Addressing Reliability Estimates When the Population is Small: Can Social Cognition Help?

Abstract

One of the issues facing researchers studying very select populations is how to obtain reliability estimates for their instruments. When the populations and resulting samples of studies are very small and select, gathering data for typical reliability estimates becomes very difficult. As a result, many researchers ignore the concern about reliability of their instrumentation and forge ahead collecting data. In response to this concern, Bandura's model of social cognition and Wolpe's model of systematic desensitization were combined and applied to a group of 90 undergraduates completing a Communication Satisfaction Scale designed to assess the attitudes of intubated patients in a hospital Intensive Care Unit. Stimuli (text, auditory and visual) were provided to sensitize the subjects to the intubation procedure and to enable the subjects to imagine what it is like to be an intubated patient. The subjects responded to 10 items focusing on the communication issues of intubated patients on a scale in Likert format. Internal reliability (Cronbach alpha) was 0.83 for the entire scale. The results are discussed within both a social cognition and a measurement framework. While the resulting reliabilities cannot be directly applied to the intubated sample, the procedure may provide critical feedback to researchers and instrument developers prior to the actual administration of the instrument in research.