Grouping is a statistical procedure through which members of the same group are considered as a single unit of observation. There are methodological and inferential problems associated with various grouping procedures in various settings. This extensive paper focuses on making inferences about individuals when the analysis uses data that is…

Argues that educational researchers analyzing multilevel data must identify appropriate sets of research questions at and within various levels of analysis and must specify appropriate models for analyzing the data. Covers problems, terminology, and measurement and statistical issues associated with multilevel research. (PGD)

Problems associated with the analysis and interpretation of aggregated data in school effects research are discussed. The multilevel character of school effects research is described. Data from the IEA Six-Subjects Survey of Science Achievement in Australia are used to illustrate the differences in between-student and between-school models of the…

Two specific methods of analysis in large-scale evaluations are considered: structural equation modeling and selection modeling/analysis of non-equivalent control group designs. Their utility in large-scale educational program evaluation is discussed. The examination of these methodological developments indicates how people (evaluators,…

Three topics within the domain of regression-based analyses of multilevel data from quasi-experiments and field studies in educational research and evaluation are presented. The paper begins with a discussion of the general question of choice of unit of analysis, or what may be the more appropriate question of choice of analytical model. It then…

The activities of the Multilevel Analysis Project of the Center for the Study of Evaluation from February 1 to June 1, 1978 are summarized. Twelve project staff activities are listed, including papers on alternative analytical models for identifying educational effects, conference presentations, and grant proposals. A report on empirical…

Means of interpreting norm-referenced tests to lead to more accurate reporting results are discussed, with particular emphasis on state-level and district-level data. Suggestions fall into the categories of documentation, frequency norming, and multiple form use. (TJH)

This discussion of how the multilevel character of data should influence analyses of the determinants of educational achievement is grounded in the context of planning explanatory analyses for the longitudinal Second International Mathematics Study (SIMS). The intention is to describe how a multilevel perspective in the specification of…

Since problems associated with the statistical methodology of educational research are becoming increasingly important, this paper examines a subset of problems associated with the analysis and interpretation of aggregated data. Two major questions arise: (1) if a researcher knows the level (e.g., individual, teacher/classroom, school, school…

The purpose of this report is to comment on practice in the use of existing data bases in program evaluation and school improvement and to explore directions of increased and improved use. The report's impetus is the role that existing data plays in current efforts in local school districts and the possibilities for the future. To a great extent,…

This is one of a series of reports based on an ongoing reality test of systemic evaluation for instructional decision making. This feasibility study is being carried out by the Center for the Study of Evaluation with the Laboratory in School and Community Relations at a suburban Los Angeles high school (called Site A). Viewing a school as a…

Multiple sources of complications in the analysis of multilevel educational data are described with emphasis on the determination of a method of analysis that will provide accurate estimates of teacher/class effects even when there are systematic differences in within-class regressions of outcome on input associated with teacher/class…

The author contends that the secondary analyst may have a different perspective and theoretical persuasion than the original investigator, may have the advantage of more current theory and practices than those originally available, may move beyond the substantive and methodological limitations of the original investigation, and may not be faced by…

Research on school reform efforts and reform evaluation and the recognition that individual students are in constant transition and are members of multiple groups whose dynamics differ over the duration of schooling were examined. A conceptual framework for developing design and analysis strategies for investigating educational reform efforts is…

This report summarizes results from research aimed at improving the quality of information collected about school curriculum. The research sought to design and pilot a model for collecting benchmark data on school coursework. These more indepth data, such as course textbooks, assignments, exams, and teacher logs, can serve as anchors against which…