Development of an Intelligent Tutoring System That Assesses Internal Visualization Skills in Engineering Using Multimodal Triangulation

Authors:
Publisher website
Google Scholar
Abstract
In many STEM domains, instruction on foundational concepts heavily relies on visuals. Instructors often assume that students can mentally visualize concepts, but students often struggle with internal visualization skills—the ability to mentally visualize information. In order to address this issue, we developed a formal as well as an informal assessment of students’ internal visualization skills in the context of engineering instruction. To validate the assessments, we used data triangulation methods. We drew on data from two separate studies conducted in a small-scale lab experiment and in a larger-scale classroom context. Our studies demonstrate that an intelligent tutoring system with interactive visual representations can serve as an informal assessment of students’ internal visualization skills, predicting their performance on a formal assessment of these skills. Our study enriches methodological and theoretical underpinnings in educational research and practices in multiple ways: it contributes to (1) research methodologies by illustrating how multimodal triangulation can be used for test development, (2) theories of learning by offering pathways to assessing internal visualization skills that are not directly observable, and (3) instructional practices in STEM education by enabling instructors to determine when and where they should provide additional scaffoldings.
Funding Information
  • NSF (DUE 1933078)

This publication has 39 references indexed in Scilit: