ERIC Number: ED344944
Record Type: Non-Journal
Publication Date: 1991-Aug
Reference Count: N/A
Standard Errors of Linear Equating for the Single-Group Design.
Large sample standard errors of linear equating for the single-group design are derived without making the normality assumption. Two general methods based on the delta method of M. Kendall and A. Stuart (1977) are described. One method uses the exact partial derivatives, and the other uses numerical derivatives. Simulation using the beta-binomial model (500 replications) and real test data (a 150-item test divided into 2 half tests) are used to evaluate the behavior of the estimated standard errors. Comparisons with standard errors derived with the normality assumption and bootstrap method are also conducted. The results indicate that the standard errors derived in this paper agree with those computed by the bootstrap method and are more accurate than the standard errors based on the normality assumption, especially in the situation in which the score distributions are skewed. Five tables present results of analyses. There is a 13-item list of references. (Author/SLD)
Descriptors: Comparative Analysis, Computer Simulation, Equated Scores, Equations (Mathematics), Error of Measurement, Estimation (Mathematics), Higher Education, Mathematical Models, Research Design, Sample Size, Statistical Distributions
American College Testing Research Report Series, P.O. Box 168, Iowa City, IA 52243.
Publication Type: Reports - Evaluative
Education Level: N/A
Authoring Institution: American Coll. Testing Program, Iowa City, IA.