Abstract
Aim:
To determine the discriminative power of the Karen-patient and the Karen-personnel instruments for measuring quality of care, and to reduce the number of items. The aim was also to test the internal consistency.
Background:
Well tested instruments for measuring quality of care, including both the patient- and the personnel perspective, is needed in nursing care. A model was developed, called the KISAAL-system, from which the two instruments used in this study derived from.
Method:
The instruments Karen-patient and Karen-personnel were examined for discriminative power using Likert's method for analysis and Cronbach's alpha coefficient for measuring internal consistency.
Results:
The item analyses resulted in a reduction of 39 items in the Karen-patient instrument and 27 items in the Karen-personnel instrument. Thus, thirty five items remained in both instruments, considered a manageable number. The internal consistency was 0.86 for Karen-patient and 0.88 for Karen- personnel. The remaining items correspond with the original instruments.
Conclusion:
After performing the analysis the remaining items in both instrument discriminated well and the reliability was good. The contents in both scales correspond with the original items and have been reduced to a suitable number to handle in clinical practice. The development of the instruments will continue testing the construct validity.