The Development of a Computerized Interactive Teaching Assistant in Physics: The CITA on CHIP Project

Cyrus M Vandrevala, Purdue University

Abstract

One of the many roles of university instructors is to provide help to students throughout the semester - especially in the form of feedback on homework. Personalized feedback from the instructor might be possible in a small classroom setting, but becomes unmanageable when class sizes grow to dozens or even hundreds of students. As a result, universities are turning to computerized homework systems that guide students through problems and provide focused grades and feedback. The overarching goal of this project was to design a comprehensive, scaffolded set of tutorials for the homework questions in an undergraduate electricity and magnetism course for engineering majors and determine how the tutorials affected student performance. The research group developed CITA (Computerized Interactive Teaching Assistant) using a design-based research approach. This system assessed student knowledge and provided focused feedback using three structures - Shallow CITA, Immersive CITA, and Postscripts. Throughout the semester, I collected student grade data as well as student opinions of the CITA system through surveys and focus group interviews. The analysis of data informed an iterative development process. This study aimed to answer a few research questions. First, does the use of CITA tutorials improve student performance in their introductory electricity and magnetism course? Second, what are students’ views of the CITA program? Finally, what motivates students to choose CITA over some other external educational resource? Overall, the first version of CITA showed no gains in student performance in the summer of 2015. However, versions two and three of CITA showed small but statistically significant gains for students that used the system. This is especially apparent in the spring 2016 semester. Students who used versions two and three of CITA generally had positive things to say about the system. Surveys questions based on Likert scales along with open-response questions and focus group interviews all indicate that students found Shallow CITA particularly helpful and Immersive CITA helpful under most circumstances. However, students generally did not find the Postscripts helpful or worthwhile. Finally, I conducted a step-by-step analysis of the tutorials that were used in the CITA system in order to track student retention. These results were synthesized with student responses from focus group interviews to gain an understanding of what motivates students to use CITA tutorials over other external resources (e.g. Google, Yahoo Answers, the course textbook, etc.). Since students were able to choose between CITA and other external resources, they adopted strategies to efficiently solve homework problems rather than maximize learning. Students left a CITA tutorial if they encountered a step which was perceived as tedious. Thus, online tutorials must be designed to prevent students from prematurely leaving the tutorial system.

Degree

Ph.D.

Advisors

Bryan, Purdue University.

Subject Area

Physics|Educational technology|Science education

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS