Use of the Multiple Lens Approach to Multiple Regression Findings with a National Dataset

Abstract

This study builds upon the work of Nathans, Oswald, and Nimon (2012) to demonstrate how to apply variable importance metrics to an analysis of a national dataset. The paper (a) reviews the variable importance metrics; (b) details the strengths and weaknesses of national datasets as well as issues the analyst must confront when using them; (c) presents an integrated write-up of how to interpret variable importance metrics in analysis of fall of kindergarten math achievement from the Early Childhood Longitudinal Study – Kindergarten Cohort dataset; and (d) discusses how variable importance metrics inform the extant substantive literature on kindergarten math achievement.