As noted on the Academic Integrity Background page, there are three dimensions to consider in order to reduce academic misconduct: rationalization, pressure, and opportunity. Below, strategies are provided that can be employed with each dimension to reduce misconduct.
Rationalization: make integrity the right choice
- Communicate and reiterate high expectations for integrity in your course and at the university
- Include clear statements in the syllabus and with each activity (assignment, exam, etc.) regarding what constitutes acceptable and unacceptable behaviour
- On an exam, make it absolutely clear what aids and other resources students are and are not allowed to access
- Include an integrity pledge with exams and other graded assignments whereby the student affirms that they have read, understood, and will follow / have followed the academic integrity principles established for that assessment. (For sample wording, see the Academic Integrity Pledge page.)
Pressure: reduce pressure to cheat
- Utilize more more low-stakes assessments, and a variety of assessments in your course
- Provide frequent and early feedback (formative and summative) in your course
- Include policies such as “only count the best 5 of 6 quizzes” or flexible grade weighting
- Encourage a culture of collaboration with learning as the goal, rather than one of competition with grades as the goal.
Opportunity: make it difficult to cheat
Administrative Strategies for Exams
- Limit time on exams
- Exams should be scheduled for specific dates and times (note allowances may be required for students in other time zones)
- Exams should close when the time allotted for the exam is reached
- Consider designing an exam to occupy the full time available
- If a student is busy with the exam, there is less opportunity for misconduct
- This should be done with caution to avoid increasing pressure (for example, design an exam where a student can comfortably earn 90% of the marks in 3/4 of the allotted time, and they then have the rest of the time for the last 10%)
- Control student access to online exams
- Provide some extra time for internet connection issues (~15 minutes), but not too much extra time
- Display only a single question on the screen at a time (to make it more difficult to copy and share exam questions)
- Do not allow students to return to questions they have completed (i.e., “no backtracking”) – this should be used with caution as it can actually increase student stress and pressure, it goes against common recommendations for students to view the entire exam before beginning, and it may not be feasible/permitted in some platforms/institutions.
- Allow each student to access an online exam only one time
- Consider the use of proctoring software
Implementation Strategies for Exams and Assignments
- Carefully consider what resources students will be allowed and prohibited (and clearly communicate this to students, as described above).
- A closed-book exam is very difficult to run without proctoring
- A partially open-book exam requires steps to mitigate access to online resources as well as collusion between students (and therefore likely also requires proctoring)
- A fully open-book exam (including full internet access) does not require the same strict proctoring, but still requires steps to mitigate collusion between students.
- To minimize collusion:
- Generate multiple questions on a given course concept (i.e., create question “pools”), and assign a randomized subset of questions to each student (note: most LMS quiz tools can be set up to do this automatically, or you can generate multiple exam variations)
- Randomize the order in which questions are asked (this is easy to set up in an LMS using question pools, or you can distribute different versions of the test to different students)
- Lower-level questions (i.e., testing recall, comprehension, and application), are generally more suitable for generating large pools of questions, as the questions are typically shorter, thus making each student’s exam more unique. In addition,
- Randomize answer choices (for matching, multiple choice, etc.)
- Use parametric calculation questions (where numbers are randomly assigned to each student)
- Some questions may make it possible to have students randomly pick a number or use digits from their student number or similar as input variables
- Higher-level questions (i.e., evaluation, analysis, and synthesis) are generally more difficult to collude on (i.e., more difficult to fully describe to someone else or share a complete answer) compared to lower-level questions, and the answers are often more nuanced and involve judgement and more thoughtfully developed arguments.
- Some questions may allow students to pick an example from their daily life, a work term, a team project, etc. as the basis for the question
Additional (related) perspectives and Resources
For exams without remote proctoring software, see also the 2-page quick guide from the Engineering Collaborative for Online and Remote Education (E-CORE): Quick Guide for Academic Integrity Remote Unproctored Exams. Other resources to support remote education can be found on the E-CORE website.
In studying academic misconduct, Mccabe, Trevino, and Butterfield  developed ten principles for faculty to ensure academic integrity (listed below). Note the significant overlap above, but also the importance of fostering trust with students.
- Affirm the importance of academic integrity
- Foster a love of learning
- Treat students as an end in themselves
- Foster and environment of trust in the classroom
- Encourage student responsibility for academic integrity
- Clarify expectations for students
- Develop fair and relevant forms of assessment
- Reduce opportunities to engage in academic dishonesty
- Challenge academic dishonesty when it occurs
- Help define and support campus-wide academic integrity standards
Cluskey Jr., Ehlen, and Raiborn  proposed eight principles to thwart online exam cheating without proctoring, which also show significant overlap above.
- Offer the exam at one set time
- Have the online exam open for students to sign in for a short time (they suggest 15 minutes)
- Randomize the sequence of exam questions
- Only present one question at a time
- Design the exam to occupy the time allowed (in their words, “The goal is for the ‘A’ and ‘B’ students to complete the exam with only a few minutes to spare. The ‘C’ or ‘D’ students may or may not complete the exam.”)
- Only allow students to access the exam one time.
- Use a lockdown browser for online exams.
- Instructors should change at least one-third of multiple choice / objective questions on each exam every term.
- Mccabe, Donald & Trevino, Linda & Butterfield, Kenneth. (2001). Cheating in Academic Institutions: A Decade of Research. Ethics & Behavior – ETHICS BEHAV. 11. 10.1207/S15327019EB1103_2.
- Jr, Cluskey, & Ehlen, Craig & Raiborn, Mitchell. (2011). Thwarting online exam cheating without proctor supervision. Journal of Academic and Business Ethics.
|Previous (A Misconduct Framework)||Next (Remote Proctoring Tools)|