Frontiers in Psychology (Jul 2016)
Do our Means of Inquiry Match our Intentions?
Abstract
A key stage of the scientific method is the data analysis, yet despite the variety of methods that are available to researchers they are most frequently distilled to a model that focuses on the average relation between variables. Although research questions are conceived as ones of general inquiry, most regression methods are limited to comprehensively evaluate how observed behaviors are related to each other. Quantile regression is a largely unknown yet well-suited analytic technique similar to traditional regression analysis, but allows for a more systematic approach to understanding complex associations among observed phenomena in the psychological sciences. Data from the National Education Longitudinal Study of 1988/2000 are used to illustrate how quantile regression overcomes the limitations of average associations in linear regression by showing that psychological well-being and sex each differentially relate to reading achievement depending on one’s level of reading achievement.
Keywords