A Randomized Control Study to Determine the Impact of Learning Communities that Link Three Undergraduate, First-Year Science and Mathematics Courses

点击翻译按钮获取中文摘要

This project serves the national interest by working to improve the retention and performance of students in science and engineering courses. Many institutions place groups of students into the same course sections, thus linking students together in learning communities. Such learning communities are a component of many efforts to retain students in science, technology, engineering, and mathematics (STEM) fields. Although the approach is often used, thus far there have been very few rigorous scientific studies that probe the effectiveness of these linked communities. This project intends to complete such a study using randomized assignment of students to linked communities or to an unlinked control group. To set up communities for the study, groups of students enrolled in the same three introductory STEM courses, will be placed into small seminar classes focused on contemporary issues such as biofuels or access to clean water. The effects of being linked as a group into the same classes will be determined by investigating the students’ sense of belonging and their motivation to continue their STEM coursework. The project hopes to identify possible reasons why being in linked communities can improve student grades and retention. The study will also examine which groups of students are most impacted by participation in the linked community. Results of this analysis will provide guidance about whether these communities are a viable approach to increase diversity in STEM. In this way, this project has the potential to offer a greater understanding about how and why these communities work, and thus offer a possible path to replication at other institutions. Mathematics placement scores and stratified random sampling will be employed to place first-year, full-time STEM majors into one of four groups: initial STEM coursework in a linked community; initial STEM coursework unaffiliated with a community (i.e., controls); pre-STEM major coursework in a linked community; and pre-STEM major coursework unaffiliated with a community (i.e., controls). This sampling method will ensure that groups represent the gender, income, racial, first-generation status, and residential/commuter status diversity of the host institution. Cohorts will be run for two years, accumulating at least 80 students in each experimental group and at least 120 students in each control group. Hierarchical linear modeling will be used to measure the effect of the learning communities on student grades and retention. A repeated-measures ANCOVA will be used to compare the pre/post responses to survey variables (i.e., sense of belonging, social capital, STEM motivation, and communal perceptions of science). Covariates will include relevant demographic variables and pre-existing markers of academic ability and performance. A mediation analysis will be conducted to see if the psychological variables measured by the survey (i.e., sense of belonging, social capital, STEM motivation, and communal perceptions of science) mediate the relationships between participation in the learning communities and grades and retention. Finally, each of the analyses will also be conducted as group comparisons on key demographic variables, including race, gender, commuter/residential status, and first-generation status. In this way, the project will test not only whether the learning communities affected student success overall, but also whether or not they affected all groups of students equally. The NSF IUSE: EHR Program supports research and development projects to improve the effectiveness of STEM education for all students. Through the Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.