Explaining Variance: Rethinking Social Causation

580_meanvar.pdf296 KB

Abstract:

Political scientists typically conceive of causation purely in "mean–centric" terms: the statement "X causes Y" is taken to imply that an increase in the value of X changes the mean of the distribution of Y. This article challenges that point of view. Many independent variables of interest to students of politics have an effect by altering the variance, not the mean, of the distribution of the dependent variable. This form of causation is alien to methodology textbooks and greatly underappreciated in empirical work. Thinking about the causes of changing variance opens up a theoretical dimension that has heretofore been neglected; understanding the causes of changing variance can provide a more fine–grained empirical description of the political world. 

Last updated on 03/25/2015