Best-Worst scaling…reflections on presentation, analysis, and lessons learnt from case 3 BWS experiments
MetadataShow full item record
Surveys based on Likert scales and similar ratings-based scales continue to dominate market research practice despite their many and well-documented limitations. Key issues of concern for Likert scales include over- or under-reporting depending on the context, and variation in responses based on cultural background. Alternatives exist to overcome the inherent weaknesses of these scales. This paper reflects on the Best Worst (BW) Scaling method that we have recently used in eight online studies. In these studies, we employed a novel pictorial approach to capture product preferences for over 3,600 respondents. One case 3 BW experiment investigating consumer preferences for organic apples is featured and evaluated using two approaches. The first analysis treats the data as a case 1 BW experiment to outline the simplicity of case 1 analysis. Case 3 BW analysis involving multinomial logit and latent class modelling is then illustrated to highlight the rich preference insights that can be obtained from case 3 BW studies. We look at some of the drawbacks of the BW case 3 approach, including design and data processing issues, weighted against the overall positive feedback received from respondents. Overall, we do believe the BWS method has a significant potential to improve predictability in market research - the response rate and positive participant feedback speaks for itself.
Market & Social Research
© 2013 Australian Market and Social Research Society. The attached file is reproduced here in accordance with the copyright policy of the publisher. Please refer to the journal's website for access to the definitive, published version.
Marketing Research Methodology