Stephen D Phipps1, Marcia L Brackbill. 1. Bernard J. Dunn School of Pharmacy, Shenandoah University, Nashville, TN 37204, USA. steve.phipps@lipscomb.edu
Abstract
OBJECTIVE: To evaluate the relationship between assessment item formats (case-based versus noncase-based) and item performance characteristics. METHODS: Assessment items (1,575) were collected from examinations administered in several therapeutics courses over 4 academic years. Items were categorized as either "case-based" or "noncase-based" and item performance characteristics (discrimination index and level of difficulty) were evaluated. RESULTS: Noncase-based items represented approximately three-fourths of all items that were evaluated, and demonstrated a higher discrimination index than case-based items. Case-based items were generally lengthier and included more detailed information than noncase-based items; however, they were not more difficult and exhibited a lower discrimination index. Secondary analyses revealed that 5-foil multiple-choice items are more difficult and have a higher discrimination index compared to 4-foil items. CONCLUSION: The format used for an examination/test item (case-based or noncase-based) has an impact on item performance characteristics.
OBJECTIVE: To evaluate the relationship between assessment item formats (case-based versus noncase-based) and item performance characteristics. METHODS: Assessment items (1,575) were collected from examinations administered in several therapeutics courses over 4 academic years. Items were categorized as either "case-based" or "noncase-based" and item performance characteristics (discrimination index and level of difficulty) were evaluated. RESULTS: Noncase-based items represented approximately three-fourths of all items that were evaluated, and demonstrated a higher discrimination index than case-based items. Case-based items were generally lengthier and included more detailed information than noncase-based items; however, they were not more difficult and exhibited a lower discrimination index. Secondary analyses revealed that 5-foil multiple-choice items are more difficult and have a higher discrimination index compared to 4-foil items. CONCLUSION: The format used for an examination/test item (case-based or noncase-based) has an impact on item performance characteristics.