To determine the degree of association of age at school entry with reading failure, the Basic Skills Assessment Program (BSAP) reading test scores for all South Carolina students in Grades 1, 2, 3, and 6 were analyzed. The study found that the proportions of students failing to meet the standards on BSAP reading tests were higher for younger, male, black, and lunch-assisted students than for older, female, non-black, and full-paying students. Adjusted odds ratios from logistic regression indicated that risk of failure for younger students was greater than for older students in all grades when controlled for race, sex, and lunch-payment status (α = .05). However, the odds ratios as well as estimates of attributable risk showed that the effect of age at school entry, though significant through Grade 6, was minor compared to the risk associated with race, sex, or lunch-payment status.
This article has two objectives. The first is to present necessary and sufficient conditions for the validity of traditional within-subject F tests in repeated measures designs. It is shown that the Mauchly sphericity criterion (W) and possibly the Box test for the equality of covariance matrices are appropriate to judge the validity of these conditions. Valid applications of both tests are conducted on sets of orthogonal normalized variables that are associated with each cluster of within-subject mean square ratios. The second objective of the article is to present empirical results on the appropriateness of using the W criterion when the variates are not normally distributed. For light-tailed distributions, the W criterion was shown to be moderately conservative, whereas for heavy-tailed distributions, empirical Type I error rates exceeded nominal alpha. Since most social science applications typically involve light-tailed rather than heavy-tailed distributions, the W criterion should provide useful results in most cases.Traditional univariate analyses of variance for repeated measures (or mixed model) designs are used extensively in educational and psychological research (Kirk, 1968;Winer, 1971). In most situations observations are made on each subject in the sample under each combination of conditions (the within-subject factors). For example, a developmental psychology study may require measurements at several time intervals, with alternate forms used at each point in time. Thus, Time and Form would constitute two within-subject factors. If no factors that differentiate subjects are included, the design implies the testing of three within-subject hypotheses dealing with the effects of Time, Form, and the Time X Form interaction. Corresponding to each of these effects is a (nonunique) set of orthogonal normalized (orthonormal) variables.On the other hand, if the subjects are stratified on one or more independent (between-subjects) factors, the interaction of these factors and the repeated measures fac-Requests for reprints should be sent to Huynh
School effectiveness indices (SEIs) based on regressing achievement test performance onto earlier test performance and a socioeconomic status (SES) measure were obtained for eight subject–grade level combinations for a large sample of elementary schools. School means based on longitudinally matched student scores comprised the data set used in the analysis. The resulting SEIs were found to be somewhat unstable across subject areas (reading and mathematics) and very unstable across grade levels (1 through 4). Grade‐to‐grade correlations of the SEIs measuring mathematics performance, although small, tended to be statistically significant, whereas those measuring reading performance were generally nonsignificant. Thus, school effects may be more readily discernible in some subject areas than in others. Implications for research on effective schools and for the design of school recognition programs based on student test performance are discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.