Missing information in single and joint decision contexts: a structural choice formulation

Wallin, Ann and Coote, Len (2014). Missing information in single and joint decision contexts: a structural choice formulation. In: INFORMS Marketing Science 2014: The 36th ISMS Marketing Science Conference, Atlanta, GA, United States, (). 12-14 June, 2014.

Author Wallin, Ann
Coote, Len
Title of paper Missing information in single and joint decision contexts: a structural choice formulation
Conference name INFORMS Marketing Science 2014: The 36th ISMS Marketing Science Conference
Conference location Atlanta, GA, United States
Conference dates 12-14 June, 2014
Publication Year 2014
Sub-type Published abstract
Open Access Status
Language eng
Formatted Abstract/Summary
Consumers make product evaluations and subsequent choices in one of two basic modes: “single” and “joint” evaluation. Evaluation mode has been found to impact how easy and/or difficult product information is to evaluate and in turn to impact on consumer decision making and choice. Missing information further complicates this decision making process. Previous studies show missing attribute information impacts both the systematic and random components of utility for choice alternatives (Islam, Louviere and Burke 2007). Moreover, profiles with missing attribute level information can lead to intransitive preferences (Kivetz and Simonson 2000). This study contributes to the field by investigating if evaluation mode impacts the processing of missing information. Specifically, this study examines the impact of missing attribute level information on utility for choice alternatives in single vs. joint evaluation modes. This paper investigates two key questions. Firstly, does the impact of missing attribute information on utility differ between single and joint evaluation modes? Second, are there identifiable latent processes that define how consumers’ differ in their responses to missing attribute information? The current research uses discrete choice experiments (DCE’s) in a within-subjects design. A model catalogue is advanced with varying degrees of latent structure. The most fundamental of the models (McFadden’s conditional logit) explores differences in aggregate preferences due to variation in missing attribute information. Subsequent models add latent structure in one of two ways. Firstly, latent class models explore differences in aggregate preferences based on discrete segments of decision makers (in response to missing attribute information). Second, factor-analytic models are used to parameterize unobserved differences between decision makers (in response to missing attribute information). These models retrieve decision makers’ aggregate preferences and retrieve patterns in the preference heterogeneity in relation to how decision makers differ.
Q-Index Code EX
Q-Index Status Provisional Code
Institutional Status UQ

Document type: Conference Paper
Collection: UQ Business School Publications
 
Versions
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Created: Fri, 13 Jun 2014, 10:53:13 EST by Ms Ann Wallin on behalf of UQ Business School