The test misgrading is treated in this paper as a stochastic process. Three characteristic values, the expected number of misgradings, the expected inter-occurrence time of misgradings, and the expected waiting time for the nth misgrading, are discussed based on a simple Poisson model and a hierarchical Beta-Poisson model. Regular tests for classroom teaching and the Test of Written English (TWE) in the Test of English as a Foreign Language (TOEFL) are illustrated as examples in the model construction.
Get full access to this article
View all access options for this article.
References
1.
Bhat, N. U. (1984). Elements of applied stochastic processes (2nd ed). New York: Wiley.
2.
Braun, H. I. (1988, Spring). Understanding scoring reliability: Experiments in calibrating essay readers.Journal of Educational Statistics, pp. 1-18.
3.
Casella, G. and Berger, R. L. (1990). Statistical inference. Pacific Grove, CA: Brooks/Cole Publishing Company.
4.
Educational Testing Service (1992). Bulletin of information for TOEFLITWE and TSE. Princeton, NJ: Author.
5.
Lawson, A. E. and Staver, J. R. (1989). Toward a solution of the learning paradox: Emergent properties and neurological principles of constructivism.Instructional Science, 18, 169-177.
