ABSTRACT
This chapter argues that efforts to identify the antecedents of the achievement gap in reading and to assess the effectiveness of programs, policies, and practices that seek to reduce this gap can be enhanced by studying differences in variability alongside mean differences. It considers the term race with five categories (American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or Other Pacific Islander, White), and considers Black to represent Black or African American. The chapter adopts the traditional definition of the achievement gap in reading as differences in test scores between Black and White students, but departs from tradition by adopting the Hedges and Nowell definition of the achievement gap as a difference in score distributions for Black and White students. Such differences could take the form of means but also includes variances. Statistical procedures for estimating the variability between samples are illustrated using simulated reading scores via regression analyses.
