The Use and Non-Use of Experiments in the Social Sciences

Along with the call to test the causality of our theoretical claims goes a trend to use experimental methods in the social sciences, which have traditionally relied more on observational studies. Experiments allow to "hold everything else constant" and hence provide a concrete test situation for a theoretical claim, which is often called "internal validity". However, to hold everything else constant is more difficult than it seems and also poses the question whether results of experimental research can be generalized to everyday life, which is often called "external validity". Since social science is concerned about everyday life we shall discuss the use of experimental methods and whether the trade-off in gaining internal in exchange for external validity is worthwhile.

Resources:
McDermott, Rose (2002): “Experimental Methods in Political Science.”
In: Annual Review of Political Science, Vol. 5, pp. 31-61.

Wilson, Rick K. (2011): “The Contribution of Behavioral Economics to
Political Science.” In: Annual Review of Political Science, Vol. 14, No. 1, pp 201-223.