This article addresses three questions: Does reduced class size cause higher academic achievement in reading, mathematics, listening, and word recognition skills? If it does, how large are these effects? Does the magnitude of such effects vary significantly across schools? The authors analyze data from Tennessee's Student/Teacher Achievement Ratio…

In organizational studies involving multiple levels, the association between a covariate and an outcome often differs at different levels of aggregation, giving rise to widespread interest in "contextual effects models." Such models partition the regression into within- and between-cluster components. The conventional approach uses each cluster's…

The authors propose a strategy for studying the effects of time-varying instructional treatments on repeatedly observed student achievement. This approach responds to three challenges: (a) The yearly reallocation of students to classrooms and teachers creates a complex structure of dependence among responses; (b) a child's learning outcome under a…

The question of how to estimate school and teacher contributions to student learning is fundamental to educational policy and practice, and the three thoughtful articles in this issue represent a major advance. The current level of public confusion about these issues is so severe and the consequences for schooling so great that it is a big relief…

Studied the efficiency and robustness of alternative estimators of regression coefficients for three-level data. A simulation study shows that, as expected, the hierarchical model analyses produced more efficient point estimates than did analyses that ignored the covariance structure in the data, even when the normality assumption was violated.…

Uses data from the Trial State Assessment of the National Assessment of Educational Progress to describe and illustrate a two-stage statistical model for investigating state-to-state variation in mathematics achievement. Results reveal considerable state-to-state heterogeneity in mathematics proficiency, but most heterogeneity is explainable on…

Gibbs sampling was applied to obtain Bayes inferences in the case of unbalanced multilevel data when the homogeneity of variance assumption fails and when interest focuses on inferences for some or all of the groups' variances. This approach is compared to a more standard analysis based on restricted maximum-likelihood statistics. (SLD)

The specification and estimation of school effects, the variability of effects across schools, and the proportion of variation in student outcomes attributable to differences in school context and practice are considered. A statistical model is presented that defines school effects for parents choosing a school and for agencies evaluating school…