Abstract
More and more judgments conducted by raters are involved in psychological and educational assessment and testing situations. Two measurement theories, CTT (classical test theory) and MFRM (many-facet Rasch model), were implemented in the analyses of rating scale data. Parameters of item difficulty, item discrimination, and reliability were compared between CTT and MFRM. One hundred and twenty-four applying proposals were evaluated by 64 raters on a 24-item rating form. The results showed that item difficulty indices estimated by CTT and by MFRM were highly consistent, but item discrimination indices were remarkably uncorrelated. The values of reliability provided by CTT and MFRM were very consistent.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
