Dental and medical students’ self-directed learning and motivation - An evaluation of two multiple-choice questions systems using machine learning

Aug 3, 2026

This comparative case study investigates how dental and medical students evaluate multiple-choice questions (MCQs) as a learning tool, applying machine learning to text responses. The study focuses on self-directed learning and motivation and evaluates two Aarhus University systems: MEDMCQ for medical students and MCQ anatomy for dental students. SurveyXact gathers responses from 126 medical students and 70 dental students, and topic modeling identifies two student groups with distinct motivational and learning-facilitating experiences.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages27

Reading Time1 h 8 min

AccessFree to download

Introduction

MCQs as assessment formats in health science education

Debates on learning depth vs objective level

Applications: higher-order thinking and clinical vignettes

MCQs for formative, self-paced, and adaptive assessment

Tidsskriftet Læring og Medier Dental and medical students’ selfdirected learning and motivation: An evaluation of two multiple-choice questions systems using machine learning Emilie Leth Rasmussen, Centre for Educational Development, Aarhus University Malthe Have Musaeus, It-University, Denmark Mads Ronald Dahl, Centre for Educational Development, Aarhus University Henrik Løvschall, Department for Dentistry and Oral Health, Aarhus University Peter Musaeus, Centre for Educational Development, Aarhus University Dansk abstrakt Dette sammenlignende casestudie beskriver en undersøgelse, der ved hjælp af maskinlæring undersøger studerendes evaluering af multiple-choice spørgsmål (MCQ) som middel til læring. Med et fokus på selvstyret læring og motivation undersøger vi to systemer udviklet på Aarhus Universitet: "MED MCQ", der bruges af medicinstuderende, og "MCQ-anatomi", der bruges af tandlægestuderende. Med svar fra 126 medicinstuderende og 70 tandlægestuderende evaluerer vito spørgeskemaer via SurveyXact. Ved anvendelse af topic modeling identificerer maskinlæringsmodellen to grupper af studerende, som på forskellig vis oplever systemet som motiverende og fremmende for deres læringsproces. De studerende oplever en øget selvstyretlæring ved selv at have indflydelse på præsentation af spørgsmål og muligheden for at besvarespørgsmål uafhængigt af en underviser. Vi diskuterer, hvordan undervisere og udviklere kan bruge MCQ til at fremme studerendes læring. Artiklen diskuterer, hvordan man kan bruge maskinlæring til at analysere tekst-svar fra åbne spørgsmål. Engelsk abstract This comparative case study reports a study investigating student evaluation of Multiple-Choice questions (MCQ) through machine learning as a means of learning. The focus is on self-directed learning and motivation. The study evaluates two systems developed at Aarhus University: "MEDMCQ" used by medical students, and "MCQ anatomy" used by dental students. The study evaluates two surveys in SurveyXact with responses from 126 medical students and 70 dental students. We use topic modeling over free text responses. The machine learning model identifies two groups of students who, in different ways, experience interacting with the system as motivating and facilitating their learning process. The students' experience increases self-directed learning by being able to choose the form of presentation of questions and answer questions independently of the instructor. The article discusses how educators and developers can use MCQs to promote student learning and how to analyze open-ended questions with machine learning. Tidsskriftet Læring & Medier (LOM)– nr. 29-2023 ISSN: 1903-248X Tidsskriftet Læring og Medier (LOM), Nr. 29, 2023 ISSN: 1903-248X Introduction Multiple-Choice Questions (MCQs) represent a widely utilized assessment format characterized by a student selecting one or more answers from a predetermined list, as thoughtfully curated and presented by the test designer (Downing & Haladyna, 2004). Student learning can be facilitated in-class and outof-class with MCQs and other written test questions (McDermott et al., 2014; Billings et al., 2021) . Teachers and students evaluate MCQs positively, and research has demonstrated that they are a valid, economical, and effective means of summative assessment (Downing & Haladyna, 2004) and formative assessment, i.e., assessment for learning, perhaps because of their ability to cover a wide range of content in a relatively short period (Einig, 2013; Javaeed, 2018) . However, there are conflicting viewson whether MCQs mainly assess lower learning objectives or shallow learning. MCQs and other formats of written test questions (Billings et al., 2021) have been extensively described in the context of health science education. Chandra et al. (2018) found that MCQs are a time-efficient assessment format displaying high levels of reliability in both undergraduate and postgraduate medical education. Some

FAQs

What is the main purpose of the study?

To evaluate how dental and medical students judge MCQs for learning using machine learning analysis.

Which two MCQ systems are compared in the study?

MEDMCQ for medical students and MCQ anatomy for dental students.

How are students’ open-text answers analyzed?

Topic modeling is applied to free-text survey responses to identify groups of students with different experiences.

Aug 3, 2026

Upload by Evangeline