Social simulation is a new method for the social sciences that combines some of the advantages of statistics and qualitative research. The SIMIAN project has been funded over three years from September 2008 by the Economic and Social Research Council to develop expertise in simulation within UK social science.
The idea of simulation is to construct a computer program that has some of the properties of a ‘real world’ social process and observe what happens when the program runs. Depending on the level of detail and other factors, this can be thought of as a method of theory development (if you can specify a computer program corresponding to your theory, this can help to clarify just what the theory means); a method of theory checking (a program will only run if it is complete and consistent), a way of gaining an understanding of the implications of behaviour (for example, macro social structures and social features might ‘emerge’ from the modeled micro social interactions) or, in its most familiar form, a way of making predictions about what might happen under given circumstances.
What can simulation in the social sciences be used for? There is a rapidly growing body of research based on simulation, spanning the whole range of the social sciences. A few examples are: in political science, studies of the effect of the inter-personal transmission of opinions; in economics, studies of stock markets and other types of market; in sociology, research on social movements; in geography, work on optimal location; in linguistics, research on the evolution of language; in demography, studies of marriage and divorce; and many, many others. A sample of current work can be found in the online journal, JASSS, at http://jasss.soc.surrey.ac.uk/ .