ERIC Number: ED464955
Record Type: Non-Journal
Publication Date: 2002-Apr
Reference Count: N/A
How Adequate Are Adequate Yearly Progress (AYP) Measures? Regression Artifacts and School AYP.
Lee, Jaekyung; Coladarci, Theodore
This study explored the extent to which current adequate yearly progress (AYP) indices reflect real change or a statistical artifact. The objective, in more specific terms, was to investigate regression artifacts in evaluating schools' academic progress and explore methods to overcome their effects in developing AYP measures. Schools in Kentucky and Maine were studied, using eighth-grade achievement data from the two states' student assessments, the Kentucky Instructional Results Information System and the Maine Educational Assessment. Three different types of school achievement gain estimates were computed: (1) the 1-year gain measure; (2) a 2-year gain measure; and (3) a 3-year gain measure. Time-reversed analyses were conducted to determine whether gain score really was a regression artifact. Findings for both states show that higher performing schools tend to gain less while lower performing schools gain more. This illustrates the well-known regression to the mean status phenomenon. Results also reveal regression to the mean growth phenomenon. Schools that gained more in the past tend to gain relatively less. The first force tends to make higher and lower performing schools appear convergent in their status, and the latter force may make more and less improving schools appear convergent in their growth. These two forces as statistical artifacts may confound AYP measures and need to be addressed. (Contains 4 figures, 11 tables, and 9 references.) (SLD)
Publication Type: Numerical/Quantitative Data; Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Sponsor: National Science Foundation, Arlington, VA.
Authoring Institution: N/A
Identifiers - Location: Kentucky; Maine