Files in this item

FilesDescriptionFormat

application/pdf

application/pdfLEE-DISSERTATION-2020.pdf (3MB)Restricted Access
(no description provided)PDF

Description

Title:Using grant applications to measure the evolution of collaborative and non-collaborative research
Author(s):Lee, Joo Ho
Director of Research:Blake, Catherine
Doctoral Committee Chair(s):Blake, Catherine
Doctoral Committee Member(s):Twidale, Michael; Carpenter, Catherine; Darch, Peter; McInnes, Bridget
Department / Program:Information Sciences
Discipline:Information Sciences
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Scientific collaboration
Biomedical informatics
Semantic similarity
knowledge representation
Abstract:Collaboration has become popular across scientific and other disciplines for creating innovative products or acquiring knowledge. Government funding agencies actively encourage collaborative work, supporting thousands of research projects. The increase in collaboration has led to innovative ways of measuring research dynamics, and the impacts of collaboration with respect to co-authorship, citation analysis, and network analysis has been investigated in many science disciplines. While large-scale data have been analyzed to discover patterns of collaboration among whole populations and specific disciplinary groups, we still have much to learn about how collaboration affects an individual’s scientific progress. Grant funding is critical for advancement in the science disciplines, and this study focuses on unravelling the properties of collaboration and collaborators of biomedical research grants by finding patterns underlying semantic representations of text, using three research questions: 1) Is the degree of change in the research question smaller between projects of the same collaboration status than among projects of different collaboration status?; 2) Do research questions asked in non-collaborative projects change less than questions in collaborative projects?; and 3) Do the principal investigators publish their works in interdisciplinary journals after collaboration begins? A key goal of this study is understanding similarities and differences of collaborative and non-collaborative projects in grant applications with a novel approach of semantic similarity. The findings on this study indicate that grant applications in general shows no change within a grant cycle regardless of collaboration status. This study uses three factors to evaluate project-to-project change data: 1) activity types of research projects; 2) transit status of a pair of projects; and 3) chronological order of a project in the PIs’ trajectory during grant career with the National Institute of Health (NIH). PIs start a new project with topics which somewhat overlap with the ones from the past project. Once PIs start to collaborate, more than half of them keep collaborating on their next projects. The degree of change between projects is greater in non-collaborating projects than collaborating projects for some factors. Individual investigators may be better placed to pursue a rather different direction than they have studied earlier in their projects. For the second research question in examining year-to-year change in a grant project cycle, the factor activity types, number of conducted projects, and support year are explored along with similarity measures in population, disease, and intervention facets of Population, Intervention, Comparison, and Outcome (PICO) elements. The analysis implies that the forming and storming components of a collaborative project take place before the grant is submitted and the investigators have reached consensus on what to study in their research project. The administrative changes or revision in PICO elements are applied to projects regardless of the collaboration status or factors, but each factor displays contradictions between collaborative and non-collaborative projects. The analysis on the third research question indicates that interdisciplinary collaboration, or collaboration in translational research, or any collaboration in general might lead researchers to publish their works in multidisciplinary science journals. But multidisciplinary science covers a broad range of science disciplines, which does not provide enough information to reveal the actual disciplinary information that is integrated in the study. This study provides fine-grain analysis of grants by examining how the grant evolves over the grant cycle of a grant project with and without PICO elements in short timeframes in PIs’ career trajectory in NIH grant. Further efforts should be made to explore larger timeframes to increase an understanding of the evolution in collaboration in the context of grant applications. Lastly, this study may provide better insight for funding agencies while making policies and allocating grant funding for scholars in collaborative research.
Issue Date:2020-11-10
Type:Thesis
URI:http://hdl.handle.net/2142/109575
Rights Information:Copyright 2020 Jooho Lee
Date Available in IDEALS:2021-03-05
Date Deposited:2020-12


This item appears in the following Collection(s)

Item Statistics