Positive Relationships Required: Advisor and Peer Relationships Improve Degree Completion Intentions for Graduate Engineering Students

• In Event: Attrition (Optional) and Relationships (Required): Educational Relationships and Decision Making to Increase Engineering Graduate Student Persistence

Abstract

Graduate engineering students leave doctoral programs without their intended degrees at high rates with students from minoritized groups leaving at higher rates than their peers (NSF, 2018). The individual loss results in talented and well-qualified people not able to achieve their academic and career goals and potential. The societal loss means advances and innovations vital to society remain undiscovered. While attrition can be attributed to numerous factors, the interpersonal relationships of graduate engineering degree programs is one malleable component that can significantly hinder or bolster attrition. Researchers have found that doctoral research advisors have strong influence on student persistence to degree, where positive relationships can support degree completion while negative experiences contribute to dropout (Berdanier et al., 2020; Burt et al., 2018). In addition, lab group and research peers uniquely contribute to students’ intentions to complete their doctoral degree. While the direct influence of advisors and peers on degree completion has received some attention, the mechanism of influence remains under researched. Professional identity or in this case graduate engineering identity presents one potential mechanism. In the current study we examine graduate engineering identity as a mediator of the association between research advisor and peer relationships and students’ doctoral degree completion intentions. We hypothesize that engineering identity then relates to intentions to finish the graduate engineering program. Given previous research on the variations in degree completion by race/ethnicity and gender, we examine whether this mediation model functions similarly for students from well represented and underrepresented racial/ethnic and gender groups.
In a national sample of 913 graduate engineering students from 22 engineering disciplines, we used structural equation modeling (SEM) to test our research questions. First, we tested the mediation model. Given previous research that shows engineering identity and degree completion vary across engineering discipline, we applied cluster robust error terms to control for the nested nature of our data. Overall, the model fit the data well and explains a good amount of variation (SRMR = .008; CD =.270). We found that advisor and peer relationships support different aspects of graduate engineering identity and degree completion intentions. For instance, advisor relationships generally support engineer identities while peer relationships support researcher identities. Next, we examined a multigroup model across six groups: underrepresented minority (URM) women, Asian women, white women, URM men, Asian men, and white men. The model fit remains strong, and the amount of variance explained increased for URM women (SRMR = .028; CD = .333), white women (SRMR = .017; CD = .361), URM men (SRMR = .038; CD .450), Asian men (SRMR = .025; CD = .358), but not for Asian women (SRMR = .033; CD = .264) and white men (RMR = .011; CD = .260). For each group, the significant direct effects vary in how variables connected and the strength of the connection. The model provides insights for ways engineering systems can engage advisors and peers to better support students as an avenue to increasing persistence to degree. The findings highlight the importance of equitable systems, structures, and relationships that serve underrepresented students.

Authors

• Matthew Bahnson, Pennsylvania State University

• Elan Hope, North Carolina State University

• Derrick James Satterfield, University of Nevada - Reno

• Adam Kirn, University of Nevada - Reno