![]() | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
OTHER CALENDARSABOUT THE CALENDARMORE RESOURCES |
Reporting of Subscores: A Review of the Status Quo and Some Future DirectionsSeminar | September 11 | 2-4 p.m. | Tolman Hall, 2515 Tolman Hall Sandip Sinharay, CTB/McGraw-Hill There is an increasing interest in subscores because of their potential diagnostic value. This presentation provides a critical evaluation of reporting of subscores in the context of educational tests. The presentation starts by raising questions about the quality of the currently reported subscores. A review is provided of a recent classical-test-theory-based method (Haberman, 2008) that examines whether subscores provide any added value over the total score. Analysis of data from several operational data sets shows that subscores have to be based on a sufficient number of items and have to be sufficiently distinct from each other to be worth reporting and that several operationally reported subscores actually do not have any added value over the total score. A discussion is provided of the extension of the method to subscores based on multidimensional item response theory models (Haberman & Sinharay, 2010) and to other contexts. The presentation concludes with some recommendations on reporting of subscores. melinda@berkeley.edu, 510-642-4206 |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
GSE | Contact Us | Sitemap
Copyright © 2013 UC Regents
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||