ERIC Number: ED185118
Record Type: RIE
Publication Date: 1980-Jan
Reference Count: 0
ESEA Title I Linking Project. Final Report.
Holmes, Susan E.
The Rasch model for test score equating was compared with three other equating procedures as methods for implementing the norm referenced method (RMC Model A) of evaluating ESEA Title I projects. The Rasch model and its theoretical limitations were described. The three other equating methods used were: linear observed score equating, linear true score equating, and equipercentile equating. Data used in the study were obtained by testing 6,000 students in grades three through eight with the norm-referenced California Achievement Tests, Form C (CAT/C), reading and mathematics, and with tests made up from items from two Rasch calibrated item banks in reading and mathematics. The results of the four equating procedures were compared with actual observed score distributions on the CAT/C for each item bank raw score. All four equating techniques yielded similar results with the least accuracy at the low end of the score distributions. It appeared that use of the Rasch model was feasible, but the accuracy of the results it produced was questionable. (Author/CTM)
Publication Type: Reports - Research
Education Level: N/A
Sponsor: Office of Education (DHEW), Washington, DC.
Authoring Institution: Oregon State Dept. of Education, Salem.
Identifiers: California Achievement Tests; Elementary Secondary Education Act Title I; Northwest Evaluation Association; Rasch Model