Making Sociological Inferences by Quantifying Data:

Society and Social Science Shown by “New” Mathematical Methods

Abstract

Several recently introduced methods in the social sciences, such as statistical causal inference and Bayesian statistics, have had profound effects on the science of sociology as well as the analysis of effect size. These methods share some fundamental concepts with traditional methods, especially Weber's adequate causation and interpretive sociology, and can be applied not only to numerical data but also to discourse.

For example, statistical causal inference is a method that accurately estimates the average causal effect based on the assumption that causality varies at the individual level. By assuming the joint distribution of potential outcomes, causal candidates, and all covariates, we can define the adequate causation more accurately and consistently. Thus, theoretically, this method can be applied to a one-time-only phenomenon. Additionally, interpretive sociology can be reformulated as a Bayesian update, which modifies subjective and hypothetical assumptions through iterative process applying objective and observable data.

These methods have both factual and assumptive aspects; therefore, the sociological phenomena inferred through these methods also have subjective and objective aspects.