Abstract:
|
The purpose of this study is to analyze the statistical power of the NAEP combined national and state samples in order to address the practical significance of results. Samples from the 2000 NAEP mathematics national and state assessments will be used in the analysis. This study will examine factors involved in merging NAEP national and state samples, such as: stratification, clustering, unequal weighting, and post-stratification.
This study will analyze several issues related to statistical power for combined samples. We will compare the effect sizes of the combined sample with those of the national sample, where the statistics covered in the study are comparisons of mean-scale scores and achievement-level percentages. The focus will be on comparing the statistical power of tests between current assessments and combined assessments, and evaluating the magnitude of the observed results with knowledge of statistical power for combined samples. In addition, we will inspect the relationship between statistical power and relative precisions for complex samples.
|