ERIC Number: ED389710
Record Type: Non-Journal
Publication Date: 1988-May
Reference Count: N/A
Scale Drift in On-Line Calibration.
Stocking, Martha L.
Recent advances in psychometrics and computer technology encourage the development of model-based methods of individualized testing on a microcomputer, where each examinee receives short tests and where the number of pretest items that can be administered is severely restricted. On-line (i.e., data is collected on operational equipment) methods for calibrating pretest items in this setting face new challenges. In the context of adaptive testing (CAT), two LOGIST-based methods of on-line calibration that were developed are described. These two methods were applied and compared in a simulation study using data from the CAT Armed Services Vocational Aptitude Battery. One of the methods was superior to the other in terms of statistical properties, but the superior method was more expensive to implement in terms of the numbers of items required and either the numbers of examinees or numbers of items per examinee required. Both methods showed scale drift, although the drift was in opposite directions for the two methods. Ways to reduce scale drift are explored. An appendix contains 4 tables and 22 figures, and 17 references are attached. (Author/SLD)
Publication Type: Reports - Evaluative
Education Level: N/A
Sponsor: Office of Naval Research, Arlington, VA. Cognitive and Neural Sciences Div.
Authoring Institution: Educational Testing Service, Princeton, NJ.
Identifiers - Assessments and Surveys: Armed Services Vocational Aptitude Battery