Assigning Functional Groups: The Influence of Group Size, Academic Record, Practical Experience, and Learning Style

Abstract

AbstractThis paper summarizes the results of a four‐year study (September 1992 ‐ December 1995) concerned with the performance of student groups in a senior engineering laboratory course. The investigation was conducted in two stages. In the first two years, the effect of group size, incoming GPA, practical experience, and the gender distribution of each group was investigated. During this period we recorded the Kolb Learning Style Inventory (LSI) scores at the end of the semester and asked students to report on the performance of their groups given their knowledge of the LSI distribution within their team. In the second stage of this study (1994–95) we evaluated the effect of grouping according to LSI, in addition to continuing our study of the effect of group size, academic record, practical experience, and gender distribution. In the final year of the study we took advantage of the disparity in the incoming GPAs of the two sections of the class (Tuesday and Thursday) to evaluate if incoming GPA influenced course grade. The study consisted of four senior classes totaling 110 students in 33 groups. The learning styles distribution of the students resulted in 6% “Type 1,” 42% “Type 2,” 42% “Type 3,” and 10% “Type 4” learners. The metric used to quantify performance was the average final course grade of students within given groups. This course grade was equally weighted between technical and writing components. Our results indicate that the most important positive correlating factor in a group's performance was the group size (four member groups statistically outperformed three member teams at α = 0.05). Although not statistically significant, observable higher average group grades indicated that the following may have an effect on group performance: the inclusion of academically outstanding individuals, the number of members with “good hands,” and the GPA history of the group. Specifically, the inclusion of a student with a GPA above 3.6 improved the performance (average group grade) of the group relative to their abilities as characterized by their average incoming GPA. Students who were good with equipment or had some practical hands‐on experience had a similar positive influence on the group performance. The gender distribution within a group did not have a significant effect on either group performance or dysfunction. Insufficient data were collected to ascertain the relative performance of homogeneous and mixed learning style groups. Since group incoming GPA may be a variable in group performance, student self‐selection is not recommended since it would result in an amplified disparity in the course grades. Indeed, we observed that grouping students by GPA, group size, and LSI resulted in a large number of functional teams, with the final variance in the course grade within a class reduced relative to other courses which have grouped activities.