Evaluation and Research in Education
Editor: Professor Keith Morrison, Inter-University Institute of Macau
Associate Editor: Professor Stephen Gorard, University of York
Statistical Adviser: Professor Colin Baker, University of Wales Bangor
Reviews Editor: Dr. Emma Smith, University of York
|
Volume: 21 Number: 1 Page: 18–42
doi:10.2167/eri418.0
|
|
|
|
|
Using Multi-level Analyses to Study the Effectiveness of Science Curriculum Materials
|
Vasuki Rethinam1, Curtis Pyke2 and Sharon Lynch2
1, Montgomery County Public Schools, USA and 2Graduate School of Education and Human Development, The George Washington University, USA
|
|
This paper explores the use of HLM analysis in determining the implementation effects of two science curriculum units on student learning. It also compares HLM results with prior results from ANCOVA analyses. HLM analyses were considered as an alternative to ANCOVA because student data were nested within classrooms. The data for this study are from the Scaling-up Curriculum for Achievement, Learning and Equity Project (SCALE-uP). The sample consists of Grade 6 and Grade 7 students from five matched pairs of middle schools in a large, diverse, metropolitan school district in the US. The HLM null model indicates that approximately 15% of the total variance in student gain scores was found between classrooms. Curriculum treated as a classroom factor was significant for both units tested, as was also the case when treated at the individual level in ANCOVA. However, there were large differences in the effect sizes reported by the two analysis techniques. A major implication is that when the students are nested in classrooms and schools, researchers should consider multi-level analysis and account for classroom/school contexts.
Keywords: curriculum effectiveness, multi-level versus single-level analysis, HLM, classroom context, science education, gain scores
© 2008 V. Rethinam et al.
Access this article
|
