NotesFAQContact Us
Search Tips
ERIC Number: ED314423
Record Type: Non-Journal
Publication Date: 1989-Mar
Pages: 15
Abstractor: N/A
Reference Count: N/A
A Comparative Analysis of Achievement Gains and Related School Characteristics.
Huff, Kimberley; And Others
This study examined the differences between actual and predicted school achievement on a grade 5 standardized reading posttest. The primary objective was to identify which schools are performing above expectation on a posttest by examining multiple regression residuals and, concurrently, which school level variables are associated with that exceptional achievement. The data consisted of pretest and posttest scores on the Degrees of Reading Power test for almost 1300 fifth grade students enrolled in 20 schools in a Community School District in New York City, or about 86% of the total enrolled. Pretest scores were from an administration of the test in the spring of 1984 and the posttest scores were from a second administration in the spring of 1986. The dependent variables were the gain score adjusted by analysis of covariance (ANCOVA), posttest scores adjusted by the method of adjusted averages, and the unadjusted mean gain score. Two school level variables--poverty index and percent of limited English proficient students--were significant. Three schools had large enough residuals to be considered extraordinary in their achievement gains. Gain scores adjusted by ANCOVA had little effect on controlling pretest differences among schools because of the lack of a pooled linear relationship between pretest scores and gain scores. The method of adjusted averages, however, requires no such relationship and represents an alternative for comparing achievement gains while controlling for pretest performance. Four data tables are provided. (Author/TJH)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A