Abstract
This chapter describes the kernel equating framework. The five steps that characterize kernel equating are illustrated using the Math20EG, Math20SG, CBdata, and KB36 data sets that were introduced in Chap. 2 and which have been previously analyzed in the literature. We also illustrate the methods using the ADM admissions test data set. The R packages kequate (Andersson et al., J Stat Softw 55(6):1–25, 2013) and SNSequate (González, J Stat Softw 59(7), 1–30, 2014) are used throughout the chapter.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Further details about \(\boldsymbol{C}\) matrices are given in Appendix B.2.
- 2.
Negative score values can arise for instance when in a multiple choice test, a fraction of the total score in the test is discounted for each wrong answer in the test.
- 3.
Note that this is only true for the EG design. For other designs, the vector of score probabilities has to be further transformed using the design functions in order to obtain \(\hat{r}_{j}\) and \(\hat{s}_{k}\).
- 4.
Although variances instead of standard deviations are plotted in the right panel of Fig. 4.5.
- 5.
It might be possible to obtain estimated score probabilities from fitted frequencies for designs different than the EG, but additional steps might be necessary (i.e., the DF should be explicitly programmed and applied).
- 6.
Estimated score probabilities are obtained by making a call to the loglin.smooth() function.
- 7.
Values from lines 6 to 75 have been omitted in order to save space.
- 8.
The quantity inside the norm sign in Eq. (4.15) is called an SEE vector.
- 9.
Values from lines 6–75 have been omitted to save space.
References
Akaike, M. (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19, 716–723.
Andersson, B., & von Davier, A. A. (2014). Improving the bandwidth selection in kernel equating. Journal of Educational Measurement, 51(3), 223–238.
Andersson, B., Bränberg, K., & Wiberg, M. (2013). Performing the kernel method of test equating with the package kequate. Journal of Statistical Software, 55(6), 1–25.
Attali, D. (2016). ggExtra: Add marginal histograms to ‘ggplot2’, and more ‘ggplot2’ enhancements. R package version 0.3.4.
Bishop, Y., Fienberg, S., & Holland, P. (1975). Discrete multivariate analysis: Theory and practice. Cambridge, MA: MIT Press.
Cid, J. A., & von Davier, A. A. (2015). Examining potential boundary bias effects in kernel smoothing on equating: An introduction for the adaptive and Epanechnikov kernels. Applied Psychological Measurement, 39(3), 208–222.
González, J. (2014). SNSequate: Standard and nonstandard statistical models and methods for test equating. Journal of Statistical Software, 59(7), 1–30.
González, J., & von Davier, A. A. (2017). An illustration of the Epanechnikov and adaptive continuization methods in kernel equating. In L. A. van der Ark, M. Wiberg, S. A. Culpeppe, J. A. Douglas, & W.-C. Wang (Eds.), Quantitative psychology – 81st annual meeting of the psychometric society, Asheville, North Carolina, 2016. New York: Springer.
Häggström, J., & Wiberg, M. (2014). Optimal bandwidth selection in observed-score kernel equating. Journal of Educational Measurement, 51(2), 201–211.
Hall, P., Marron, J., & Park, B. U. (1992). Smoothed cross-validation. Probability Theory and Related Fields, 92(1), 1–20.
Holland, P., & Thayer, D. (1989). The kernel method of equating score distributions (Technical report). Princeton, NJ: Educational Testing Service.
Jiang, Y., von Davier, A. A., & Chen, H. (2012). Evaluating equating results: Percent relative error for chained kernel equating. Journal of Educational Measurement, 49(1), 39–58.
Johnson, D. H., & Sinanovic, S. (2000). Symmetrizing the Kullback-Leibler distance (Technical report). IEEE Transactions on Information Theory.
Kendall, M., & Stuart, A. (1997). The advanced theory of statistics (4th ed., Vol. 1). London: Griffin.
Lee, Y., & von Davier, A. (2011). Equating through alternative kernels. In A. von Davier (Ed.), Statistical models for test equating, scaling, and linking (Vol. 1, pp. 159–173). New York: Springer.
Lehmann, E. L. (1999). Elements of large-sample theory. New York: Springer.
Liang, T., & von Davier, A. A. (2014). Cross-validation: An alternative bandwidth-selection method in kernel equating. Applied Psychological Measurement, 38(4), 281–295.
Moses, T., & Holland, P. (2009). Selection strategies for univariate loglinear smoothing models and their effect on equating function accuracy. Journal of Educational Measurement, 46, 159–176.
Moses, T., & von Davier, A. (2005). Using PROC GENMOD for loglinear smoothing. Paper SA0605, 1–21.
Rao, C. R. (1973). Linear statistical inference and applications. New York: Wiley.
Silverman, B. (1986). Density estimation for statistics and data analysis (Vol. 3). London: Chapman and Hall.
von Davier, A. A. (2013). Observed-score equating: An overview. Psychometrika, 78(4), 605–623.
von Davier, A. A., Holland, P., & Thayer, D. (2004). The kernel method of test equating. New York: Springer.
Wiberg, M., & Bränberg, K. (2015). Kernel equating under the non-equivalent groups with covariates design. Applied Psychological Measurement, 39(5), 349–361.
Wiberg, M., & González, J. (2016). Statistical assessment of estimated transformations in observed-score equating. Journal of Educational Measurement, 53(1), 106–125.
Wickham, H. (2009). ggplot2: Elegant graphics for data analysis. New York: Springer.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this chapter
Cite this chapter
González, J., Wiberg, M. (2017). Kernel Equating. In: Applying Test Equating Methods. Methodology of Educational Measurement and Assessment. Springer, Cham. https://doi.org/10.1007/978-3-319-51824-4_4
Download citation
DOI: https://doi.org/10.1007/978-3-319-51824-4_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-51822-0
Online ISBN: 978-3-319-51824-4
eBook Packages: EducationEducation (R0)