Date of Award


Document Type

Campus Access Dissertation


Educational Leadership and Policies


Educational Administration

First Advisor

Doyle Stevick

Second Advisor

Peter Moyi


This study raised the basic question about if the school size is associated with student academic achievement, in schools without the impact of nested factors. Thus, the main purpose of this study is to investigate the effects of nested factors, school compositional variables, in particular, family structure, family learning engagement, and school learning climates. This examines how the impact of school size on academic achievement is explained by the nested factors. School size studies, in recent, have reported more positive effects of smaller size schools on accountability, equity, school climates, and academic achievement and showed that there were not differences between smaller schools and larger schools in the matter of the advantages of economic effects and program comprehensiveness (Sergiovanni, 1995; Cotton, 1996; Lee & Smith, 1997; Bowen, Bowen, & Richman, 2000).

The data used in this study were longitudinal, nationally representative data, the Educational Longitudinal Study (ELS) 2002, from National Center for Educational Statistics (NCES). Student population 16,252 sampled was high school 10th graders of spring 2002. Surveyed questionnaires were about the perceptions of 10th graders, their parents, teachers, and administrators on school operations, functions, and others. This study measures the school-level (level 2) variables influencing the impact of school size on student academic achievement, reading and math, after controlling for individual students' background variables at student-level (level 1). The explanatory variables at school level (level 2) are school size, the level of concentration of students from non-intact families, the percentage of students with the high level of family engagement, the percentage of students and parents with high perceptions of negative school learning climates. The analysis techniques used was quantitative statistics; descriptive method, Z-scores, and multilevel analysis, Hierarchical Linear Modeling (HLM).

The findings indicate that schools with smaller size had positive impact on academic achievement when controlling for only individual students' background variables at student level (level 1). However, after taking into account of the nested variables at school level (level-2) along with individual students' background variables at student level (level 1), the positive effect of smaller schools on academic achievement were no longer statistically significant. In fact, the larger size of schools in the study indicated more positive outcome than smaller size of schools when controlling for both the nested factors at school level and individual students' background variables at student level. This indication proved that the nested factors did impact the negative effect to larger schools on school outcomes. For students attending schools with enrollment size more than 400 tenth graders, for instance, family learning engagement and school learning climates were particularly critical because larger size schools had higher concentration of students from single parent family who have less time for their child learning.

In conclusion, this study provides the policy makers, researchers, and educators with new ideas to look at the negative effect of larger size schools. The implications of the results may motivate educators to investigate the nested factors discouraging school success. That is, administrators may incline to hire more social workers, to find child role models, or tutors to help single mothers who do not have enough time to engage in their child learning activities. Policy makers and researchers may be more interested in making new policies or programs on students' class disciplines, attendance, and school safety to improve school performance.