Laws, Causality and Statistics: Positivism, Interpretivism and Realism

Abstract

At the end of his Introduction to this symposium, after highlighting some specific problems that arise in attempting either to generate causal laws by means of the statistical manipulation of social data or to test theories statistically, Stephen Turner sets a basic question: what relations hold between laws, causality and statistics? My answer is that the view one takes of these relations is fundamentally influenced by one's conception of the nature of the sociological enterprise. To justify this answer, I will explore some general issues about statistics and about laws in order to locate the statistical sociologists' specific problems within wider debates about the nature of sociology. As a preliminary, it is as well to remember that the term statistics refers to a number of distinct ideas (Halfpenny, 1982: chap. 2). One distinction is between descriptive statistics and inductive statistics. Descriptive statistics refers to the measurement and enumeration of the characteristics of populations. Inductive statistics refers to methods used to infer from