Investigating the consistency of conventional ranking versus best worst scaling generated preferences for attributes of quality of life in younger and older person general population samples

Abstract: Best worst scaling is gaining increasing prominence in health economics as a method for eliciting patient and general population preferences for health and health care. The method was originally developed as an efficient method of data collection, the primary purpose of which, via repeated rounds of best-worst choices, is simply to obtain a full ranking of items in a manner that is comparatively easier for respondents to undertake relative to a traditional ranking exercise. One way in which this hypothesis may be formally tested is to examine the relative level of choice consistency (the variability in choice outcomes not explained by attributes and their associated preference weights) for best worst scaling generated responses relative to traditional ranking. This presentation will report the methods and preliminary results from a study that sought to address this issue through simultaneous application of successive best worst and conventional ranking methods to assess the relative importance of key dimensions of quality of life in two samples of the Australian general population differentiated by age (younger and older person). Whilst the findings indicate broad agreement overall, some inconsistencies are evident highlighting that these two methods of data collection may not be interchangeable.