Miriam Feigenbaum

Observed Score Equating Using a Mini-Version Anchor and an Anchor with Less Spread of Difficulty: A Comparison Study:

Two different types of anchors are investigated in this study: a mini-version anchor and an anchor that has a less spread of difficulty than the tests to be equated. The latter is referred to as a midi anchor. The impact of these two different types of anchors on observed score equating are evaluated and compared with respect to systematic error (bias), random equating error (SEE), and total equating error (RMSE) using SAT operational data. The results suggest that the overall bias, SEE, and RMSE when the midi anchor is used are either smaller than or very similar to those when the mini anchor test is used. The findings suggest that a midi anchor test would be preferred to a mini anchor test if equating accuracy at the ends of the score scale is not a primary concern.