Shared understanding in engineering design capstone teams

This study explores how Latent Semantic Analysis (LSA) can be used to understand the shared content of open-ended reflections (RQ1) as a means of identifying the typical elements of a collaborative design team’s shared understanding of a design problem (RQ2). Undergraduate student design teams of 5-6 were recruited from a capstone design course in a Mechanical Engineering department at a large midwestern university in Spring 2022. Eight teams were selected for further study based on response rates. All students were asked to submit reflections throughout the semester, with one prompt repeating three times. Additionally, a survey was conducted to collect demographic information and additional context about the students’ perceptions of their project. An emergent coding scheme was used to label utterances from responses to the repeated reflection prompt according to the type of content contained within each utterance (n=664). Utterances were also processed using LSA to approximate each utterance (n=5162) as a vector in the k-reduced semantic space. The sharedness of each utterance was computed as the cosine similarity between the vector representing the utterance and a composite vector representing the sum of utterances from teammates in response to the same reflection prompt. The most common type of utterance found in design team reflections was Design Considerations. Design Considerations (U=61337, p<0.001) and Project Descriptions (U=61337, p<0.001) were found to have higher sharedness than the overall group of utterances using the Mann-Whitney U-Test. Process (U=21128, p=0.010), Perspective (U=42216, p<0.001) and Obstacles (U=12961, p<0.001) were found to have lower sharedness than the overall group of utterances using the Mann-Whitney U-Test. Three descriptive cases were explored representing the approximate middle, lowest, and highest sharedness averages. Taken together, the descriptive cases demonstrated that sharedness varies dynamically over a project timeline, influenced by factors such as project clarity and team member specialization. Teams emphasized the importance of external representations such as block diagrams or rough prototypes in managing a shared understanding, not just among team members but also with stakeholders. The Team-Task-Design Mental Model Framework was used to examine these results, finding that utterances with high sharedness represented aspects of design-related mental models, whereas lower sharedness utterances represented aspects of task-related mental models.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Mechanical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2024

Author dc:creator

• Pagano, Alexander

Contributors dc:contributor

• Goldstein, Molly H.
• Jacobi, Anthony
• Mercier, Emma
• West, Matthew

• Engineering Education
• Design
• Engineering Design
• Latent Semantic Analysis
• Shared Mental Model
• Collaboration
• Teamwork

Statement dc:rights

• Copyright 2024 Alexander Pagano

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/125595

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F125595