Understanding interaction network formation across instructional contexts in remote physics courses

Engaging in interactions with peers is important for student learning. Many studies have quantified patterns of student interactions in in-person physics courses using social network analysis, finding different network structures between instructional contexts (lecture and laboratory) and styles (active and traditional). Such studies also find inconsistent results as to whether and how student-level variables (e.g., grades and demographics) relate to the formation of interaction networks. In this cross-sectional research study, we investigate these relationships further by examining lecture and lab interaction networks in four different remote physics courses spanning various instructional styles and student populations. We apply statistical methods from social network analysis—exponential random graph models—to measure the relationship between network formation and multiple variables: students’ discussion and lab section enrollment, final course grades, gender, and race or ethnicity. Similar to previous studies of in-person courses, we find that remote lecture interaction networks contain large clusters connecting many students, while remote lab interaction networks contain smaller clusters of a few students. Our statistical analysis suggests that these distinct network structures arise from a combination of both instruction-level and student-level variables, including the learning goals of each instructional context, whether assignments are completed in groups or individually, and the distribution of gender and major of students enrolled in a course. We further discuss how these and other variables help to understand the formation of interaction networks in both remote and in-person physics courses.

Figure 1

Network of five nodes with all edges present (left) and one possible realization of this network (right).

Figure 2

Diagrams and densities of interaction networks for M-Eng and M-Phys. Nodes are colored by gender and sized proportional to total degree (number of adjacent edges). Thick edges represent reciprocal edges (students A and B both reported interacting with one another) and thin edges represent one-way edges (student A reported interacting with student B, but student B did not report interacting with student A). Densities are the proportion of observed to possible edges, with standard errors of the last digit shown in parentheses. These same network diagrams with nodes colored by race or ethnicity are in the Supplemental Material [76].

Figure 3

Diagrams and densities of interaction networks for EM-Eng and EM-Phys. Nodes are colored by gender and sized proportional to total degree (number of adjacent edges). Thick edges represent reciprocal edges (students A and B both reported interacting with one another) and thin edges represent one-way edges (student A reported interacting with student B, but student B did not report interacting with student A). Densities are the proportion of observed to possible edges, with standard errors of the last digit shown in parentheses. These same network diagrams with nodes colored by race or ethnicity are in the Supplemental Material [76].

Figure 4

Plot of ERGM coefficients for the homophily on lab section and homophily on discussion section variables. A more positive (negative) coefficient estimate indicates that more (fewer) edges occur between students in the same section. Error bars indicate the standard error for each estimate and asterisks indicate statistical significance.

Figure 5

Plot of ERGM coefficients for the main effect of final course grade on degree variable for each observed network. A more positive (negative) coefficient estimate indicates that students with higher final course grades have more (fewer) total connections in the network than students with lower final course grades. Error bars indicate the standard error for each estimate and asterisks indicate statistical significance.

Figure 6

Plot of ERGM coefficients for the two predictor variables related to gender. A more positive (negative) coefficient estimate for the main effect of gender on degree variable indicates that women have more (fewer) connections than men. A more positive (negative) coefficient estimate for the homophily on gender variable indicates that more (fewer) edges occur between students of the same gender. Error bars indicate the standard error for each estimate and asterisks indicate statistical significance.

Figure 7

Plot of ERGM coefficients for the two predictor variables related to race or ethnicity. A more positive (negative) coefficient estimate for the main effect of race or ethnicity on degree variable indicates that URM students have more (fewer) connections than non-URM students. A more positive (negative) coefficient estimate for the homophily on race or ethnicity variable indicates that more (fewer) edges occur between students of the same race or ethnicity. Error bars indicate the standard error for each estimate and asterisks indicate statistical significance.

Figure 8

Goodness-of-fit plots for the M-Eng lab interaction network. The horizontal axis represents the value of the network measure and the vertical axis represents frequency. Plots compare the distribution of each measure for the observed data (thick black line) to that for 10 network simulations generated using the coefficient estimates of the model (box plots).