Nye, B., Hedges, L. V., & Konstantopoulos, S. (2000). The effects of small classes on academic achievement: The results of the Tennessee class size experiment. American Educational Research Journal, 37(1), 123-151.
The research problem that was addressed in Nye, Hedges and Konstantopoulos' (2000) study was the determination of the most effective allocation of scarce educational resources. For most schools, one of the most critical decisions is to decide how to distribute instructional staff to the classes, which is in turn related to the determination of class size. More specifically, the research questions that were addressed in this research study were:
What are the effects of small class sizes on academic achievement?
How do the effects of small class sizes influence schools with diverse classes and teacher characteristics?
Data were collected from students from kindergarten to grade three, who were attending 79 elementary schools in 42 Tennessee school districts. In this long-term research study, kindergarten students were first randomly assigned to three conditions: small classes (13-17 students), large classes (22-26 students), or large classes with a classroom aide. Following the changes in the actual composition of the classes from kindergarten to grade 3, the attrition rate and the switching of the students from one class type to another, the researchers assessed the impact of class type on average reading and mathematics achievement test scores (Nye et al., 2000).
The data were analyzed by using a clustering or hierarchical structure in order to account for the variance in student achievement within schools, as well as between schools. In the within-school model, the variation in student achievement was measured in terms of the following variables: "student characteristicsà treatment group assignment, and the interaction of assignment with gender" (Nye et al., 2000, p. 137). The following variables wer...