New study explores the risks for premature match closure in an online mentoring program

Uebler, C., Emmerdinger, K. J., Ziegler, A., & Stoeger, H. (2023). Dropping out of an online mentoring program for girls in STEM: A longitudinal study on the dynamically changing risk for premature match closure. Journal of Community Psychology. 

https://doi.org/10.1002/jcop.23039

Summarized by Ariel Ervin

Notes of Interest:

• Although mentoring is an efficient approach to promoting youths’ academic, professional, and personal development, improper implementation of it will leave a small or moderate impact.
• Given the salience of premature match closures, it is essential for scholars and practitioners to get a more comprehensive understanding of how they come about.
  • Many researchers who studied this issue examine premature match closure after it occurred.
• This study longitudinally assessed the risks for premature match closure in an online mentoring program for girls in STEM.
• Findings indicate that age isn’t a notable predictor of premature match closure.
• Mentees’ compliance and interest in the program content can decrease the risk of premature match disclosure.
• Compliance with program specifications plays an essential role in mentorship duration and mentoring outcomes.
• Having continuous and frequent communication (e.g., emails and chat sessions) with mentors can also reduce the chances of premature match closures, especially if the conversations are relevant to the program content.
• Mentors’ previous experience and mentee-to-mentee networking mitigate the risk of premature match closure.

Introduction (Reprinted from the Abstract)

Premature closure of mentoring relationships decreases positive effects of mentoring or can even lead to negative effects for mentees. Past studies retrospectively investigated mechanisms of premature match closure. However, a deeper understanding of the dynamics that lead to premature match closure is still missing. In our study, we longitudinally examined the preprogram characteristics, program adherence, as well as program communication and networking behavior of girls (N = 901, M = 13.80 years) who took part in a 1-year online mentoring program in science, technology, engineering and mathematics (STEM), comparing girls who dropped out of the program prematurely (N = 598) with girls who were considered as non-dropouts (N = 303). We used survival analysis methods to simultaneously analyze time-independent characteristics and time-dependent dynamics of mentees’ communication and networking behavior. Besides mentees’ interest in STEM and compliance with program specifications, a frequent and steady communication with their mentors decreased the risk for premature match closure, especially, if it focused on STEM. Mentors’ mentoring experience, mentees’ program-wide networking and their networking with other mentees reduced the risk for premature match closure. Regarding the STEM focus of networking, we found competing influences, which need to be further explored in future research.

Implications (Reprinted from the Discussion)

In the present study we investigated the dynamically changing risk for premature match closure in an online mentoring program for girls in STEM. For this purpose, we longitudinally observed the development of mentoring relationships as well as mentees’ communication and networking behavior, analyzing the role of various potentially influencing factors and their changes over time. First, we investigated whether individual-level characteristics, namely mentees’ age and STEM interest, predict premature match closure. Second, we investigated the influence of mentees’ compliance with program specifications on premature match closure. As proxy for initial program adherence, we chose mentees’ willingness to fill out a questionnaire before entering the program. As another proxy for program adherence, we used the number of logins of mentees on the platform, as all participants committed to communicating on the platform at least once a week when registering for the program. Third, we examined the role of various aspects of the mentoring process for premature match closure. More specifically, we explored the predictive power of the frequency and kind (and here especially the STEM focus) of communication between mentees and their personal mentors for premature match closure. We also investigated the role of mentors’ previous mentoring experience. Fourth, we investigated how mentees’ networking and STEM-focused networking relate to premature match closure, taking into account platform-wide networking, networking with other mentors and networking among peers.

Mentees’ Individual-Level characteristics and premature match closure

Our first aim was to investigate the role of mentees’ individual characteristics, namely age and STEM interest, for premature match closure. In contrast to previous research (Grossman & Rhodes, 2002; Kupersmidt et al., 2017b; Spencer et al., 2020), we did not find any consistent effect of age on premature match closure. Although age was not a significant predictor of premature match closure in most of our models, our results suggest that if age mattered at all, older mentees were at lower risk for premature match closure than younger mentees. This trend contrasts with previous research, which indicates that older mentees are at higher risk for premature match closure than younger mentees (Grossman & Rhodes, 2002; Kupersmidt et al., 2017b; Spencer et al., 2020). One possible explanation lies in the specific group of mentees and the setting of our study: Unlike previous research that focused mostly on youth mentoring with mentees at risk, we investigated premature match closure in an online mentoring program for girls in STEM. In this setting, two opposing age effects might come into play that cancel each other out: On the one hand, students’ — and especially girls’ — interest in STEM decreases with age (Frenzel et al., 2010; Kerr & Robinson Kurpius, 2004), which would imply that older mentees are at higher risk for early dropout of a STEM program. On the other hand, since older students use digital tools more frequently than younger students (Medienpädagogischer Forschungsverbund Südwest, 2020), and therefore might be more familiar and experienced with online platforms, older mentees’ risk to drop out could be decreased compared to younger ones. In addition, selection effects could come into play. Since STEM interest generally decreases with age (Frenzel et al., 2010; Kerr & Robinson Kurpius, 2004), older students registering for a STEM program might be those who are especially committed to the STEM domain compared to their age group, thus decreasing risk for early drop out. This assumption is supported by a study that showed that CyberMentor participants had higher STEM achievements, STEM interest, STEM activities and higher elective intentions for STEM (e.g., to choose a STEM career) than a same-aged random-sample control group (Stoeger et al., 2016). In contrast, younger mentees might be less committed to STEM and rather register for a STEM program to get to know various STEM areas for the first time, bringing with it a higher risk to drop out of the STEM program early on when compared to older mentees. This interpretation also aligns with our findings concerning the second individual-level characteristic we investigated: mentees’ STEM interest. Mentees who rated their STEM interest with a higher score when registering for the program, were less at risk for premature match closure. This finding is in line with previous research on the role of mentees’ motivation for premature match closure. For example, extrinsic reasons for program participation as well as mentees’ disinterest increase the risk for premature match closure (DeWit et al., 2016; Herrera et al., 2013; Spencer, 2007).

To access this article, click here.