Cluster Analysis Methods and Future Time Perspective Groups of Second-Year Engineering Students in a Major-Required Course

Justine Chasmar; Katherine M. Ehlert

Download Paper | Permalink

Conference: 2018 ASEE Annual Conference & Exposition
Location: Salt Lake City, Utah
Publication Date: June 23, 2018
Start Date: June 23, 2018
End Date: July 27, 2018
Conference Session: Motivation, Attitudes, and Beliefs
Tagged Division: Educational Research and Methods
Page Count: 21
DOI: 10.18260/1-2--30195
Permanent URL: https://peer.asee.org/30195
Download Count: 715

Paper Authors

biography

Justine Chasmar Goucher College

visit author page

Justine Chasmar is an Assistant Professor in the Center for Data, Mathematical, and Computational Sciences and the Director of the Quantitative Reasoning Center at Goucher College. Her research focuses on tutoring, student learning, motivation, and professional identity development. Through her background in learning centers, she has applied this research to undergraduate students and peer tutors. Her education includes a B.S. and M.S. in Mathematical Sciences and Ph.D. in Engineering and Science Education from Clemson University.

visit author page

biography

Katherine M. Ehlert Clemson University

visit author page

Katherine M. Ehlert is a doctoral student in the Engineering and Science Education department in the College of Engineering, Computing, and Applied Sciences at Clemson University. She earned her BS in Mechanical Engineering from Case Western Reserve University and her MS in Mechanical Engineering focusing on Biomechanics from Cornell University. Prior to her enrollment at Clemson, Katherine worked as a Biomedical Engineering consultant in Philadelphia, PA. Her research interests include identity development through research experiences for engineering students, student pathways to engineering degree completion, and documenting the influence of co-op experiences on academic performance.

visit author page

Download Paper | Permalink

Abstract

The primary purpose of this research paper is to identify homogeneous groups of second-year engineering undergraduates in a major-required course in terms of their motivations and attitudes. Motivations and attitudes were evaluated using a survey that has been validated on other college-aged students. Specifically, we are utilizing Future Time Perspective (FTP), as our measure of motivation in this paper. FTP has been shown to have a connection to student strategies and how they approach learning in the present. FTP can be measured quantitatively with general, major, or task-specific instruments. However, it is often difficult for engineering educators to select appropriate quantitative analysis methods, as there is a lack of literature comparing types of analytic methods. Thus, the secondary purpose of this paper is to fill this gap by discussing multiple types of cluster analysis (CA) methods, selecting the best clustering method and solution, and using the clustering results to discuss domain- and context-specific differences between the FTPs of second-year undergraduate engineering majors. The research question for this paper is: What are the motivational (FTP) characterizations of second-year undergraduate engineering majors within the context of a major-required course? This study focuses on comparing cluster analysis techniques of data from a distribution of the Motivation and Attitudes in Engineering (MAE) survey, specifically the FTP Likert-type items from five domain- and context-specific factors. The survey was distributed in class and submitted online in two separate courses at a four-year, land grant institution in the southeast. Students were either enrolled in a Materials Science and Engineering or Industrial Engineering sophomore-level course during the same semester. These two sections were compared using Fisher’s Exact and Chi squared testing to ensure data could be aggregated during analysis. An exploratory factor analysis was conducted to assess the latent correlation structure of the survey items for validation. After survey validation, participants were put into homogenous groups utilizing two CA methods: Ward’s hierarchical and k-means. ANOVA and pairwise t-tests were run to look for significant differences between factor scores for each cluster within a CA method as well as across methods. Each method suggested that three clusters would be an ideal solution for the data, agreeing with theoretical predictions. The k-means cluster solution provided two clusters of similar size (Cluster 1 n=56, Cluster 2 n=58) and one larger cluster (Cluster 3 n=109). The Ward’s clustering solution provided three clusters of slightly different size (Cluster 1 n=86, Cluster 2 n=37, Cluster 3 n=100). The goodness of fit scores showed that k-means was the best fit for this set of data. Visually, both methods appeared to have achieved distinct clusters with strong clustering solutions. While Ward’s and k-means provided similar means and standard deviations for the three clusters, k-means provided a more robust and theoretically accurate measure, possibly because it allowed for movement of data between iterations. In future work, the MAE survey should be further studied for this population. Additionally, while this paper supports the creation of foundational methods papers for engineering education researchers, more literature in this area is needed.

Citation
Format

Chasmar, J., & Ehlert, K. M. (2018, June), Cluster Analysis Methods and Future Time Perspective Groups of Second-Year Engineering Students in a Major-Required Course Paper presented at 2018 ASEE Annual Conference & Exposition , Salt Lake City, Utah. 10.18260/1-2--30195

TY - CPAPER
AB - The primary purpose of this research paper is to identify homogeneous groups of second-year engineering undergraduates in a major-required course in terms of their motivations and attitudes. Motivations and attitudes were evaluated using a survey that has been validated on other college-aged students. Specifically, we are utilizing Future Time Perspective (FTP), as our measure of motivation in this paper. FTP has been shown to have a connection to student strategies and how they approach learning in the present. FTP can be measured quantitatively with general, major, or task-specific instruments. However, it is often difficult for engineering educators to select appropriate quantitative analysis methods, as there is a lack of literature comparing types of analytic methods. Thus, the secondary purpose of this paper is to fill this gap by discussing multiple types of cluster analysis (CA) methods, selecting the best clustering method and solution, and using the clustering results to discuss domain- and context-specific differences between the FTPs of second-year undergraduate engineering majors. The research question for this paper is: What are the motivational (FTP) characterizations of second-year undergraduate engineering majors within the context of a major-required course?
This study focuses on comparing cluster analysis techniques of data from a distribution of the Motivation and Attitudes in Engineering (MAE) survey, specifically the FTP Likert-type items from five domain- and context-specific factors. The survey was distributed in class and submitted online in two separate courses at a four-year, land grant institution in the southeast. Students were either enrolled in a Materials Science and Engineering or Industrial Engineering sophomore-level course during the same semester. These two sections were compared using Fisher’s Exact and Chi squared testing to ensure data could be aggregated during analysis. An exploratory factor analysis was conducted to assess the latent correlation structure of the survey items for validation. After survey validation, participants were put into homogenous groups utilizing two CA methods: Ward’s hierarchical and k-means. ANOVA and pairwise t-tests were run to look for significant differences between factor scores for each cluster within a CA method as well as across methods.
Each method suggested that three clusters would be an ideal solution for the data, agreeing with theoretical predictions. The k-means cluster solution provided two clusters of similar size (Cluster 1 n=56, Cluster 2 n=58) and one larger cluster (Cluster 3 n=109). The Ward’s clustering solution provided three clusters of slightly different size (Cluster 1 n=86, Cluster 2 n=37, Cluster 3 n=100). The goodness of fit scores showed that k-means was the best fit for this set of data. Visually, both methods appeared to have achieved distinct clusters with strong clustering solutions. While Ward’s and k-means provided similar means and standard deviations for the three clusters, k-means provided a more robust and theoretically accurate measure, possibly because it allowed for movement of data between iterations.
In future work, the MAE survey should be further studied for this population. Additionally, while this paper supports the creation of foundational methods papers for engineering education researchers, more literature in this area is needed.
AU - Justine Chasmar
AU - Katherine M. Ehlert
CY - Salt Lake City, Utah
DA - 2018/06/23
PB - ASEE Conferences
TI - Cluster Analysis Methods and Future Time Perspective Groups of Second-Year Engineering Students in a Major-Required Course
UR - https://peer.asee.org/30195
DO - 10.18260/1-2--30195
ER -