Statistical Characteristics of the Weighted Inter-Rater Reliability Index for Clinically Validating Nursing Diagnoses.

PURPOSE To analyze the statistical characteristics of the weighted inter-rater reliability index for validating nursing diagnoses. METHODS From a computational algorithm, the index values were compared with the sample size, frequency of the clinical indicator, number of agreements between raters, and three other agreement estimators. FINDINGS The index moderately correlated with the frequency of the clinical indicator and strongly correlated with the number of agreements and disagreements between the raters. This index showed a poor correlation with sample size. CONCLUSION This index selects those clinical indicators with high frequency and/or absolute agreement and therefore has a high sensitivity. IMPLICATIONS FOR NURSING PRACTICE The high sensitivity may result in incorporating false positive clinical indicators.