Optimizing Mentoring: A Research-Based Application for Ideal Mentor-Mentee Relationships
Location
SU-003
Start Date
2-5-2025 11:20 AM
Department
Other
Abstract
Mentoring is one of the most valuable professional relationships for a mentee’s career development, while also benefiting the mentor. However, establishing a strong and effective mentorship connection remains challenging, as there is no blueprint for success. This research explores the development of a research-based application to optimize mentor-mentee relationships, ensuring long term engagement and meaningful professional development. We hypothesize that personality type, demographics, career, and interests contribute to ideal mentoring relationships. Using these four dimensions, we calculate a matching percentage between mentors and mentees. To do so, we designed an experimental algorithm using a tier-based approach to test which dimensions carry more weight and what matching percentage is needed." The experimental algorithm is implemented in the development of an automated application for the mentor-mentee matching process. For Phase 1 of the experiment, a hundred (132) participants were asked to sign up as a mentor, a mentee, or both. The algorithm matched participants' responses. Mentors and mentees are blindly matched into pairs to prevent bias; the pairs undergo a guided mentoring experience for two months, later asking participants to complete a user experience survey—responses are used for algorithm adjustments. Out of fifty-three (53) participants, (48) completed the post-experience survey. Results show that our current algorithm provides a mentor-mentee match percentage within a +/- 10% range when compared to our perceived mentor-mentee match percentage from the survey, indicating a close approximation. To further optimize the current algorithm we plan to expand the sample size by recruiting additional mentors and mentees until it is able to produce percentage matchings within a +/- 5% range. The increased sample size is developing to achieve the Phase 1 experiment and adjust the algorithm for Phase 2. We expect the research-based mentor-mentee app to be an effective approach to mentoring, with the wide use of a well-tested algorithm.
Faculty Sponsor
Doris Espiritu
Optimizing Mentoring: A Research-Based Application for Ideal Mentor-Mentee Relationships
SU-003
Mentoring is one of the most valuable professional relationships for a mentee’s career development, while also benefiting the mentor. However, establishing a strong and effective mentorship connection remains challenging, as there is no blueprint for success. This research explores the development of a research-based application to optimize mentor-mentee relationships, ensuring long term engagement and meaningful professional development. We hypothesize that personality type, demographics, career, and interests contribute to ideal mentoring relationships. Using these four dimensions, we calculate a matching percentage between mentors and mentees. To do so, we designed an experimental algorithm using a tier-based approach to test which dimensions carry more weight and what matching percentage is needed." The experimental algorithm is implemented in the development of an automated application for the mentor-mentee matching process. For Phase 1 of the experiment, a hundred (132) participants were asked to sign up as a mentor, a mentee, or both. The algorithm matched participants' responses. Mentors and mentees are blindly matched into pairs to prevent bias; the pairs undergo a guided mentoring experience for two months, later asking participants to complete a user experience survey—responses are used for algorithm adjustments. Out of fifty-three (53) participants, (48) completed the post-experience survey. Results show that our current algorithm provides a mentor-mentee match percentage within a +/- 10% range when compared to our perceived mentor-mentee match percentage from the survey, indicating a close approximation. To further optimize the current algorithm we plan to expand the sample size by recruiting additional mentors and mentees until it is able to produce percentage matchings within a +/- 5% range. The increased sample size is developing to achieve the Phase 1 experiment and adjust the algorithm for Phase 2. We expect the research-based mentor-mentee app to be an effective approach to mentoring, with the wide use of a well-tested algorithm.