Multimodal Large Language Models (MLLMs) in Educational Research: Analyzing Text, Audio, and Visual Data
Year: 2026
Type: book-chapter
Abstract: Artificial intelligence applications have had a significant impact on the conduct of educational research in recent years.Researchers have moved beyond simply collecting and analyzing data and are now using these applications to support data acquisition, coding, analysis, and hypothesis generation.This offers significant advantages, such as the rapid analysis of large datasets (thousands of student responses, lengthy forum discussions, etc.) that were previously very difficult to obtain, thus expanding the scope of research.However, the use of these applications also raises debates about data reliability and the researcher's role in the studies.As a result, artificial intelligence applications have gone beyond being merely a tool offering new possibilities in educational research; they have become a factor requiring a re-evaluation of our research methods and ethical understanding.Initially, LLMs (Large Language Models) were text-focused applications whose inputs and outputs consisted solely of written expressions.While these models were quite effective for tasks such as grading student assignments and interpreting survey data, they could not process visual or auditory data directly.The transition to MLLMs (Multimodal Large Language Models) largely overcame this limitation.These models can process text while simultaneously understanding and interpreting data such as photographs, screenshots, tables, short videos, and audio recordings within the same context.Thus, for example, it became possible to analyze a lecture video, its slides, and notes in a single system.This diversified the types of data educational research focuses on and brought many previously overlooked learning indicators into the analysis.This chapter examines the possibilities offered by MLLMs, such as Gemini and Claude, in educational research and how they influence research processes.
Author: Erhan Yokuş
Institution: Gaziantep University
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: gold
OpenAlex ID: W7223096251