Final Exams in bCourses: Strategies and Effective Practices

December 11, 2025

Many instructors find it valuable to deliver a final exam or end-of-semester assignment in bCourses. Delivering an exam in bCourses makes it easy for students to access the exam content and convenient for the instructors to grade at the end of the term. Yet some instructors worry about safeguarding academic integrity within bCourses, particularly since students can access Generative AI tools while they’re taking exams within bCourses. While Berkeley does not currently support any solutions that would block student access to external websites while students are in bCourses, there are some effective practices that instructors can implement that may mitigate academic dishonesty while facilitating online and/or remote exams:

Just-in-Time Strategies

  1. Remind students of the Berkeley Honor Code at the start of the exam. Consider including a first question that asks students to acknowledge that they’ve read the Honor Code. You might even encourage students to write out their own statement to the effect of: “I swear on my honor that I have neither given nor received aid on this exam.” Recent studies have found that students are less likely to use AI when they are reminded of academic integrity policies and the negative consequences of cheating if caught (see Zhao et. al 2023, Malesky et. al 2022, Tatum 2022).
  2. Remind students of your course AI (or your exam AI) policy. For example, the "No AI" policy language provided by the Academic Senate includes the statement, "If unauthorized AI use on a particular assessment is suspected, the instructor may require you to complete a short, in-person examination orally or on paper, related to the content and skills tested in the original assessment" or "A student who is suspected of violating the GenAI policy will be reported to the Center for Student Conduct and will be required to redo the assignment in person." Consider adding in additional language about the extent to which students should (or shouldn’t) consult with GenAI during their exam.
  3. Test Assignments with Generative AI: Before finalizing your assignments, use GenAI tools to attempt them. This can help to identify if the assignments can be easily completed by AI, potentially bypassing deep understanding or genuine student engagement. Learn more from Northern Michigan University’s guidance.
  4. Use a timed exam in bCourses. This feature allows instructors to create a maximum time limit in a bCourses exam, which may reduce student usage of GenAI to vet responses while taking an exam. If implementing this strategy, be mindful that some students may need to bypass it if they have accommodations allowing them for extended time on exams (see our Accommodations in bCourses Quick Guide).
  5.  Use “One Question at a Time” exams. In bCourses quiz settings, consider checking the "Show one question at a time" and "Lock questions after answering” options. This prevents students from opening the whole test and copying it. If they skip a question to look it up, they cannot come back to it.

Longer-Term Strategies

  1. Eliminate curved exam grading schemas. Grading on a curve can create significant anxiety for students, as students’ grades impact each other. When students feel anxious about their grades, such as in curve-based grading schemas, they are more likely to cheat (Gallant and Rettinger 2024, Krou et. al 2020, Lang 2013). Eliminating curve-based grading should mitigate students’ desires to use GenAI to get an advantage over their peers.

  2. Structure the exam to use assessment strategies that incorporate personal or course-specific relevance. Write questions that are closely and specifically connected to course content, focus more on students’ reasoning and decision-making and less on simple memorization of basic facts, or even consider an open book/open notes exam. For more examples, see our page on Redesigning Assignments and Assessments to be responsive to GenAI. 

  3. Administer an in-person proctored exam for maximum exam security. There is no one technical approach that will completely prevent the use of AI or retrieving answers from the web. For this reason, some instructors choose to administer the bCourses exam in the classroom so that students can be monitored as they complete the test.

  4. Consider administering written exams using Gradescope. Gradescope is an online grading tool that allows paper-based assignments to be scanned, submitted, graded, and analyzed quickly and efficiently. Set up a handwritten assignment or online assignment, scan and grade in Gradescope, then pass the grades back to bCourses.


References

Gallant, T. B., & Rettinger, D. A. (2025). The Opposite of Cheating: Teaching for Integrity in the Age of AI. University of Oklahoma Press.

Krou, M. R., Fong, C. J., & Hoff, M. A. (2021). Achievement Motivation and Academic Dishonesty: A Meta-Analytic Investigation. Educational Psychology Review, 33(2), 427–458. https://doi.org/10.1007/s10648-020-09557-7

Lang, J. M. (2013). Cheating lessons. Harvard University Press. 

Malesky, A., Grist, C., Poovey, K., & Dennis, N. (2022). The effects of peer influence, honor codes, and personality traits on cheating behavior in a university setting. Ethics & Behavior, 32(1), 12-21. 

Tatum, H. E. (2022). Honor Codes and Academic Integrity: Three Decades of Research. Journal of College and Character, 23(1), 32–47. https://doi.org/10.1080/2194587X.2021.2017977

Zhao, L., Peng, J., Yang, X., Yan, W., Ke, S., Dong, L. D., ... & Lee, K. (2023). Effects of honor code reminders on university students’ cheating in unproctored exams: A double-blind randomized controlled field study. Contemporary Educational Psychology, 75, 102213.