Abstract
No one has applied Computerized Adaptive Testing (CAT) to Student Evaluation of Teaching (SET). We introduce three algorithms for CAT of SET (Marginal, Joint, Iterative) and evaluate them via simulations. The Iterative method showed the highest reliability, lowest RMSE, and little bias. Joint or Marginal were worse. Random was the worst. Copyright © 2019 NCME conference.
Original language | English |
---|---|
Publication status | Published - Apr 2019 |