Intra-rater reliability

About: Intra-rater reliability is a research topic. Over the lifetime, 2073 publications have been published within this topic receiving 140968 citations.

Validation of an Assessment Instrument for Pediatric Basic Life Support.

Abstract: OBJECTIVES To develop a valid and reliable instrument for the assessment of pediatric basic life support (PBLS). METHODS An assessment instrument for PBLS was developed, based on 3 existing scoring systems and the European Resuscitation Council PBLS guideline. We tested if experienced PBLS instructors performed better than medical students on a standard PBLS examination on a low-fidelity pediatric manikin (construct validity). To pass the examination, 15 penalty points or less were required. The examinations were videotaped. One researcher assessed all videos once, and approximately half of them twice (intrarater reliability). A second researcher independently assessed part of the videos (interrater reliability). The time needed to assess 1 examination was determined. RESULTS Face and content validity were established, because PBLS experts reached consensus on the instrument and because the instrument incorporated all items of the European Resuscitation Council algorithm. Of the 157 medical students that were scored, 98 (62.4%) passed the examination. Fourteen PBLS instructors were scored; all passed (100%). Pass rate (62.4% vs 100%) and median penalty points (15 [interquartile range, 10-22.5] vs 7.5 [interquartile range, 1.25-10]) were significantly different between students and instructors (P = 0.005 and <0.001, respectively). Reassessment demonstrated a κ for intrarater reliability of 0.62 (95% confidence interval, 0.45-0.81) (substantial agreement); κ for interrater reliability was 0.51 (95% confidence interval, 0.09-0.93) (moderate agreement). It took approximately 3 minutes to assess 1 videotaped examination. CONCLUSIONS Our instrument for the (video-based) assessment of PBLS is valid and sufficiently reliable. It is also designed to be practical, time-efficient, and applicable in various settings, including resource limited.

Study on the Assurance Ability of Statistical Test on Software Reliability

Abstract: This paper studies statistical test's assurance ability on different programs and brings forward to incorporating effectiveness information into software reliability assessment. This can improve the accuracy of reliability assessment. This paper empirically investigates the reasonableness of this method. The conclusion provides a new way for high reliable software assurance.