causal inference in the social sciences社会科学中的因果推论精品.pdfVIP

Causal Inference in the Social Sciences: Variance Theory, Process Theory, and System Dynamics Don R. Morris Miami-Dade County Public Schools 1500 Biscayne Blvd., Ste. 225, Miami, Florida 33132 Tel: 1 305 995 7531 / Fax: 1 305 995 7521 donr.morris@worldnet.att.net The social sciences are in need of an alternative to the variance approach to causal inference, which—because it requires a counterfactual—restricts the claim of valid inference to conclusions drawn from experimental and quasi-experimental methods. Process theory proposes an alternative by way of demonstrating the presence of observable characteristics of the causal mechanism, a method that, while accepted in principle, has proven elusive in practice. It is suggested that system dynamics can improve the process approach, and in so doing open a path for itself to wider application in the social sciences. Educational research is suggested as a place to start, and results from two models drawn on topics from that field are offered as examples. Keywords: causal mechanism; causal process; educational research; modus operandi; nonexperimental methods; program theory I have read with much interest Nelson Repenning ’s (2003) account of his experiences in trying to introduce system dynamics to the other social sciences. I am encouraged by his attempts, and impressed by his candor. The rejection of contributions to the journals of the various social sciences is something many have experienced, even while articles by their own members embrace