Abstract:

Data Structures and Algorithms (CS2430) is offered by the University of Houston Department of Computer Science and is a fundamental class for a career in computer science. On average, seventy percent of this class consists of transfer students. In past, an acquaintance matrix pairs students based on how well they know each other. This method causes problems however, with the high rate of transfer students, using this method might be difficult. Also, from a previous report (N. Rizk, N. Bacchus – Using Data Science Algorithms to Enhance Student Group Performance), there is no evidence that groups formed off of acquaintances perform better than groups randomly put together.

Measures are put in place to ensure that data is suitable to pair the students for group formation before measuring their performance and effectiveness. Measurement of team performance defines the definition of team effectiveness. This metric also entails the likelihood of the teammates working together in the future.

Relatively speaking, the solution would be to give out two surveys to the students. One survey to handle the personality and working type of the student; this survey would measure pro-activity and the likelihood to work within a group. The second survey measures programming skills and is self-reported. This document will state their initial findings and experiment with general pairing strategies for participants. Pairing the students would be based on similarities in working personalities, but differences in skills. Pairing them will also allow for students to help each other in their weak spots while maintaining fluidity in group performance and time management.

Measures of effectiveness are based on self-reported results of satisfaction with their team members, their self-reported score of their performance on the team, and an account of their performance with their group assignment scores. A high level of team effectiveness can benefit the output of product/assignment, relations within the team. It can create a culture of collaborative success for the students

Keywords:

Data science, Unsupervised learning, Clustering.