The effects of including/excluding attributes on consumer choice: An empirical study of new food packaging

This thesis is an investigation of the impact of including/excluding product packaging attributes on consumer product choice. Two newly developed packaging formats, provided by a multi-national food packaging company, were used in choice experiments. Eight interrelated choice experiments were designed to systematically and independently vary the inclusion/exclusion of attributes. Covariance heterogeneity models were applied to model effects on systematic and random components. The empirical results showed that inclusion/exclusion of attributes in discrete choice experiments not only impacts consumers' mean preference (the "systematic utility component"), but also impacts the consistency of choices (the "random utility component"). Results also suggest that failure to take choice consistency into account in comparing or pooling the results of choice experiments is likely to lead to biased conclusions about mean attribute effects. The findings add new insights for academic researchers, policy makers and managers in the area of consumer product choice with incomplete information.