NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED318784
Record Type: Non-Journal
Publication Date: 1990-Apr
Pages: 126
Abstractor: N/A
Reference Count: N/A
ISBN: N/A
ISSN: N/A
Current Norms: Do They "Deflate" Test Scores? A Study of Normative Trend Data for an Urban School District.
Lanese, James F.
A strategy that has been initiated to respond to assertions that out-of-date norms distort standardized achievement test results involves annually updating the norms for achievement tests to avoid the production of inflated scores through aging norms. The effect of the application of normative trend data to the obtained test results in an urban school district was evaluated. The analysis included data for the Spring 1988 administration of the California Achievement Test (CAT), Form E, in reading (n=54,871) and mathematics (n=17,722). The district participated, with a number of other schools, in CTB/McGraw Hill's Normative Trend Data (NTD) project, which involved renorming the CAT for that year. Data were then applied to the original results to transform the obtained frequency distributions for each subtest at each grade. Redistributed scores were then compared to the original distributions to assess the impact of the updated norms. The use of NTD data resulted in lowering the measure of achievement levels in reading and mathematics. NTD scores positively skewed the district's grade level achievement distributions. Annual national performance appeared to improve relative to the original standardization sample. Longitudinal comparisons for the district must, however, rely on the original standardization, and the interpretation and explanation of two sets of scores in a district compromise the utility of the results. An appendix contains 55 figures illustrating test score distributions. (SLD)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Identifiers - Assessments and Surveys: California Achievement Tests