesser extends schools-within-schools tend to foster this in a "small town" way. Students know who is cheating, and they pressure them or at least don't enable them.

    Grade work, not answers. Given a sheet of math problems to answer, the student who made a stupid mistake in the first step but propagated that error successfully should get almost full credit; the student who shows no work, or whose work leads to an answer completely different from what they write down at the end, should get no credit at all.

    Investigate abnormal commonalities amongst answers. Given long-form answers (see above) there should be a fairly large degree of variance between one student and the next. Where there is not, a red flag should go up (but it's not necessarily a problem; the commonality may come from both cribbing too closely to the example set in a book or the wording of the textbook).

    Open book exams. Make bringing in extra materials meaningless. You have your books and notes. But you also have a time limit. If you have to look everything up you will run out of time.

    Never "grade" homework (but check it). Homework is asking for "cheating". A dominant school of thought indicates that one-size-fits-all homework assignments are almost always counterproductive to learning if they compose a portion of the students' final grades. Homework should be a tool for learning, and the exam a test that the learning has been accomplished. A student who "gets it" after the first page of homework shouldn't have to stick with the next five because another student takes six pages to understand this particular concept.

    Test cumulative understanding. This is to avoid the main consequence of "high-stakes" exams, which is that a student will cram hard for a particular test and then forget everything the next day. A test towards the end of the year should include a smattering of questions requiring topics covered at the beginning. Ideally, this would not just include the single course but would extend throughout the curriculum (a first-quarter exam should include topics covered the previous year).

    Why wouldn't these be done? It's more expensive to have the small classrooms required for the first. It's more expensive to hire someone to grade work rather than scan down a list of answers or better slip a scantron sheet into a machine, so the second loses on cost as well. If you aren't grading work, you have little evidence to go on for commonalities, so the third isn't implemented. Open book exams are harder to write than "random fact memorization" exams, and open book exams including cumulative topics are harder yet. Finally, if they don't include significant cumulative knowledge checks then using homework as a teaching aid rather than an uncontrolled "lite" test causes educational problems, and it's highly institutionally engrained that homework must be graded, so homework tends to count as a significant part of the grade.

    I'd recommend the work of Ken O'Connor on the subject.