Introduction
Does your organization worry about employees or students cheating on exams?
Many organizations implement countermeasures such as proctoring and browser lockdown. But these measures are not actually very effective. They can also create, in Dr. Stephanie Moore’s words, “a culture of distrust and surveillance.”
Rather than having a reactive approach, organizations should be proactive about addressing cultural and environmental conditions that lead to cheating.
This post summarizes a presentation by Dr. Stephanie Moore about how to prevent cheating while improving learning outcomes. Throughout her presentation, Dr. Moore cites the book Cheating Lessons: Learning from Academic Dishonesty by James Lang, which has a higher-education focus but is widely applicable.
Is Cheating Rampant?
Before we get into specifics about how to prevent cheating, first let’s consider whether it’s really a widespread problem. As Lang says in his book, “alarmist arguments about the state of cheating . . . lead to alarmist responses.” But do we need to be alarmed?
Most people will admit to having cheated at least once in their academic career. (Depending on the study, the number is 75 to 87 percent.) When asked if they’d cheated three or more times, the number drops to 19 to 38 percent (again, depending on the study).
But the number of people who cheat regularly is very low. In addition, students who cheat are rarely dishonest in other contexts. So what makes people cheat on a test when they wouldn’t behave dishonestly elsewhere?
Conditions That Lead to Cheating
Cheating is strongly driven by the learning environment. The system (policies, resources, rewards, and consequences) can create conditions that lead to cheating.
"If you put a good performer against a bad system, the system will win almost every time."
The Princess Alice experiment showed that most children would not cheat at a game if they believed an invisible princess was watching them. But, before the princess was introduced, the first part of the experiment also clearly demonstrated that a significant number (42%) would cheat if success was impossible otherwise.
Dr. Moore’s presentation listed the following conditions that lead to cheating:
- Strong emphasis on performance
- Extremely high stakes
- Extrinsic motivation (e.g., grades, payment, or entry into profession)
- Low self-efficacy (i.e., learners lack confidence in their ability to complete a task successfully)
These conditions are not conducive to learning.
Workplace Learning Example
As a real-world example with adult learners, let me tell you about a training center I worked with as part of a team. For many participants, the stakes were very high. If they failed a class, they lost their job. No second chances, no do-overs. (The training center had no control over that policy.)
If that weren’t bad enough, for many years, the training center had operated without instructional designers. Subject matter experts developed and taught the courses. They also wrote the exam questions, which weren’t mapped to learning objectives. In fact, many of the courses didn’t even have learning objectives.
Participants had no way of knowing what content from a week-long course would be on the exam. Furthermore, exam questions often didn’t match what was taught in the class. In fact, some instructors took pride in developing “tricky” test questions.
To make matters worse, the end-of-course exam was the sole determining factor in whether participants passed or failed the class.
It was a recipe for cheating.
Conditions That Reduce Cheating
No amount of proctoring was going to fix the problem at that training center. Instead, we had to address the learning conditions.
Here’s a list of conditions that reduce cheating and improve learning, as shared by Dr. Moore:
- Emphasis on mastery
- Low, more frequent stakes
- Intrinsic motivation
- High self-efficacy
Instructional Design Strategies to Prevent Cheating
Let’s look at some instructional design strategies recommended by Dr. Moore and how we used them in the redesign of the training center courses to improve learning and prevent cheating.
- Use multiple, more frequent assessments. For the training center, we redesigned courses to include multiple exercises throughout the week. Participants performed real-world job tasks in the context of scenarios. Instructors evaluated their performance using rubrics. Successful course completion was based on a combination of scores on various exercises and tests. Thus, we de-emphasized exams and helped build self-efficacy.
- Use formative assessment. Rather than waiting until the end to evaluate learning, we incorporated quizzes, tests, and exercises throughout the courses. This helped instructors gauge participants’ mastery at various checkpoints all week, so they could clarify concepts as needed.
- Use feedback and opportunities to apply feedback for improvement. We built in multiple practice opportunities, and instructors gave feedback during exercise debriefs and on performance evaluation rubrics. This helped participants understand what skills they needed to improve along the way, again building learner confidence.
Why These Strategies Work
In Dr. Moore’s presentation, she explains that these strategies lead to better learning for a number of reasons:
- Frequent testing is an effective learning strategy that outperforms studying alone.
- Spaced repetition and retrieval are highly effective for learning.
- Knowledge gained in these ways is less likely to be inert. Students are less likely to “binge and dump.”
"Learning is a delta, not a result. And so, how do we facilitate that delta—or that change—over time?"
The course design strategies we implemented at the training center led to a shift in the overall culture. Participants began to focus more on learning throughout the week than on achieving a certain score on an exam. They stopped asking, “Will this be on the test?” and started asking questions that would help them apply what they were learning to their jobs.
Priming and Timing
Another effective strategy Dr. Moore mentioned was “Priming and Timing.” Students are primed to recall a code of ethics, which is timed directly before a test. She referenced the Ten Commandments experiment, in which some students were asked to list the Ten Commandments before taking a test. Meanwhile, others recalled ten books they had read in school. No students who thought of the Ten Commandments right before a test cheated, even if they were not Christian themselves.
At the training center, we asked students to sign a code of conduct at the beginning of the class. Then, right before taking a computer-based test, they checked a box reaffirming adherence to the code of conduct. I don’t have any data to tell me if it helped reduce cheating, but it didn’t hurt.
(It’s worth mentioning that a more recent study was unable to replicate the results of the Ten Commandments experiment.)
Summary
If your organization is concerned about cheating, consider whether the learning environment is creating conditions that lead to cheating and hinder learning.
Conditions that increase cheating and impede learning:
- Strong emphasis on performance
- Extremely high stakes
- Extrinsic motivation
- Low self-efficacy
Conditions that prevent cheating and improve learning:
- Emphasis on mastery
- Low, more frequent stakes
- Intrinsic motivation
- High self-efficacy
"While we should never let cheating students off the hook, or blame ourselves for the problem, we owe it to them—and to ourselves—to consider how our course design and classroom practice can give them all the help they need in order to support academically honest work.”
Resources for Learning More
Many thanks to Dr. Stephanie Moore for giving me permission to summarize the information in her presentation. I invite you to watch the full presentation (from the first link in the list below) and check out the other resources from which the information was gatherred.
- Assessing Retention and Comprehension Online and Addressing Cheating, by Dr. Stephanie Moore (presentation)
- Cheating Lessons: Learning from Academic Dishonesty, by James M. Lang (book)
- Improving Performance, by Geary Rummler and Alan Brache (book)
- Our Buggy Moral Code, TED Talk by Dan Ariely (He begins talking about cheating at about 04:15 and the Ten Commandments experience at about 7:18.)
- Princess Alice is Watching You: Children’s Belief in an Invisible Person Inhibits Cheating, by Jared Piazza, Jesse M. Bering, and Gordon P. D. Ingram (journal article)
- The Dishonesty of Honest People: A Theory of Self-Concept Maintenance, by Nina Mazar, On Amir, and Dan Ariely (journal article)
- The Honest Truth About Dishonesty, by Dan Ariely (article)
- Universities Could Be Encouraging Students to Cheat Without Even Knowing It, by James M. Lang (article)
As an Amazon Associate, I earn a small amount if you purchase a book from one of the above links. This does not affect the price you pay and it helps to support this blog.
Comments