Generalising Social Behaviour and Theory of Social Behaviour: When Is It Statistical and When Not?

Abstract

Researchers and philosophers interested in findings pertaining to social behaviour, or theory of social behaviour, are necessarily concerned with generalising findings, theory or both. There are statistical issues that are ignored at one's peril, pertaining to generalising from a sample to the population from which that sample was drawn. However, if the goal is to generalise to other populations, more conceptual issues come into play. Moreover, if the goal is to test a theory's ability to generalise or be useful for an applied goal, yet more conceptual issues come into play. The present aim is to clarify some of these issues, including relevant questions, so researchers and philosophers can better understand that although certain statistical issues are always relevant, there are many conceptual issues that are sometimes relevant and sometimes not. Those who are interested in social behaviour must necessarily be interested in generalising something, and so the issues discussed are ubiquitously germane.