The Research And Development Of Behavioral Observation Scales For Appraising The Performance Of Charge Nurses Of Class Three Hospital In Sichuan

Objective: The purpose of the study is to develop a scientific, quantitative, comparative and simple behavioral observation scales(BOS) for appraising the performance of charge nurses of Class three hospital in Sichuan based on a job analysis known as the Critical Incident Technique.Methods: First, the performance criteria of charge nurses were determined by experts' consultation. Charge nurses and supervisors were interviewed in accordance with the Critical Incident Technique. Then collected critical incidents were categorized into overall performance criteria to form one original BOS. It was revised and polished by objective and subjective methods. Applying item analysis, such as: Distribution Responses, Item Discrimination Coefficient, Item—Tota Correlation, Correlation Matrices, Cronbach's Alpha Coefficient, Factor Analysis to select the discriminating behavioral items , in this way, we eliminated ambiguous items and determined performance criteria. Thus BOS was constructed. Finally, thereliability and validity of one were tested by practicing it in real situation. Results:1. 41 charge nurses reported 370 incidents and 15 supervisors reported 128 incidents for a total of 498 incidents by interviewing. One original behavioral observation scales was composed of 15 criteria and 169 behavior items by categorizing all the collected incidents.2. 48 supervisors observed and rated 190 charge nurses on a five—point scales as to the frequency with which they engaged in the behavior described by each incident. One formal behavioral observation Scales for appraising performance of charge nurses was composed of 12 performance criteria and 54 behavioral items which was derived from item analysis.3. To test the reliability and validity of behavioral observation scales for appraising performance of charge nurses, we operated it in real situation. 48 supervisors observed and rated 192 charge nurses on a five—point scales as to the frequency with which they engaged in the behavior described by each incident. The values for Test-Retest reliability ranged between 0.502 and 0.758; the values for Internal- Consistency Coefficient(Cronbach's Alpha coefficient) ranged between 0.654 and 0.927; Content Validity: items had higher correlations with corresponding performance criteria than any other criteria; Construct Validity: 9 factors with eigenvalues greater than one were selected and accumulative proportion was 68%.Conclusion: Behavioral observation scales for appraising the performance of Charge nurses which is developed based on a job analysis known as the Critical Incident Technique has moderately high reliability and validity.Interestingly, it is not only offer scientific grounds for rating performance of charge nurses, but also can exploit and pilot their job.Suggestion for further study1. We should further explore the performance criterion and the equilibrium distribution of items of behavioral observation Scales.2. We should develop other version of behavioral observation scales for overall rating of charge nurses' performance.