A Predictive Analysis Approach to Adaptive Testing [electronic resource] / Levent Kirisci and Tse-Chi Hsu.
The predictive analysis approach to adaptive testing originated in the idea of statistical predictive analysis suggested by J. Aitchison and I.R. Dunsmore (1975). The adaptive testing model proposed is based on parameter-free predictive distribution. Aitchison and Dunsmore define statistical predict...
Saved in:
Online Access: |
Full Text (via ERIC) |
---|---|
Main Author: | |
Other Authors: | |
Format: | Electronic eBook |
Language: | English |
Published: |
[S.l.] :
Distributed by ERIC Clearinghouse,
1988.
|
Subjects: |
Summary: | The predictive analysis approach to adaptive testing originated in the idea of statistical predictive analysis suggested by J. Aitchison and I.R. Dunsmore (1975). The adaptive testing model proposed is based on parameter-free predictive distribution. Aitchison and Dunsmore define statistical prediction analysis as the use of data obtained from an informative experiment in the past to make some reasonable statement about the outcome of the future experiment. Use of the approach's predictive density function and item selection procedure and terminating criteria is discussed. A small-scale exploration study compared the approach with A. Wald's sequential probability ratio test (1947), M. F. Lord's flexi-level test (1971), R. J. Owen's Bayesian strategy (1975), and F. C. Samejima's maximum likelihood strategy (1977). The various approaches could not be placed on equal base in terms of data used. Results indicate that: (1) final predictive probabilities were significantly correlated with total scores; (2) the predicted adaptive testing performed better than sequential probability testing and almost as well as the Bayesian strategy in the area of mastery classification; (3) the benefit of adaptive testing could not be demonstrated in the area of the number of test items required; and (4) there was no effect on the number of misclassifications of students when different priors were used in predictive testing. Three tables are included. (TJH) |
---|---|
Item Description: | ERIC Document Number: ED295982. ERIC Note: Paper presented at the Annual Meeting of the American Educational Research Association (New Orleans, LA, April 5-9, 1988). |
Physical Description: | 12 p. |