An examination of the effect of asymmetrically dominated decoys on consumer choice behaviour by using judgments and choice : a thesis presented in partial fulfilment of the requirements for the degree of Master of Business Studies in Marketing at Massey University

Abstract

The Purpose of this research was to test assumptions underlying the three asymmetrically dominated decoy effect (ADE) models described by Wedell and Pettibone (1996): the Weight Change Model, the Value Shift Model, and the Emergent Value Model. Of particular interest was whether there is support for an alternative interpretation of the Weight Change Model, as proposed by Bonaccio and Reeve (2006). Wedell (1991). Wedell and Pettibone (1996) conducted a series of studies on the three models. Their results found no support for the Weight Change Model but support both the Value Shift Model and the Emergent Value Models. These results have been widely accepted (see Highhouse, 1996; Pettibone & Wedell, 2000; Tenbrunsel & Diekmann, 2002). However, Bonaccio and Reeve (2006) proposed a different interpretation of the Weight Change Model. They argued that the lack of the supporting evidence for the model is a result of misinterpretation of the original Weight Change Formulation derived from Huber. Payne and Puto (1982). They suggested that the Weight Change Model might be a viable explanation for the ADE. This study examined this issue by replicating the analysis of both Wedell and Pettibone (1996) and Bonaccio and Reeve (2006). using both a choice task and judgment tasks from their study, but with a different set of products and attribute values. The research used a mail survey of 960 New Zealand residents selected from the Electoral Roll of a median size New Zealand city. The research demonstrated strong ADD effect in both choice and judgment tasks and found a strong relationship between choice and judgment. The study also found little support for either of the Weight Change Models (i.e.. that of Wedell and Pettibone's, 1996. or that of Bonaccio and Reeve, 2006). but strong support for the Value Shift Model and some support for the Emergent Value Model. The use of No-decoy comparison with the decoy condition in current study also provides evidence that Wedell and Pettibone's (1996) analysis is a weak test of the ADE and should not be used to support the decision of whether to use ADD in a choice set.