ERIC Number: ED468074
Record Type: Non-Journal
Publication Date: 2002-Jul
Reference Count: N/A
Using Multiple Measures To Evaluate the Performance of Students and Schools: Learning from the Cases of Kentucky and Maine. Research Report. Statewide Systemic Initiatives (SSI) Study.
Lee, Jaekyung; Coladarci, Theodore
This report is the product of the first year of study of the Statewide Systemic Initiatives (SSI) Study on exploring data and methods to assess and understand the performance of states participating in the SSI. The report describes three sets of studies. The first studied valid and reliable measures of student achievement, and how to combine multiple measures of achievement, as well as how results of certification based on a single-measure compare with those based on a multimeasure analysis. The study addressed these questions through analyses of state and local data from two sites in Maine. The second study focused on valid and reliable ways to estimate school performance and identify value-added school effects on student achievement, as well as how the results of school evaluation based on a unilevel analysis compare with those based on a multilevel analysis. The study addressed these questions through multilevel analyses of National Assessment of Educational Progress assessment data in Maine and Kentucky. Results suggest that school-level compositional effects and student-level background characteristics need to be taken into account for fair evaluation of school performance. The third study explored ways to estimate academic progress and how to identify true achievement gains without statistical artifacts. These questions were addressed through the analysis of Maine Educational Assessment and Kentucky Instructional Results Information System assessment data. Results show that to set realistic expectations about schools academic progress, past trends need to be taken into account, and that combining multiple years of data improves the reliability of school performance measures. (Contains 24 tables and 28 references.) (SLD)
Publication Type: Numerical/Quantitative Data; Reports - Research
Education Level: N/A
Sponsor: National Science Foundation, Arlington, VA.
Authoring Institution: Maine Univ., Orono.
Identifiers - Location: Kentucky; Maine