THE KERNEL METHOD OF EQUATING SCORE DISTRIBUTIONS

Paul W. Holland, Dorothy T. Thayer

ETS Research Report Series · 1989 · 75 citations · 5 references

Concepts

Abstract

ABSTRACT A new and unified approach to test equating is described that is based on log‐linear models for smoothing score distributions and on the kernel method of non‐parametric density estimation. The new method contains both linear and standard equipercentile methods as special cases and can handle several important equating data collection designs. An example is used to illustrate the new method for the random groups and external anchor‐test designs.

References

5