ERIC Number: ED375147
Record Type: Non-Journal
Publication Date: 1993-Nov
A Study of Various Statistical Analyses Applied to School Report Cards.
Bobbett, Gordon C.; French, Russell L.
As the latest in a series of studies analyzing the data reported on Tennessee's school district report cards, this paper focuses on finding a common statistical treatment to examine the impact of predictor variables on dependent variables. Investigations of the 1988-89 report card data explored the relationships among eight school district variables and the relationship between each variable and the average student test scores at the school district level. In 1990-91, Tennessee began using the Tennessee Comprehensive Assessment Program (TCAP) outcome measures, making comparisons of current data with the 1988-89 data impossible. The following four statistical treatments were compared for their usefulness in analyzing the data: (1) Pearson product moment correlation; (2) Guttman's partial correlation; (3) stepwise regression; and (4) multiple regression. Independent variables do not have the same impact on student outcome at all outcome levels using the four methods, and it was evident that choice of method has an impact on study conclusions. The contribution of each variable is considered. Four tables in the text and eight presented as appendixes present analysis results. (Contains 9 references.) (SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A
Identifiers - Location: Tennessee