A Comparison of Combined Scoring and Standard-setting approaches applied to an Interim Assessment in Predicting Statewide Assessment Performance

• In Event: AER Roundtable Session Friday 1:45 pm - Gold 4
  In Roundtable Session: Innovations in Item Response Theory for Modeling Student Performance and Growth

Abstract

This study examines the combined utility of various scoring (number correct, two-parameter logistic, and nominal response model) and standard setting methods in accurately predicting performance on an external assessment. The study highlights the implementation of the embedded standard setting method, a newer standard setting procedure that is not yet well represented in the literature. Embedded standard setting will be compared to a variation of the bookmark method and to variation of norm-referenced standard setting, the equi-distribution method. The results of the study suggest that each of the combined approach performance levels, evaluated using multilevel logistic regression analyses, were statistically significant predictors of performance on the statewide assessment as defined by whether a student passed or did not pass the assessment.

Authors

• Michelle Paul, Orange County Public School

• Stephen A. Sivo, University of Central Florida

• Parul Acharya, Columbus State University