All Score Points Are Not Created Equal: Anchor Set Composition for Licensure Examinations

• In Event: Roundtable Session 18
  In Roundtable Session: Methodological Advances in Equating

Abstract

Recent research has challenged the traditional method of creating anchor tests as “mini” tests with similar statistical properties and representative content. Instead, equating may be improved through the use of a “midi” anchor with similar mean item difficulty but less variance than the total test (Sinharay & Holland, 2007). The following study explores the performance of mini and midi anchors under a licensure/certification testing context where precision of measurement is meaningful at the cut score and high end of the score scale. Results indicate bias around the cut score to be similar, though differences at the upper end of the score scale suggest that the mini anchor may not be best positioned to support secondary uses of score information.

Authors

• Richard A. Feinberg, National Board of Medical Examiners

• Javarro Antoine Russell, Educational Testing Service