Beyond the Average Case: The Mean Focus Fallacy of Standard Linear Regression and the Use of Quantile Regression for the Social Sciences

Abstract

This article outlines the limitations and fallacies of the routine use of standard linear regression.The method confines the analysis to the effect of the explanatory variable(s) on the mean of the dependent variable, and precludes the exploration of the various other ways in which the dependent variable responds to changes on the explanatory variable(s). The implications for the discovery and development of theories in the social sciences are discussed. Using the empirical example of the relationship between income and happiness, the article shows how quantile regression can help avoiding these fallacies and allow the researcher to address a broader range of research questions.

Keywords: quantile regression, mean focus fallacy, research methodology, happiness, life satisfaction, income, regression analysis

JEL Classification: C00