Machine-Assisted Knowledge Creation: Reading and Writing for the 21st Century Information Economy (Pending)
Project Description: This SBIR Phase I project addresses a major gap in information technology and a related need in educational technology. By combining learning sciences with recent advances in computer science and user experience design, the HyLighter team is building a system that will enable rich interaction between both technical and nontechnical people, intelligent machines and ideas, and will be widely available through a freemium/premium Software-as-a-Service (SaaS) distribution model. Among other desirable effects and outcomes, users will be able to (a) improve their capacity to understand and synthesize information from multiple sources, (b) recognize non-obvious and meaningful associations across large and varied collections of documents, (c) generate new ideas and knowledge, and (c) communicate through clear, concise technical writing.
Motivation, Threat, and Engagement Intensity in Cross-Disciplinary Health, Biomedical, Policy, and Education Teams (Pending) Cross Disciplinary Research Fund, GWU
Gaetano R. Lotrecchiano, EdD, PhD, Assistant Professor of Clinical Research and Leadership, GWU School of Medicine and Health Sciences; Sean Cleary, PhD, Associate Professor, Associate Professor of Epidemiology & Biostatistics at the Milken Institute, School of Public Health; Kevin Cleary, PhD, Professor, Radiology, GWU School of Medicine and Health Sciences; Trudy Mallinson, PhD, Associate Professor, Clinical Research and Leadership, GWU School of Medicine and Health Sciences; Shelley Brundage, PhD, Associate Professor, Hearing and Speech, Columbian College of Arts and Sciences.
Project Description: Academe and national trends are urging team science scholars to provide greater and more specific directions about how stakeholders might better work together to encourage greater team fit and secure innovative scientific outcomes that positively impact society.1-3 The development of validated psychometric assessments is critical to the advancement of research tools that will address these issues. To date, there is widely available comprehensive assessment tool that assesses individual motivations and threats to health sector collaborations in teams with the goal of applying knowledge gained through assessment to understand degrees of engagement that accompany team goals. The Motivation Assessment for Team Readiness, Integration, and Collaboration (MATRICx) is a psychometric tool developed by our research team that measures individual motivations and threats to scientific collaborations for the purpose of informing these priorities.4,5 Now that a pilot instrument has been developed, the next phase of research is to test the tool by conducting a comparative analysis with cross-disciplinary teams who are working on issues that transcend historical or disciplinary boundaries. Our team is prepared to embark on comparative analyses between dedicated groups of diverse teams working on cross-disciplinary (CD) problems. This CDRF-funded phase of research will expand the MATRICx data pool, increase validity and reliability of the tool, allow us to consider interventions that cultivate more collaboration-ready workforces, and advance the research for larger scale funding opportunities that develop scientific readiness and training interventions.
Cognitive Task Analysis (Pending) Cross Disciplinary Research Fund, GWU
Ryan Watkins, PhD, Professor, Educational Leadership, GWU Graduate School of Education and Human Development; Tara Behrend, PhD, Associate Professor, Industrial/Organizational Psychology, GWU Columbian College of Arts and Sciences; Larry Medsker, PhD, Department of Physics, GWU Columbian College of Arts and Sciences. Project Description: Professionals are increasingly using the outputs of data analytics (i.e., machine learning, predictive analytics, artificial intelligence, big data) to inform a variety of decisions. From algorithms informing the scheduling hospital rooms, to complex models shaping how universities approach admissions, the role of data in our decision-making continues to increase sharply. However, it can be argued that outputs of data analytics are ideally operationalized when used to complement the judgment of experienced practitioners instead of replacing that judgment (Agrawal, Gans, Goldfarb, 2016). This optimal combination of analytical results and human judgment requires that professionals are prepared to both (a) understand the basics of the data analysis models that are generating algorithms and analytical results, and (b) interpret those analytical results in conjunction with their professional judgment. Our first hypothesis is that the confidence of professionals in their decisions would benefit from having the skills to open up the “black boxes” of data analytics, understand the basics of how the models and algorithms are generating results (including the biases that are inherently part of these), and then utilize that knowledge to appropriately complement their professional judgement with the outputs of data analytics. A supporting hypothesis is that experienced domain experts (such as healthcare professionals) do not have to become data scientists to adequately understand analytical models. Rather, we can use computational thinking as a framework for developing the required foundational knowledge and skills to integrate data analytics knowledge into decision-making.
