Search

Variance heterogeneity is a common feature of educational data when treatment differences expressed through means are present, and often reflects a treatment by subject interaction with respect to an outcome variable. Identifying variables that account for this interaction can enhance understanding of whom a treatment does and does not benefit in ways that can inform and improve the treatment. Even in the absence of a treatment effect expressed through means studying variance heterogeneity offers insight into a treatment by identifying subject characteristics related to heterogeneity. This study illustrates four methods of modeling variance heterogeneity for data from a study of the impact of an engineering design-based STEM curriculum on student achievement with a focus on multilevel models.

Commercial data analysis software has been a fixture of quantitative analyses in education for more than three decades. Despite its apparent widespread use there is no formal evidence cataloging what software is used in educational research and educational statistics classes, by whom and for what purpose, and whether some programs should be recommended over others. This paper argues that the rise of the R data analysis software has intensified the need for rigorous evaluations of these programs to identify their strengths and weaknesses in ways that provide educators with guidance in choosing programs. Examples of research activities to produce a literature to guide these choices are described.

The relationship between high school mathematics curricula and the likelihood of students who enroll in a developmental (non-credit bearing) course in college taking additional mathematics courses was studied. The results showed that high school mathematics curriculum, years of high school mathematics completed, and ACT mathematics scores were related to developmental mathematics course-taking, but curriculum was not related to the subsequent mathematics course-taking of students who began college with developmental mathematics. The results have important implications for educational researchers and policymakers at the college and high school levels.

Meta-analytic methods were used to examine the moderating effect of institutional factors on the relationship between high school mathematics curricula and college mathematics course-taking and achievement from a sample of 32 colleges. The findings suggest that the impact of curriculum on college mathematics outcomes is not generally moderated by institutional characteristics such as selectivity and educational profile, providing evidence that the relationships between curriculum and college mathematics outcomes generalize to a range of colleges. The results inform college policies and practices for advising students on mathematics course-taking including enrollment in developmental courses, and high school mathematics curriculum selection. [Note: The title on the PDF is incorrect.]

A scale-up study in education typically expands the sample of students, schools, districts, and/or practices or materials used in smaller studies in ways that build in heterogeneity. Yet surprisingly little is known about the factors that promote successful scaling up efforts in education, in large part due to the absence of empirically supported theories of scaling up. A literature for scale-up studies in education is growing but is years away from providing research-supported practices in planning and conducting these studies. Following the suggestion of Schneider and McDonald (2006) to import relevant knowledge from other fields into the scale-up literature in education, this paper examines the multisite public health and nursing literature in search of a multidisciplinary knowledge base that can inform scaling up efforts in education. Five strategies and practices identified in these literatures as critical to scaling up success are described.

Multivariate and univariate analysis of variance methods (MANOVA and ANOVA, respectively) are compared for their relative value in educational research. The favoritism shown multivariate techniques is questioned. Criteria for the selection of the appropriate technique are outlined. The relationships among research hypotheses, statistical hypotheses, and underlying assumptions are crucial. (TJH)