Back

Measuring Institutional Research Culture: An Analysis of Accepted Abstracts at the Academic Surgical Congress

Sathe, T. S.; Sathe, O. S.

2024-05-22 surgery
10.1101/2024.05.21.24307727 medRxiv
Show abstract

IntroductionUnderstanding which institutions have a strong research culture is important for trainees, but existing metrics such as NIH funding do not always reflect a trainees lived experience. Here, we present a descriptive analysis of accepted abstracts at the Academic Surgical Congress (ASC). MethodsWe analyzed accepted abstracts at the Academic Surgical Congress between 2016 and 2024. We identified institutions with the most accepted abstracts, overall and by year. We also identified institutions with the greatest year-over-year improvement between 2022 and 2024. Finally, we calculated associations between accepted abstracts and NIH funding and also between accepted abstracts and leadership in the ASCs hosting societies. ResultsA total of 12,142 abstracts were accepted by the ASC between 2016 and 2024. The University of Alabama and the University of Michigan led overall abstract acceptances and also led in most years. Baylor and Massachusetts General Hospital had the greatest absolute and percent year-over-year increase in abstract acceptances in the post-COVID era. There was a positive association between NIH funding and abstract acceptances. Among the top 40 institutions by accepted abstracts in 2024, institutions in the top 10 were over 10 times as likely to have leadership representation in the societies that host the ASC. ConclusionsConference abstracts may serve as a useful measure of trainee-relevant research culture. Trainees can utilize this information when selecting residency programs, and programs can use this information to optimize trainee research experiences.

Matching journals

The top 5 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.