Research on mentorship, collaboration, scientific careers, incentives, and the organization of science at the University of Colorado Boulder.
What shapes the people and institutions that produce science?
Science is made by people working within teams, institutions, and systems of recognition and support. We use publication records, mentorship networks, and computational methods to study how those systems relate to scientific activity.
Mentorship and scientific careers
Our work studies mentorship as a relationship that extends beyond coauthorship. The open mentorship dataset connects academic-family records to publications and representations of research, with documented coverage and estimation limits.
Teams, recognition, and opportunity
The publication archive includes studies of collaboration, shared leadership, recognition, and disparities in scientific careers. These questions require care about what the available records capture and which people or contributions they leave out.
Funding under uncertainty
Research programs face uncertain outcomes. Our decision-theoretic work makes costs and benefits explicit when considering research evidence, complementing work on funding, incentives, and the direction of scientific activity.
An open question
How can we build useful measures of scientific activity without treating incomplete records or predicted demographic attributes as a complete account of a person's contribution?
This data descriptor introduces a resource linking academic mentorship relationships to publication records, research representations, and demographic estimates. It is designed to support analysis of mentorship and scientific careers.
Jean F Liénard, Titipat Achakulvisut, Daniel E Acuna, et al. · 2018
This observational study examines how graduate and postdoctoral mentorship relate to later academic careers. It asks whether combining ideas from mentors with different expertise predicts a trainee's subsequent success.
A study's chance of being correct is only one part of deciding whether a research program is worth pursuing. This paper develops a decision-theoretic framework that makes potential costs and benefits explicit.
This study evaluates whether journal websites and publication metadata can support large-scale screening for questionable open-access journals. It treats automated predictions as a way to focus expert investigation.