The following list of journal articles and books provide extended reading on topics covered in chapter 36 in the second edition. Please note that journal articles are free to access, whereas book extracts (denoted by methods.sagepub.com URLs) require your university to have a subscription to SAGE Research Methods.
Hong, G., & Raudenbush, S. W. (2005). Effects of kindergarten retention policy on children’s cognitive growth in reading and mathematics. Educational Evaluation and Policy Analysis, 27(3), 205-224.
In this article, Hong and Raudenbush use multilevel models in analyzing data from ECLS to investigate the causal effects of retention in Kindergarten on students’ subsequent academic achievement. An important contribution and feature of this article is that it extends key concepts and techniques in the causal inference literature (e.g., propensity score matching and sub-classification) to settings that involve the analysis of multilevel data.
Rumburger, R. W., & Palardy, G. J. (2004). Multilevel models for school effectiveness research. In D. Kaplan (Ed.). The handbook of quantitative methods for the social sciences (pp. 235-255), Thousand Oaks, CA: Sage.
This chapter provides a comprehensive overview of the use of multilevel models in research on school effectiveness involving the analysis of data from large-scale surveys such as NELS and HSB. The authors begin by presenting a valuable conceptual framework for investigating school effectiveness, and discussing key decisions that must be made regarding data and sample selection. They then take the reader through the logic and use of various types of multilevel models used in school effectiveness research, including models for investigating differences within and between schools in rates of change in student achievement, and models for binary outcomes (e.g., whether or not a student drops out of school).
Seltzer, M. (2004). The use of hierarchical models in analyzing data from experiments and quasi-experiments conducted in field settings. In D. Kaplan (Ed.), The handbook of quantitative methods for the social sciences (pp. 259-280). Thousand Oaks, CA: Sage.
This chapter focuses on the value of multilevel models in obtaining more appropriate standard errors in multisite studies of programs and interventions, and in investigating factors that might be critical to a program’s success. The chapter presents in-depth analyses of the data from the Transition Mathematics study, and from a study of reform-minded mathematics instruction. The importance of collecting data on implementation and the continual need to attend to possible confounding factors receive special emphasis.