Student Learning Groups: Does Group Composition Matter?

Abstract

There has been a shift in university teaching over the past decades to emphasize student achievement and persistence through high-impact practices and collaborative learning. While research supports the efficacy of these pedagogical strategies, it can be difficult to implement them on large campuses. Yet, many criminal justice professors assign students to work in small groups. This study, conducted at a university with a large underrepresented student population, found that the composition of student groups in one criminal justice class can affect the perceived benefits students receive in that class, in other classes, and in general. We found that students placed in a small group with classmates who shared some criminal justice classes reported more student-initiated collaborative learning, expanded support networks, and improved grades in shared courses compared to students in small groups with classmates who shared no classes. This simple intervention has the potential to mimic learning-communities with similar benefits.

Notes

¹ Institutional Review Board approval was obtained each year.

² Random Assignment was attempted but was impossible because there was too much crossover between this course and other courses each semester. It was impossible to create control groups from the students randomly assigned to the control condition in which no students shared courses and experimental groups from the students randomly assigned to the experimental condition in which students shared most classes.

³ The response rate does not account for student attrition and therefore is slightly higher than stated for most semesters.

⁴ Each semester there were four or five comparably sized sections of the course offered, of which the first author taught two.

⁵ Students were not asked for their demographic information at the beginning of the course so that students were not alerted to the experiment and the research was not compromised.

⁶ It is possible that survey respondents differ from non-survey respondents. However, the consistency of respondent characteristics from semester to semester and the congruence between survey respondent characteristics and the characteristics of the general criminal justice student population lend credibility to our assumption that survey respondents are similar to non-survey respondents and representative of typical criminal justice majors at CSUF.