Development of instruments to assess assaultive behavior in nursing homes.

The use of psychometric techniques is not usually considered in the development of non-scale tools. In the development of these tools, both content validation and inter-rater reliability were assessed in an effort to produce better quality tools than would have occurred otherwise. The procedure employing content experts for content validation followed the usual protocol for both of the tools. In each case, valuable information about the tools was obtained, in addition to data supporting the content validity of the tools. The development of vignettes for the inter-rater assessment would probably not have occurred if the authors were not familiar with the different techniques for assessing inter-rater reliability. The only way data concerning inter-rater reliability could be obtained was by having each respondent view the same stimuli. Creating vignettes was the best way to accomplish this goal in a cost-effective manner. The additional data obtained from the respondents also provided invaluable information for the development of the tools. This pilot study was conducted to assess all aspects of the tools in the nursing home setting. Clarity of items and usability of the tools were major considerations. Because NAs were involved in the pilot study, both readability by most NAs and clarity of the items in the revised tool were assured. Considerable time was spent assessing these tools, but the completed tools are evidence that the time was well spent. The use of psychometric techniques resulted in objective evidence that led to quality improvement decisions. The revised tools have the potential for use in obtaining quality data for research projects, in policy development by an agency, and for training agency personnel.