Sage Journals: Discover world-class research

Abstract

Von Davier, Holland, and Thayer (2004) laid out a five-step framework of test equating that can be applied to various data collection designs and equating methods. In the continuization step, they presented an adjusted Gaussian kernel method that preserves the first two moments. This article proposes an alternative continuization method that directly uses the log-linear function from the smoothing step. This alternative method has the advantages of being simple and direct and producing smoother continuous distributions. When the number of score points is large, it can better preserve the moments. Two real test examples are provided to demonstrate these properties and to compare equating functions based on different methods. The standard error of equating for this method is derived.

Keywords

test equating kernel equating method continuized log-linear method log-linear smoothing

Get full access to this article

View all access options for this article.

References

Holland, P.W. , King, B.F. , & Thayer, D.T. (1989). The standard error of equating for the kernel method of equating score distributions (Technical Report No. 89-83). Princeton, NJ: Educational Testing Service.

Holland, P.W. , & Thayer, D.T. (1987). Notes on the use of the log-linear models for fitting discrete probability distributions (Technical Report No. 87-79). Princeton, NJ: Educational Testing Service.

Holland, P.W. , & Thayer, D.T. (2000). Univariate and bivariate loglinear models for discrete test score distributions. Journal of Educational and Behavioral Statistics, 25, 133-183.

Kolen, M.J. , & Brennan, R.L. (2004). Test equating, scaling, and linking: Methods and practices (2nd ed.). New York : Springer-Verlag.

Thisted, R.A. (1988). Elements of statistical computing: Numerical computation . New York: Champman & Hall /CRC.

von Davier, A.A. , Holland, P.W. , & Thayer, D.T. (2004). The kernel method of test equating. New York: Springer-Verlag.