Reconsidering the Use of Oral Exams and Assessments: An Old Way to Move Into a New Future

Synthesis: Fenton (2025) makes the case that oral exams and assessments deserve renewed attention as an authentic assessment strategy in the face of generative AI. Arguing that the real-time, interactive nature of an oral assessment prevents students from using AI to generate and memorize answers, the article reviews the history, types, benefits, challenges, and practical recommendations for oral assessments across high school and higher education. Key benefits include personalization, authenticity, work-readiness, deeper knowledge, and the preservation of Academic Integrity; key challenges include scheduling, anxiety, and potential bias.

Key Findings

• Oral assessments reduce AI-related academic integrity issues because the real-time, interactive format prevents students from using AI to generate answers and memorize them ahead of time, particularly when questions are unknown in advance (drawing on Nikolic et al., 2023).
• Oral assessments are authentic — they mirror professional practice (job interviews, workplace challenges, clinical/patient communication), build communication skills, confidence, and deeper understanding, and prepare students for careers where verbal communication is vital.
• They probe depth over memorization: as a dialogue, oral assessments let educators test comprehension, critical thinking, Problem Solving, and reasoning — reaching the highest levels of Bloom's taxonomy — rather than rote recall.
• Benefits include personalization, authenticity, work-readiness, deeper knowledge, enjoyment, and higher grades; oral assessments can also support struggling students and shift study habits away from memorization toward understanding.
• Challenges include scheduling/technical logistics, student anxiety, and potential bias (gender, ethnicity, language, speed of answering, and non-anonymous marking), though evidence suggests oral assessments can be as inclusive as written exams (e.g., accommodating dyslexia).

Study Design & Method

This is a review/essay (not an empirical study) in Educational Researcher. Fenton synthesizes the literature on oral exams and assessments, covering their history (from ancient Greek/Socratic practices through the "viva voce" tradition at Oxford and Cambridge), Joughin's (1998) typology of oral assessments (six dimensions: primary content type, interaction, authenticity, structure, examiners, orality), and evidence from disciplines including business, medicine, engineering, marketing, chemistry, and biology. The article distills benefits and challenges from the research and offers practical design considerations and 13 numbered recommendations for implementation.

What this means for practice

• Instructors. Publish the schedule, criteria, and question style in advance and give students a practice run at the format — unfamiliarity, not the format itself, drives much of the reported anxiety, and students were less anxious in later oral assessments once they knew what to expect (Roecker, 2007).
• Instructors. Weight the assessment toward explaining and reasoning rather than recall: because the format is a dialogue, it tests comprehension and lets students demonstrate problem-solving and communication skills that written exams with static answers miss (Gaudet, 2015; Sayre, 2014).
• Instructors. Add oral assessment alongside AI Detection rather than instead of it: it is a low-tech, human-centered complement that makes unauthorized AI use impractical because answers are constructed live.
• Assessment designers. Build a clear rubric keyed to levels of mastery for the skill being assessed, plus a bank of questions linked to each learning objective, so every student faces a comparable task, and consider online delivery, which cuts travel time and cost for students and educators and keeps recordings available for reviewing fail grades (Starr-Egger, 2023).
• Assessment designers. Train assessors on a shared prompting taxonomy (presenting the task, repeating information, clarifying, probing, leading) and on reflexivity, then prepare carefully for fairness: gender, ethnicity, language, speed of answering, and non-anonymous marking are all potential bias sources, while the format can still accommodate students with dyslexia (Akimov & Malin, 2020).

Limitations

• As a review/essay, it does not provide new empirical evidence: it is a position/review piece published in a major research journal (Educational Researcher) rather than a study presenting original data.
• The author notes that more research is needed on how students and educators actually experience oral assessments in the current AI environment, and on more practical implementation guidance.

Citation

Fenton, A. (2025). Reconsidering the use of oral exams and assessments: An old way to move into a new future. Educational Researcher, 54(7), 430–436.