Abstract The United States' inability to achieve equitable workforce development in biomedical career pathways is well- recognized and has been attributed to poor retention of a diverse stream of students in academia.1,2 National data examining retention of individuals from historically underrepresented (HU) groups at the undergraduate, graduate, and faculty levels in science, technology, engineering, and mathematics (STEM) show that the disparity between the HU and historically overrepresented (HO) groups increases as people progress through these career pathways.3 One theoretical model applied to understand and address this issue is the Tripartite Integration Model of Social Influence (TIMSI)15,16, which will be used to inform the design of a Social Inclusion Intervention (SII) within the context of the Small World Initiative (SWI). The scope of the SWI program provides an ideal context in which to conduct this rigorous multisite, longitudinal, randomized experiment with HU and HO faculty and students. Having grown rapidly since 2012, SWI now includes 151 undergraduate institutions in the US, with 29 HSIs, 5 HBCUs, and 2 TCUs. In this context, the first aim of the proposed study is to experimentally test the hypothesis that among HU and HO faculty participants the SII will result in higher quality mentorship from the assigned peer mentor and short and long-term growth in mentor networks, mentorship activities, improvements in scholarly productivity, and joy in work. The second aim is to test the hypothesis that the SII with undergraduate HU and HO students, implemented by faculty in their classrooms (cluster randomly assigned), will result in a higher degree of identification with and support from their faculty mentor, the development of stronger mentor networks, as well as growth in interest and retention in biomedical fields. The third aim is to test the hypothesis that indices of integration into the scientific community will mediate the relationship between the quality of mentorship and inclusion and short- and longer-term outcomes for both faculty and student participants. Results will be derived from longitudinal data from 200 biomedical faculty participants, a short-term sample of 3,000 student participants in SWI courses (with a minimum of 25% being from HU groups), and a long-term stratified random subsample of 1040 student participants with 2 years psychosocial data, providing robust power to utilize multilevel growth curve analysis, social network analysis, and structural equation modeling based mediation analysis to test hypotheses that extend previous findings regarding TIMSI for both HU and HO faculty and students. The proposed study will contribute to the DPC goals of advancing the science of workforce development by (a) identifying a theoretically guided low cost and scalable intervention to increase mentor networks and inclusion experiences, (b) measuring for whom this intervention has the strongest impact (e.g., moderation by HU/HO status), (c) providing a deeper understanding of why an intervention may impact outcomes (i.e., mediation), and (d) extending knowledge about the effects of inclusion and kindness for faculty and students in the context of broadening biomedical career pathways.

Narrative The United States' inability to achieve equitable workforce development in science career pathways is well- recognized and has been attributed to poor retention of a diverse stream of scholars in academia. Using a theory-driven, longitudinal experimental design, results will determine when, how, and for whom a scalable social inclusion intervention improves faculty and student mentorship, professional networks, and integration into biomedical career pathways. The results will inform the field of mentorship science and broadly impact future mentorship programs that aim to broaden participation in the biomedical workforce.