WIP: How do visual representations affect how engineering students learn and solve problems within and across disciplines?

This work-in-progress paper describes the preliminary synthesis of two data sets that explore how students solve engineering problems and access information from visual representations. This synthesis is a first step toward building theory about how students use visual representations across engineering disciplines.

By graduation, engineering students are expected to solve problems and communicate engineering ideas with several types of visual representations (e.g., free-body diagrams, graphs, or schematics). This skill requires students to identify conceptually-relevant features from visual representations and translate them into other useful forms to solve problems. Results from the cognitive science and physics education research literatures suggest that certain features of visual representations and students’ level of domain knowledge determine how effectively students access information from these representations. These two bodies of literature model knowledge acquisition and problem solving as generalized functions that are independent of the discipline. However, engineering concepts are visually represented differently depending on conventions within a discipline. These differences suggest that problem solving could also be discipline dependent. Thus, we should consider conventions of a discipline when redesigning visual representations for pedagogical interventions.

Our goal through this study is to develop theory that explores similarities and differences in students’ knowledge acquisition and problem-solving strategies using visual representations across different engineering disciplines. Our research question for this work in progress is how do certain features in a visual representation affect the types of mistakes students make during problem solving?

Our initial theory is built from the synthesis of results obtained from two previous qualitative studies that analyzed engineering students’ problem-solving behavior in Statics and Digital Logic using think-aloud interviews. We use the Constant Comparative Method to synthesize these data sets to determine whether previously identified themes and patterns are discipline specific or whether they transcend disciplinary boundaries.

In this work-in-progress paper, we present one emergent theme, “Informationally incomplete representations leads to coordinating multiple representations.” This theme describes how the presence of representations that do not contain enough information to translate to another representation result in students coordinating multiple representations during their problem solving. Results from our study can be used to better inform classroom interventions within the discipline-based education research community.