NotesFAQContact Us
Search Tips
ERIC Number: ED174673
Record Type: Non-Journal
Publication Date: 1979-Apr
Pages: 10
Abstractor: N/A
Reference Count: N/A
The Equipercentile Assumption as a Pseudo-Control Group Estimate of Gain.
Powell, George; And Others
The tenability of the equipercentile growth assumption was investigated. This assumption is the basis of the norm referenced evaluation model (Model A) for Title I program evaluations, and supposes that a cohort of students, given no special educational intervention, will maintain the same percentile rank over the course of a school year, as measured on a standardized achievement test. Model A also assumes that a good estimate of the no-treatment expectation is the percentile corresponding to the average pretest score. Data were analyzed from reading and mathematics scores on the Metropolitan Achievement Tests; pre- and post tests were administered to elementary school students in grades two through four. For reading test results, the equipercentile growth method of estimation was found to be valid within the limits of the sample size. The comparisons between pretest and post test means for the mathematics scores, however, were significantly different from what would be expected using the equipercentile assumption. (GDC)
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Identifiers - Laws, Policies, & Programs: Elementary and Secondary Education Act Title I