Generative AI-Enhanced Oral Exams in Economics Education
Open accessAbstract
†Georgia Gwinnett College, ‡University of Houston-Clear Lake, *SUNY Cortland Cheating on exams is a growing concern in higher education, especially with the rise of Generative AI (GenAI).This challenge is more pronounced in online/hybrid settings where traditional monitoring struggles to keep up with AI tools.We developed an innovative oral exam using GenAI in hybrid economics classes.A GenAI prompt was designed that incorporated course learning objectives and generated personalized questions that evaluated students' knowledge and provided real-time encouraging and formative feedback.Students collaborated in pairs using ChatGPT during Zoom sessions.The use of this prompt simulated a structured oral exam that elicited detailed responses, and reinforced comprehension, critical thinking, and application of key macroeconomic concepts.The approach promoted engagement and minimized opportunities for academic dishonesty, making it a practical alternative.
AI snapshot
Written from the paper’s full text.
TL;DR
The paper describes a GenAI-enhanced oral exam implemented in hybrid principles of macroeconomics classes, using a custom ChatGPT prompt to ask personalized questions in paired Zoom breakout rooms. It reports positive student perceptions, reduced cheating opportunities, and scalability potential, while noting the absence of comparison data and implementation challenges.
Key findings
- Students described the format as more engaging and less stressful than traditional exams, with quotes such as “a cool change,” “way more interactive than just filling in bubbles,” and “It felt more like a conversation than a test.”
- The GenAI prompt provided real-time encouraging and formative feedback, asked follow-up or similar questions after incorrect answers, and prompted real-world connections; survey responses and instructor observations showed students felt more motivated and preferred the approach over traditional tests.
- The paired Zoom breakout format added accountability and peer dialogue, reduced the temptation to let GenAI do all the work, and supported students’ sense of connection in an online/hybrid environment.
- AI-generated errors sometimes occurred; when students identified them, they became teachable moments about hallucinations and critical thinking, but without guidance students risked accepting incorrect information as accurate.
Methods
- Implemented in Fall 2024 and Spring 2025 across multiple sections of a hybrid, flipped-format principles of macroeconomics course; each section enrolled approximately 40 students and met once a week for 75 minutes, using the free version of ChatGPT for equitable participation.
- An initial one-on-one Zoom oral exam without GenAI was time-consuming and difficult to scale; the revised approach used a custom GenAI oral-examiner prompt aligned with course learning objectives, with students working in pairs in Zoom breakout rooms, instructor drop-ins, and a follow-up 10-question multiple-choice quiz on Gradescope.
- Assessment used the GenAI Teaching Assistant Blueprint for a grading prompt; no personally identifiable information was included in GenAI transcripts or prompts, pseudonyms were used, and the study was IRB approved. Analysis relied on informal course evaluations, survey responses, and instructor observations.
Limitations
- Stated: the study did not include a traditional or non-oral exam comparison group because the GenAI-enhanced oral exam was implemented across all course sections in the same semester; findings reflect student experiences rather than comparisons across formats.
- Stated: most pairs were self-selected, while only a few were instructor-assigned; because the number of instructor-assigned pairs was small, the authors cannot make meaningful statistical comparisons between group types.
- Stated: the approach may not work well for all learners, especially students with delayed processing disorders; non-native speakers may face additional challenges; AI-generated errors require well-crafted prompts and instructor oversight; prompt design and review take time, and faculty training is essential. Evident: the paper relies on informal evaluations and instructor observations, with no structured outcome data or statistical analysis reported.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
7 of 18 cited sources checked · 3 supported · 0 flagged · accuracy not yet measured
- Premisecitation
ChatGPT scores in the 90th percentile on standardized exams.
“In economics education, Geerling et al. (2023) demonstrate that ChatGPT scores in the 90th percentile on standardized exams, showing how easily it can produce high-quality academic work without genuine student engagement.”
Quote in the paperNot a substantial claim1 of 1 sources checked- cites Geerling et al. (2023) source
The Conclusion (p. 9) reports that ChatGPT ranked at the 91st percentile in microeconomics and the 99th percentile in macroeconomics on the standardized TUCE exams. This supports the claim of high exam performance, but “the 90th percentile” obscures the distinct results, and the study does not establish performance on standardized exams generally. (LLM, conf=0.85)
Experimental — accuracy not yet measured
- Premisecitation
Misuse of GenAI often stems from unclear guidance rather than intentional dishonesty.
“Scholars such as Bhullar, Joshi, & Chugh (2024), Cassidy (2023), Eke (2023), and Plata, De Guzman, & Quesada (2023) further argue that misuse often stems from unclear guidance rather than intentional dishonesty, underscoring the need for explicit institutional policies.”
Quote in the paperNot a substantial claim1 of 3 sources checked- cites Bhullar, Joshi, & Chugh (2024) source
Section 4.1 (Implications; page unavailable) calls for academic-integrity training and clearer institutional policies on appropriate and inappropriate AI use. Section 3.3.1 (Thematic clusters; page unavailable) likewise calls for clear boundaries. Neither passage says misuse often results from unclear guidance rather than intentional dishonesty. (LLM, conf=0.87)
Experimental — accuracy not yet measured - cites Cassidy (2023)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites Eke (2023) source
- Interpretationcitation
When GenAI is used in a structured, student-centered way, students perceive it as supporting engagement and deeper thinking about learning tasks.
“What we saw confirmed that when GenAI is used in a structured, student-centered way, students perceive it as supporting engagement and deeper thinking about learning tasks (Chan & Hu, 2023).”
Quote in the paperSubstantial claim1 of 1 sources checked- cites Chan & Hu, 2023 source
In “Personalized and immediate learning support” (p. 9), a student says GenAI feedback on homework could improve their depth of thinking and understanding; the same section reports that students considered GenAI capable of improving motivation. The paper does not establish that these perceptions depend on GenAI being used in a structured, student-centered way. (LLM, conf=0.84)
Experimental — accuracy not yet measured
- Findingnone given
The GenAI-enhanced oral exam format helped reduce opportunities for cheating.
“The format also helped reduce opportunities for cheating and showed potential for scalability in larger classes.”
Quote in the paperSubstantial claim- Rests on an assumption: Cheating on exams is a growing concern in higher education, especially with the rise of Generative AI.
- Findingown data
Students' evaluations praised the adaptive nature of the GenAI-enhanced oral exam experience.
“Further, the students' evaluations also praised the adaptive nature of the experience.”
Quote in the paperNot a substantial claim
- Findingreasoning
The identification of GenAI errors turned into valuable teachable moments about GenAI hallucinations and the need for critical thinking when using GenAI.
“The identification of GenAI errors turned into valuable teachable moments about GenAI hallucinations and the need for critical thinking when using GenAI.”
Quote in the paperNot a substantial claim - Study design and conductreasoning
The study did not include a traditional or non-oral exam comparison group.
“This study did not include a traditional or non-oral exam comparison group, as the GenAI-enhanced oral exam was implemented across all course sections in the same semester.”
Quote in the paperSubstantial claim - Interpretationreasoning
When students identify AI-generated errors, the process can deepen their knowledge.
“When students are able to identify these errors, the process can actually deepen their knowledge.”
Quote in the paperNot a substantial claim - Study design and conductreasoning
The GenAI-enhanced oral exam approach was aligned with FERPA and institutional privacy policies.
“This approach allowed us to experiment with GenAI in a way that was altogether innovative, respectful of student rights, and aligned with both the Family Educational Rights and Privacy Act (FERPA) and institutional privacy policies.”
Quote in the paperNot a substantial claim - Interpretationreasoning
The GenAI-enhanced oral exam format showed potential for scalability in larger classes.
“The format also helped reduce opportunities for cheating and showed potential for scalability in larger classes.”
Quote in the paperSubstantial claim - Interpretationown data
The GenAI-enhanced oral exam approach has proven effective in a hybrid setting.
“Now that the approach has proven effective in a hybrid setting, we are beginning to explore its adaptation for asynchronous online courses.”
Quote in the paperSubstantial claim - Study design and conductreasoning
Because the number of instructor-assigned pairs was small, we cannot make meaningful statistical comparisons between self-selected and instructor-assigned groups.
“Because the number of instructor-assigned pairs was small, we cannot make meaningful statistical comparisons between the two types of groups.”
Quote in the paperNot a substantial claim - Premisecitation
Current plagiarism detectors often fail to identify AI-generated content.
“Wiley (2024) and Susnjak and McIntosh (2024) add that current plagiarism detectors often fail to identify AIgenerated content, signaling an urgent need to redesign assessments for the realities of digital learning.”
Quote in the paperSubstantial claim0 of 2 sources checked- cites Wiley (2024)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites Susnjak and McIntosh (2024)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Interpretationreasoning
The GenAI-enhanced oral exam offers a more engaging and pedagogically meaningful alternative for smaller or moderately sized classes.
“Similarly, the oral exam also presents scalability challenges, but a GenAI-enhanced version offers a more engaging and pedagogically meaningful alternative for smaller or moderately sized classes.”
Quote in the paperNot a substantial claim- Rests on an assumption: Oral exams are a powerful way to truly understand what students know, more so than traditional written tests.
- Study design and conductreasoning
The findings reflect student experiences using GenAI-enhanced oral exams rather than comparisons across formats.
“The findings therefore reflect student experiences using GenAI-enhanced oral exams rather than comparisons across formats.”
Quote in the paperSubstantial claim
- Classified as probably not a substantial claim. 18
- May combine two claims. 32
- The quote only partly states this claim. 1
Probably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationAI-generated content challenges traditional definitions of authorship and originality.
“As Cotton, Cotton, & Shipway (2024) note, AI-generated content challenges traditional definitions of authorship and originality, a concern echoed by Perkins (2023).”
Quote in the paperNot a substantial claim1 of 2 sources checkedClassified as probably not a substantial claim. May combine two claims.
PremisecitationDependence on GenAI undermines critical thinking and reduces opportunities for deep learning.
“Bhullar, Joshi, & Chugh (2024) warn that such dependence undermines critical thinking and reduces opportunities for deep learning.”
Quote in the paperSubstantial claim1 of 1 sources checked- cites Bhullar, Joshi, & Chugh (2024) source
Section 3.3.2 says ChatGPT could reduce students’ motivation to explore topics and draw conclusions independently. Section 3.4 warns of possible learning impediments and stresses the importance of critical thinking. These passages support a concern about learning, but they do not establish that dependence undermines critical thinking or specifically reduces opportunities for deep learning. (LLM, conf=0.82)
Experimental — accuracy not yet measured
May combine two claims.
PremisecitationWhen students talk through their answers with a peer, they think more carefully about their reasoning, making assessment more collaborative and learning oriented.
“Prior research on peer instruction shows that when students can talk through their answers with a peer, they think more carefully about their reasoning, and this makes the assessment more collaborative and learning oriented (Tullis and Goldstone, 2020).”
Quote in the paperSubstantial claim1 of 1 sources checked- cites Tullis and Goldstone, 2020 source
May combine two claims.
PremisecitationStudents who rely on tools like ChatGPT for support may experience reduced social interaction, lower wellbeing, and a weaker sense of belonging.
“Crawford et al. (2024) found that students who rely on tools like ChatGPT for support may experience reduced social interaction, lower wellbeing, and a weaker sense of belonging.”
Quote in the paperSubstantial claim1 of 1 sources checked- cites Crawford et al. (2024) source
Probably not a substantial claim: may not pass S3 (load-bearing).
Findingnone givenAI-generated responses to in-class prompts could sometimes contain errors.
“One common challenge is that AI-generated responses to in-class prompts could sometimes contain errors.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
FindingreasoningWorking in pairs encouraged collaboration and reduced the temptation to let GenAI do all the work.
“Working in pairs encouraged collaboration and reduced the temptation to let GenAI do all the work.”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown dataMost pairs in the GenAI-enhanced oral exams were self-selected, while only a few were assigned by the instructor.
“Most pairs were self-selected, while only a few were assigned by the instructor.”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown dataSurvey responses and instructor observations showed that students felt more motivated and preferred the GenAI-enhanced oral exam over traditional tests.
“Survey responses and instructor observations showed that students felt more motivated and preferred this new approach over traditional tests.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataStudents appreciated the conversational tone and opportunity to think aloud in the GenAI-enhanced oral exams.
“They appreciated the conversational tone and opportunity to think aloud.”
Quote in the paperNot a substantial claimMay combine two claims.
FindingreasoningInstructor informal check-ins helped keep students on task, and the group format added an extra layer of accountability.
“These informal check-ins helped keep students on task, and the group format itself added an extra layer of accountability.”
Quote in the paperNot a substantial claimMay combine two claims.
FindingreasoningIn the first round of one-on-one Zoom oral exams without GenAI, the format was time-consuming and difficult to scale.
“The format was timeconsuming and difficult to scale.”
- Also stated in: § Initial Exam Structure · plain
“Even with a manageable class size of 40 students, it was challenging to reach everyone, let alone imagine doing so in larger classes.”
Quote in the paperSubstantial claim- Also stated in: § Initial Exam Structure · plain
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingreasoningMany students offered more thoughtful responses because they were bouncing ideas off a peer in real time.
“In fact, many students offered more thoughtful responses simply because they were bouncing ideas off a peer in real time.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Findingown dataWe did not encounter language-related issues in our implementation of GenAI-enhanced oral exams.
“While we did not encounter language-related issues in our implementation, we acknowledge that oral exams may pose additional challenges for non-native speakers.”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown dataStudents in the GenAI-enhanced oral exams experienced a more engaging and interactive format, with questions tailored to their understanding and real-time feedback that encouraged deeper thinking.
“In this study, students experienced a more engaging and interactive format, with questions tailored to their understanding and real-time feedback that encouraged deeper thinking.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
Findingown dataInformal observations suggested that self-selected partners tended to collaborate more smoothly due to prior rapport, whereas instructor-assigned pairs sometimes required additional guidance to work effectively together.
“Informal observations suggested that self-selected partners tended to collaborate more smoothly due to prior rapport, whereas instructor-assigned pairs sometimes required additional guidance to work effectively together.”
Quote in the paperNot a substantial claimMay combine two claims.
FindingreasoningThe GenAI-enhanced oral exam format helped students feel more at ease, boosted their confidence, and led to thoughtful conversations.
“This format helped students feel more at ease, boosted their confidence, and led to some genuinely thoughtful conversations.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
FindingreasoningIn one case, a student was granted extended time and allowed to complete the exam individually outside of class.
“In one case, the student was granted extended time and allowed to complete the exam individually outside of class.”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown dataIn informal course evaluations, students described the GenAI-enhanced oral exam format as more engaging and less stressful than traditional exams.
“In informal course evaluations for the principles of macroeconomics course, students described the format as more engaging and less stressful than traditional exams.”
Quote in the paperNot a substantial claimMay combine two claims.
FindingreasoningIn the first round of oral exams, not all students were able to finish within the scheduled class time, requiring extra sessions outside of class.
“Not all students were able to finish the exam within the scheduled class time, which meant extra sessions had to be arranged outside of class.”
Quote in the paperSubstantial claimMay combine two claims.
PremisecitationGenAI tools are becoming easier to use and can produce accurate and detailed human-like responses to academic assignments.
“According to Lo (2023), these GenAI tools are becoming easier to use and can produce accurate and detailed human-like responses to academic assignments”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Lo (2023)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Probably not a substantial claim: may not pass S2 (checkable), S4 (based).
InterpretationreasoningThe GenAI-enhanced oral exam approach offers a practical model for faculty interested in integrating GenAI tools like ChatGPT into oral assessments.
“This approach offers a practical model for faculty interested in integrating GenAI tools like ChatGPT into oral assessments.”
Quote in the paperNot a substantial claimMay combine two claims.
PremisecitationRemote settings weaken student-instructor connections and make academic dishonesty easier for students to rationalize.
“Studies by Swartz and Cole (2013) and Agha, Zhu, & Chikwa (2022) show that remote settings weaken student-instructor connections and make academic dishonesty easier for students to rationalize.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites Swartz and Cole (2013)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites Agha, Zhu, & Chikwa (2022) source
May combine two claims.
InterpretationreasoningBy guiding students through personalized, interactive questions, the GenAI-enhanced oral exam approach makes assessments more engaging while helping protect academic integrity.
“By guiding students through personalized, interactive questions, this approach makes assessments more engaging while helping protect academic integrity.”
Quote in the paperSubstantial claim- Rests on an assumption: Oral exams are a powerful way to truly understand what students know, more so than traditional written tests.
Probably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationMany college students enjoy using GenAI as part of their academic experience.
“A recent survey confirms that many college students enjoy using GenAI as part of their academic experience (Mowreader, 2025).”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Mowreader, 2025
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
May combine two claims.
Study design and conductnone givenNo personally identifiable information was included in GenAI transcripts or prompts, and pseudonyms were used to ensure confidentiality.
“No personally identifiable information was included in GenAI transcripts or prompts, and we used pseudonyms to ensure confidentiality.”
Quote in the paperNot a substantial claimMay combine two claims.
InterpretationreasoningGenAI-enhanced exams can support deeper learning and uphold academic integrity, particularly in hybrid and online courses.
“By fostering more interactive, personalized, and ethical evaluations, GenAI-enhanced exams can support deeper learning and uphold academic integrity, particularly in hybrid and online courses.”
Quote in the paperSubstantial claim- Rests on an assumption: Oral exams are a powerful way to truly understand what students know, more so than traditional written tests.
The quote only partly states this claim.
InterpretationreasoningThe GenAI-enhanced oral exam approach may not work well for all learners, especially those with delayed processing disorders.
“Despite these benefits, this approach may not work well for all learners, especially those with diverse needs.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable). May combine two claims.
Interpretationown dataWhen used thoughtfully, GenAI can strengthen the learning experience, help students think more deeply, stay more engaged, and feel more connected to the material without taking the place of meaningful instructor involvement.
“When used thoughtfully, GenAI can strengthen the learning experience, help students think more deeply, stay more engaged, and feel more connected to the material without taking the place of meaningful instructor involvement.”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim. May combine two claims.
InterpretationreasoningThe GenAI-enhanced oral exam approach provides a scalable alternative that still draws on the instructor's expertise and judgment.
“While the approach is not without challenges, it provides a scalable alternative that still draws on the instructor's expertise and judgment.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim.
InterpretationreasoningFor students with delayed processing disorders, the fast-paced, conversational format can create added pressure, making it difficult to organize thoughts or respond confidently without additional time or accommodations.
“For students with delayed processing disorders, this fast-paced, conversational format can create added pressure, making it difficult for them to organize their thoughts or respond confidently without additional time or other accommodations.”
Quote in the paperSubstantial claimMay combine two claims.
InterpretationreasoningIn a hybrid or online environment, the GenAI-enhanced oral exam approach creates a more connected and authentic learning experience while maintaining academic rigor and reducing the temptation to rely on shortcuts.
“In a hybrid or online environment, where it can be easy to disengage or feel isolated, this approach creates a more connected and authentic learning experience, while maintaining academic rigor and reducing the temptation to rely on shortcuts.”
Quote in the paperSubstantial claim- Rests on an assumption: Oral exams are a powerful way to truly understand what students know, more so than traditional written tests.
Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based). May combine two claims.
Interpretationnone givenThe GenAI-enhanced oral exam approach is flexible, and the template can be adapted for other disciplines and course formats.
“The approach is flexible, and the template can be adapted for other disciplines and course formats.”
Quote in the paperNot a substantial claimMay combine two claims.
InterpretationreasoningThe GenAI prompt simulated a structured oral exam that elicited detailed responses and reinforced comprehension, critical thinking, and application of key macroeconomic concepts.
“The use of this prompt simulated a structured oral exam that elicited detailed responses, and reinforced comprehension, critical thinking, and application of key macroeconomic concepts.”
Quote in the paperSubstantial claimMay combine two claims.
InterpretationreasoningUtilizing breakout rooms for student pairs to complete the oral exams fostered collaboration and peer-to-peer dialogue, supporting students' sense of connection within an online environment.
“Utilizing breakout rooms for student pairs to complete the oral exams, fostered collaboration and peer-to-peer dialogue.This atmosphere supported students' sense of connection within an online environment.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Interpretationnone givenOral exams may pose additional challenges for non-native speakers.
“While we did not encounter language-related issues in our implementation, we acknowledge that oral exams may pose additional challenges for non-native speakers.”
Quote in the paperNot a substantial claimMay combine two claims.
PremisecitationTraditional proctored exams can help maintain oversight and uphold academic integrity but are often costly and logistically demanding.
“While traditional proctored exams, in-person or remote, can help maintain oversight and uphold academic integrity, they are often costly and logistically demanding (Dadashzadeh, 2021).”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Dadashzadeh, 2021
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Probably not a substantial claim: may not pass S3 (load-bearing).
Study design and conductnone givenThe GenAI-enhanced oral exam was implemented by one author across multiple sections of a hybrid-format principles of macroeconomics course.
“The GenAI-enhanced oral exam was implemented by one author across multiple sections of a hybrid-format principles of macroeconomics course.”
Quote in the paperNot a substantial claimMay combine two claims.
InterpretationreasoningThe GenAI-enhanced oral exam approach promoted engagement and minimized opportunities for academic dishonesty, making it a practical alternative.
“The approach promoted engagement and minimized opportunities for academic dishonesty, making it a practical alternative.”
Quote in the paperSubstantial claim- Rests on an assumption: Cheating on exams is a growing concern in higher education, especially with the rise of Generative AI.
Probably not a substantial claim: may not pass S3 (load-bearing).
Interpretationown dataA pilot of a related GenAI tutor-style prompt in an asynchronous Health Economics course demonstrated the potential to extend GenAI-assisted assessment into flexible learning environments.
“While this setting did not include the same oral exam format, it demonstrated the potential to extend GenAI-assisted assessment into flexible learning environments.”
Quote in the paperNot a substantial claimMay combine two claims.
Study design and conductreasoningAll students used the free version of ChatGPT, and no subscription-based features were required.
“To ensure equitable participation, all students used the free version of ChatGPT.”
- Also stated in: § 3. AI-enhanced Oral Exams -Design and Implementation · plain
“No subscription-based features were required, permitting every student to participate regardless of financial limitations.”
Quote in the paperSubstantial claim- Also stated in: § 3. AI-enhanced Oral Exams -Design and Implementation · plain
May combine two claims.
InterpretationreasoningBy automating parts of the assessment process, GenAI-enhanced oral exams help streamline assessment for instructors, allowing them to focus more on in-class activities and personalized support.
“By automating parts of the assessment process, GenAI-enhanced oral exams help streamline assessment for instructors, allowing them to focus more on in-class activities and personalized support.”
Quote in the paperNot a substantial claimMay combine two claims.
PremisecitationIncorporating real-time elements into assessments can safeguard against unauthorized GenAI use and ensure a more accurate reflection of each student's understanding.
“Incorporating real-time elements into assessments in the form of oral responses or monitored discussions can safeguard against unauthorized GenAI use and ensure a more accurate reflection of each student's understanding (Cotton, Cotton, & Shipway 2024).”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Cotton, Cotton, & Shipway 2024 source
Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.
Interpretationnone givenDesigning GenAI prompts requires time, as does reviewing students' GenAI interactions to ensure the quality of student responses and alignment with course learning outcomes.
“Designing GenAI prompts requires time, as does reviewing students GenAI interactions to ensure the quality of student responses and alignment with course learning outcomes.”
Quote in the paperNot a substantial claim
The paper relies on this without support.
Cheating on exams is a growing concern in higher education, especially with the rise of Generative AI.
“Cheating on exams is a growing concern in higher education, especially with the rise of Generative AI (GenAI).”
The paper relies on this without support.
Oral exams are a powerful way to truly understand what students know, more so than traditional written tests.
“Oral exams are known as a powerful way to truly understand what students know, much more so than traditional written tests.”
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 3
- Dadashzadeh, M. (2021). The Online Examination Dilemma: To Proctor or Not to Proctor? Journal of Instructional Pedagogies, 25.
- Swartz, L. B., & Cole, M. T. (2013). Students' perception of academic integrity in online business education courses. Journal of Business & Educational Leadership, 4(1).
- Wiley. (2024). The latest insights into academic integrity: Instructor & student experiences, attitudes, and the impact of AI. https://www.wileyplus.com/platforms/academic-integrity/ 1. "Can you describe the 'Consumption' component of GDP and give examples?" 2. "Could you explain what counts as 'Net Export' in GDP?" 3. "What do we mean by 'Government Spending' as a component of GDP?" 4. "If Michelin sells tires to Toyota for a new Camry built in the U.S., does this count toward GDP? If so, under which component?" 5. "Imagine you help your neighbor babysit her children on weekends. Does this count toward GDP? Why or why not?"
- Authors
- Onodipe, G., Ayadi, M. F., & Burke, K.
- Affiliations
- Georgia Gwinnett College (US)
- University of Houston - Clear Lake (US)
- SUNY Cortland (US)
- Venue
- Journal of Economics Teaching
- Publisher
- Journal of Economics Teaching
- Published
- 2026
- Pages
- 229-240
- ISSN
- 2690-506X (ISSN-L)
- OpenAlex
- W7131291651
- Semantic Scholar
- 75b0bd4a562dcccba19d38dfa46ad7f794b5e17f
- Cited by
- 0
- References
- 0
- FWCI
- 0.00
- Type
- journal article
- Open access
- diamond
- doi.org
- Licence
- cc-by
- Locations
- Journal of Economics Teaching (publishedVersion, cc-by)
- Topics
- Physical Sciences › Computer Science › Information Systems › Artificial Intelligence in Education
- Social Sciences › Safety Research › Academic integrity and plagiarism
- Physical Sciences › Computer Science › Information Systems › Educational Technology and Assessment
- Keywords
- generative AI, oral examination, economics education, academic dishonesty, ChatGPT, hybrid learning, online learning, formative feedback, macroeconomics, critical thinking, student engagement
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-23.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-23.
Cited sources are checked by gpt-6-sol.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review