登入選單

返回Google圖書搜尋

Priority Challenges for Social and Behavioral Research and Its Modeling

Paul K. DavisAngela O'MahonyTimothy R. GuldenOsonde A. Osoba

出版	RAND Corporation, 2018
主題	Medical / Clinical MedicineMedical / Diet TherapyMedical / Health Care DeliveryMedical / Health PolicyPolitical Science / Security (National & International)
ISBN	08330999579780833099952
URL	http://books.google.com.hk/books?id=Sqt4tAEACAAJ&hl=&source=gbs_api

註釋

This paper summarizes priority challenges for social and behavioral modeling based on a recent study building on a base of prior studies. Our focus is less on describing and hand-wringing about the current state than on identifying what is necessary for moving on. Some of the obstacles reflect inherent challenges: social systems are complex adaptive systems; they often pose "wicked problems;" and even the structure of social systems shows emergent behavior. Other obstacles reflect disciplinary norms and practices, mindsets, and numerous very difficult scientific and methodological challenges. We discuss challenges in six groups: (1) tightening links among theory, modeling, and both empirical and computational experimentation; (2) seeking more general and coherent theories while retaining alternative perspectives and narratives, and while effectively confronting multidimensional uncertainty from the outset; (3) assuring that explanatory models represent science faithfully, to include addressing aspects and determinants of behavior that have often been omitted or treated with hard-wired representations; (4) challenging experimenters to find new theory-informed (but not theory-imposing) ways to obtain and analyze relevant data in this modern era of ubiquitous data; (5) challenging theorists and technologists to provide related methods and tools; and (6) nurturing the rest of the ecology needed for overall effectiveness. We suggest identifying several national challenge problems and, for each, having a distributed and virtual social and behavioral laboratory to stimulate synthetic interdisciplinary work. These should feature mixed methods (not just classic simulation) and competition, but also frameworks, modularity, and problem-focused composition--again with competition and evolution-- rather than an imagery of standing "correct" federations. Experience shows that breakthroughs often occur as the result of solving concrete problems and then recognizing more general patterns.