Back

UCLA's Competitive Edge Program Provides an Advantage to STEM Doctoral Students from Historically Excluded and Underrepresented Groups

Del Carpio, C. A.; Azurdia, D. E.

2023-09-11 scientific communication and education
10.1101/2023.09.08.555984 bioRxiv
Show abstract

Universities benefit from recruiting and retaining diverse students, as it leads to more creative and rigorous problem solving. Efforts to improve the diversity, equity, and inclusion of graduate education have historically been focused on recruitment but are now shifting to retaining enrolled students. A low percentage of doctoral students, particularly those from historically underrepresented groups (URGs), complete their PhD within 10 years. To increase support of incoming doctoral students from URGs, UCLA established the Competitive Edge (CE) bridge program. CE provides six weeks of professional development and research training at the start of the PhD program. We surveyed 55+ first-year doctoral students (14 CE students, eight non-CE students from URGs, and 34 well-represented (WR) students) in STEM fields to understand CEs effectiveness. We found that the CE program aided students in four areas that influence graduate student attrition: mentor/mentee relationship, socialization, finances, and preparedness. At the end of their first academic year, CE students reported that the program helped them in multiple areas relating to student success, such as mental wellbeing and sense of belonging. CE students reported a larger mean growth in seven of eight skills needed in graduate school compared to NonCE URG and WR students. Short answer responses revealed that NonCE students wished for more support in areas covered by the CE program, such as managing advising relationships and protecting mental health. Additionally, CE students received significant funding during the program. The CE programs successful model at UCLA can be adapted to improve support for underrepresented doctoral students at other institutions.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.