SCISIPBIO: Constructing Heterogeneous Scholarly Graphs to Examine Social Capital During Mentored K Awardees Transition to Research Independence: Explicating a Matthew Mechanism

尽管努力实现多元化，但NIH R01著名奖项中有很大一部分授予了有限的个人和机构。这种情况是如何发生的，为什么会发生？马太效应，即成功带来成功，被认为是负责任的：申请者在一开始即使优势很小，但在最初的成功之后，他们的优势可能会成倍增加。证据与R01资金中马太效应的存在是一致的，但没有研究阐明优势的具体性质，也没有详细说明优势是如何倍增和积累的。这个项目将通过考察NIH导师职业发展奖(MK奖)获奖者的个人职业轨迹，回答关于社会资本和学术成就的哪些方面对R01的成功贡献最大，以及性别或学术活动的时机是否有贡献的问题。项目成果将有助于设计有效的干预措施，以避免意外的资金差距，同时保持严格的同行审查制度。这将是在向研究独立性过渡期间对马太机制的第一次经验测试，也是第一次利用异质学术图(HSG)。第一个目标是捕捉每个MK获奖者之间的复杂关系，他们的学术成就和社会资本，以及他们在寻求研究独立的过程中取得的R01成功。现有的书目和NIH奖数据将结合在一起，建立一个“全球”HSG数据库--将所有MK奖获得者与他们相关的学术对象联系起来。结果将是一个全面的图形结构数据库，其中节点表示所有MK获奖者及其关联的学术对象(例如，已发表的论文、期刊、主要学术机构、合著者、共同作者的学术对象)，而边表示各种类型的关系(例如，作者、被引用者、从属关系、研究主题)。通过全局、局部和超局部图形特征提取，将为HSG中的所有学术对象捕获关系上下文，以全面表征MK获奖者的学术档案。其次，将开发生存模型，以根据学术成就和社会资本的潜在变量和观察变量，以及全球、本地和超本地HSG特征来预测MK获奖者的R01成功。这项研究为研究复杂的社会过程提供了一种新的方法，将社会资本理论与不同种类的学术图表和网络科学方法结合在一起。这项研究将超越以往的研究，提供学术社会资本的多维表征(超越合著性和引文)，并将考察不同的社会资本积累在MK到R01过渡中的中介作用。以经验为基础的预测模型将被设计来探索现有的理论，并对社会资本在R01融资成功中的作用产生洞察力。这项研究将产生可操作的知识，为旨在提高效率奖和增加获奖者池的多样性的战略提供参考。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Despite efforts at diversification, an outsized proportion of prestigious NIH R01 awards go to a circumscribed group of individuals and institutions. How and why does this happen? The Matthew Effect, whereby success begets success, is thought to be responsible: applicants with even small advantage at the outset may have their advantage multiplied many times over following initial success. Evidence is consistent with the presence of a Matthew Effect in R01 funding, yet no study has illuminated the specific nature of the advantage, nor detailed the means by which advantage is multiplied and accumulated. This project will answer questions about which aspects of social capital and scholarly achievement contribute most to R01 success, and whether gender or timing of scholarly events contribute, by examining the individual career trajectories of awardees of NIH Mentored Career Development Awards (MK awards). Project results will help design effective interventions to avert unintended funding disparities, while maintaining a rigorous peer review system. This will be the first empirical test of a Matthew Mechanism during transition to research independence and the first to leverage heterogeneous scholarly graphs (HSGs). The first aim is to capture complex relationships between each MK awardee, their scholarly achievement and social capital, and R01 success during their quest for research independence. Existing bibliographic and NIH award data will be combined in the construction of a “global” HSG database - relating all MK awardees to their associated scholarly objects. The result will be a comprehensive graph structured database in which nodes represent all MK awardees and their associated scholarly objects (e.g., published articles, journals, primary academic institution, coauthors, coauthor’s scholarly objects), and edges represent relationships of various types (e.g., author of, cited by, affiliation, research topics). Relationship context will be captured for all scholarly objects in the HSG through global, local, and hyper-local graphical feature extraction to comprehensively characterize MK awardees’ scholarly profiles. Second, survival models will be developed to predict R01 success for MK awardees from latent and observed variables of scholarly achievement and social capital, and global, local, and hyper-local HSG features. The study offers a novel approach to studying complex social processes that marries social capital theory with heterogeneous scholarly graphs and network science methods. This study will go beyond previous studies in providing a multidimensional characterization of scholarly social capital (beyond coauthorship and citation) and will examine differential social capital accumulation as a mediator in MK to R01 transition. Empirically-grounded predictive models will be designed to probe existing theory and yield insights on social capital’s role in R01 funding success. This study will yield actionable knowledge to inform strategies aimed at improving efficiency awards and increasing the diversity of the awardee pool.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.