| Recent reforms of Social Studies Education (NCSS 2013) have called for increasing focus on "complex causal reasoning", and the use of computer simulations for testing policy interventions. While these focal areas mirror moves in the hard sciences, e.g. NGSS (Lead States 2013), little existing work has focused on exploring what it means to reason causally about social systems, or how to design simulation activities that encourage and improve on this kind of thinking.;In this dissertation, I first propose a definition of Complex Social Systems Thinking (CSST) which takes into account the differences between reasoning about social and non-social complex systems.;One set of differences relate to the importance of reasoning about the circumstances under which humans behave in particular ways. I present the design and implementation of a 4-day high school curriculum activity on urban planning that is designed to encourage and improve these components of CSST in students. I then present an analysis of students work in the unit, and an analysis of their conceptual change during the unit. We find that students' explanations increasingly take into account local heterogeneity, and the role of human motivation and desires when explaining the consequences of urban planning between pre- and post-responses.;Another set of differences relate to the importance of reasoning about causality at multiple levels in social systems. To address this, I present my design of LevelSpace, an extension of NetLogo that allows learners and modelers to build and 'think-with' many concurrently interacting models at many different levels. I then present the design and implementation of a 3-day curriculum activity on sustainable food production using LevelSpace. I present an analysis of students' work with these multi-level model systems, focusing on the difficulties facing students when dealing with much larger complex systems with many more complex causal-relationships. We find that students move from reasoning about smaller sets of causal-mechanisms within models to including larger sets of mechanisms between models when using the designed model-activity.;The dissertation thus makes three primary contributions: 1) a theoretical contribution by offering a framework for further investigating "complex causal reasoning" about social phenomena; 2) a design contribution by offering design artefacts, tools, and design principles for supporting students in CSST; and 3) an empirical contribution by studying conceptual change in a classroom implementation within the CSST framework. |
