Date of Award
4-1983
Degree Name
Doctor of Education
Department
Educational Leadership
First Advisor
Dr. Uldis Smidchens
Second Advisor
Dr. Carol Sheffer
Third Advisor
Dr. Tom Carey
Abstract
The Title I Evaluation and Reporting System (TIERS) consists of three basic evaluation models, the norm-referenced model, the control group model, and the regression model. The most widely used of these is the norm-referenced model. The equipercentile assumption, upon which this model is based, states that student groups, in the absence of Title I services, will maintain their percentile status over time. The validity of this assumption was tested using a data base of students who were eligible for Title I services but did not receive such services. The pre-and post-test reading and math scores of these students were aggregated and analyzed to determine to what extent average group scores change over the course of one year. Due to difficulties in identifying Title I-eligible students and the fact that not all students had both pre-and post-test scores, the data base was too small to perform analyses on grade level and group size variables. However, analyses on the overall math and reading data indicate the equipercentile assumption may not be valid for Title I students. In the absence of Title I services, some student groups increased their percentile status while other groups lost ground on their percentile status.
Access Setting
Dissertation-Open Access
Recommended Citation
Ameen, Christine Applin, "A Test of the Validity of the Equipercentile Assumption: An Analysis Using a Data Base of Title-I Eligible Students" (1983). Dissertations. 2431.
https://scholarworks.wmich.edu/dissertations/2431