OOS 13-3
Modeling qualitative social data: Collaborative approaches for and continuing challenges to crossing the qualitative-quantitative divide
Bringing together rich ecological, physical, and social data in order to study coupled human and natural systems (CHANS) has posed special problems for modeling the Anthropocene. While many natural and physical scientists can bring together quantitative data sets, the integration of social data has proved more elusive. One approach has been to embrace models of human behavior that can be quantified, particularly those that emphasize rational decision-making. Unfortunately, these models of human behavior (e.g., agent based modeling) often do a poor job of capturing the deeper understandings of human behavior studied across the social sciences – particularly, those social sciences that study human behavior empirically and inductively. Thus, the needs of CHANS modelers to find tractable “social data” has resulted in the use of models of human behavior that rarely do a good job of modeling complexity in real-world human behavior. How, then, does one integrate “the social” into ecosystem modeling in a way that better reflects complexities – both those amenable to qualitative and quantitative data outputs – of real world, empirically-observed human behavior? We illustrate a solution to this problem through a case study drawing on empirical observations of human behavior and ecological systems interacting around a threatened watershed in Oklahoma.
Results/Conclusions
We describe efforts to rethink “social data” to bridge the qualitative-quantitative divide in a coupled human and natural systems modeling effort. The lead author, an anthropologist, describes research emerging from Oklahoma’s current NSF EPSCoR-funded efforts to establish “socio-ecological observatories” across the state to understand the long-term impacts of climate change, climate variability, and severe weather. We show how qualitative and quantitative “social data” can be integrated in order to provide four functions in efforts to model coupled human and natural systems: 1) parameterization; 2) scenario generation; 3) process identification; and 4) initial conditions and distributions. Collaboration among disparate social scientists (anthropologists, sociologists, economists, and political scientists) and natural/physical scientists (ecologists, hydrologists, meteorologists, and climate modelers) on our research team has resulted in an approach to modeling that not only attempts to account for the modeling of human behavior from a rational actor approach but also can account for models of human behavior that do not assume that human behavior reflects maximizing, rational, decision-making. We conclude by arguing that it is necessary to capture greater human complexity in order to understand the Anthropocene particularly regarding efforts to model the possible futures that face coupled human and natural systems.