Methodology is suggested for several statistical procedures to detect polytomously scored items that function differently for two subgroups of examinees. The 3 methods are alternative ways of combining the data from 2 x "k" tables. Simulation results demonstrate the superiority of two of the methods, designated HW1 and HW3. (SLD)

Describes an automated scoring algorithm for a computer-based simulation examination of physicians' patient-management skills. Results with 280 medical students show that scores produced using this algorithm are highly correlated to actual clinician ratings. Scores were also effective in discriminating between case performance judged passing or…

Guidelines for developing performance assessments in science education allied with current research and reform are presented. The guidelines are applied to 3 hands-on science investigations performed by over 300 fifth and sixth graders and scored by science educators. Although difficult to develop, such assessments can be scored reliably. (SLD)