Purpose: This study aimed to create a Cardiopulmonar y Resuscitation (CPR) knowledge assessment tool for nurses and establish its reliability and validity for effectively measuring CPR knowledge in nursing simulation. Methods: This methodological study began by developing of 20 preliminary CPR-related questions in a dichotomous format, crafted through collaborative discussions with a rapid response team and a nursing education team. A sample of 226 ward nurses from a tertiary general hospital were assessed for CPR knowledge using these questions before undergoing CPR simulation training. The tool’s reliability was measured using the Kuder-Richardson Formula 20 (KR-20) and confidence intervals, and its validity was determined through difficulty and discrimination indices derived from Item Response Theory. The contrasted group validity was assessed for construction validity. Results: Correlation analysis led to the elimination of three questions with low overall correlation coefficients. Subsequent difficulty and discrimination verification removed five non-discriminatory items. A significant difference in mean scores was found between novice nurses and the rest of the participants (t=2.68, p=.008), highlighting the tool’s construct validity to experience. The final assessment com...