Impact of Undergraduate Research Training Programs: An Illustrative Example of Finding a Comparison Group and Evaluating Academic and Graduate School Outcomes.

In this study, researchers at a large, urban, comprehensive minority-serving institution used propensity score matching to identify a unique comparison group to study academic and graduate school outcomes in students served by the National Institutes of Health-funded Building Infrastructure Leading to Diversity (BUILD) Initiative. Acknowledging that students' self-selection biases may confound findings, the use of propensity methods to match students served with those who were not (but were otherwise eligible) provides a valuable tool for evaluators and practitioners to combat this challenge and better evaluate their effectiveness and impact on students' success. This study's findings indicate that BUILD participants had higher academic and graduate school success with regard to cumulative GPA, units attempted and completed, graduation status, and application and admission to graduate programs.

The BUILD Mentor Community at CSULB: A Mentor Training Program Designed to Enhance Mentoring Skills in Experienced Mentors.

The BUILD Mentoring Community (BMC) at California State University Long Beach (CSULB) was developed to enhance mentoring skills among our already experienced research faculty mentors. Designed in alignment with the published "Entering Mentoring" program, the 2015-2019 BMC trained 93 research mentors across 24 departments. Mentors discussed best practices in mentoring in a hybrid format during the first semester and completed a second semester independent project where refinements to their mentoring were piloted. Mentors were surveyed following BMC completion with a Qualtrics survey, and BMC-trained mentors (BMCT), BUILD non-BMC trained mentors (BNT), and non-BUILD (NB) faculty mentors and their students were surveyed using the Higher Education Research Institute (HERI) and Diversity Program Consortium (DPC) instruments. BMCT mentors found that the workload of the BMC was reasonable, and 86% of mentors would recommend the BMC. Most (97%) BMCT mentors stated that they were likely to make a change in their mentoring as a result of BMC participation. Both BMCT and BNT mentors rate mentoring undergraduates more highly, and present work with undergraduates more frequently, than NB mentors. Students perceive BMCT mentors as providing a better mentoring experience and being better at increasing motivation and confidence, setting expectations, and acknowledging contributions compared to the ratings of students without BMCT mentors. While students rated BMCT mentors better on many key skills taught in the BMC, some learning goals did not produce a difference, including discussing and valuing diversity, using active listening and constructive criticism, and employing communication strategies. Therefore, many aspects of mentor training at CSULB can improve. Overall, instituting online/hybrid mentor training enhanced mentoring skills even among experienced mentors, particularly when mentors were asked to apply and assess new mentoring practices as part of the program.