« back
Room
Tagungszentrum
Thursday, September 13  »  11:45 - 13:00
Symposium 6
Consumer Psychology
Host
Ulrich Hoffrage (University of Lausanne )
Chair
Ulrich Hoffrage (University of Lausanne )
Discussant
Damian Läge (University of Zurich)
All of us are consumers. The coffee we have at breakfast and the toothpaste we use before we go to bed are uncontroversial examples, but to some extent even activities such as attending scientific conferences can be considered as consumption. The present symposium presents recent studies, conducted at the Universities of Basel, Lausanne and Zürich, that try to provide some answers to the following questions: (1) How do consumers interpret health claims on food packages? (2) Which strategies do consumers use to make a choice among several products and to what extent are they able to verbalize and formalize their strategies? (3) When and why do consumers defer to make a choice? (4) How does affect influence people's willingness to invest in screening potential choice options, thereby improving their chances to obtain desirable products? (5) To what extent is it possible to classify types of products that can be linked to types of consumers? Taken together, these studies use a variety of methodologies, encompassing, in particular, conversational pragmatics analysis, process tracing methods, computational modelling, priming, and non-metric multidimensional scaling.
Speakers
Nils Reisen
Identifying Decision Strategies in a Consumer Choice Situation
Authors
Nils Reisen (University of Lausanne)
Ulrich Hoffrage (University of Lausanne)
In a study on consumer choices, we combined elements of three different process tracing techniques (Mouselab, Active Information Search, and retrospective verbal protocols) to shed light on the cognitive processes and to identify the strategies people used. After participants had repeatedly selected one out of four mobile phones, they were asked to formalize their strategy such that it could be used by someone else to make future choices on their behalf. Of the 31 participants, 28 (90%) reported strategies that used elimination of alternatives somewhere in the process, and 24 of 31 (77%) reported a strategy that added up attributes in a linear fashion. About two thirds of the participants (21 of 31; 68%) combined the two operations, elimination and addition, and most of those (18 of 21; 86%) started with elimination. The match between the choices predicted from the identified strategies and the observed choices was on average 73%. In contrast, the mean match when using one participant's strategy to predict the choices made by another participant was only 34%. Previously established decision strategies (Weighted ADDitive and Take the Best) predicted roughly half of the participants' choices correctly. It can be concluded that this particular combination of process tracing techniques is a useful research tool
Chris White
The Two-Stage, Two-Threshold (2S2T) Model of Choice Deferral
Authors
Chris White (University of Lausanne)
Ulrich Hoffrage (University of Lausanne)
William James once noted that “When you have to make a choice and don’t make it, that in itself is a choice.” Indeed, in many choice situations people do not choose one of the options, instead they defer making a choice. To improve our understanding of choice deferral behavior, we introduce the Two-Stage, Two-Threshold (2S2T) model, which is based on the following assumptions: A person may defer choice either because none of the options reaches an acceptable level of attractiveness or because more than one option exceeds the attractiveness threshold but there is uncertainty which of the remaining options is the best. These separate reasons require that preferential choice involves two stages of processing, one that computes the absolute attractiveness of each option independently of the others, and another that processes the information interactively to compute the relative attractiveness of the options. A threshold value is associated with each stage and for an option to be chosen it must exceed both thresholds, otherwise choice is deferred. Using data from a novel experimental paradigm in which deferring choice had consequences despite using a controlled laboratory environment, the 2S2T model categorized responses according to which stage(s) of processing was responsible for each. The model’s predictions were verified. Furthermore, it accounts for all previously documented choice deferral phenomena, including the tyranny of too much choice (Iyengar & Lepper, 2000).
Sebastian Hafenbrädl
Affective Decision Making in Consumer Choice
Authors
Sebastian Hafenbrädl (University of Lausanne)
Ulrich Hoffrage (University of Lausanne )
Chris White (University of Lausanne)
When making purchase decisions, how many options do people wish to have? Does this number depend on people’s affect towards the product? To address this, we combined two research areas, the “tyranny of too much choice” (Iyengar & Lepper, 2000) and “affective decision making” (Hsee & Rottenstreich, 2005).
Participants chose either 5 or 10 postcards of their university's campus and Christmas tags from a choice set whose size they determined beforehand. To alter their sensitivity towards the number they could choose, we performed a feeling versus calculation priming manipulation before the choice. The Willingness-to-Pay for the cards and the Desired-Set-Size (the number of items participants wanted to choose from) were the dependent variables. The calculation-primed participants were willing to pay significantly more for 10 cards than for 5. In contrast,the feeling-primed participants were largely insensitive to the number of items they would choose. This interaction occurred for the Desired-Set-Size, but failed to reach statistical significance.We also measured participants' satisfaction with the choosing process and the chosen items. Reutskaja and Hogarth (2005) found satisfaction to be an inverse U-shaped function of the manipulated choice set size. In contrast, we found that there was no linear or nonlinear relationship between the self-determined choice set size and reported satisfaction.This research has implications for marketing and the management of assortments.
Stefan Ryf
Visualization of preference structures in markets with robust ideal point models
Authors
Stefan Ryf (University of Zurich)
Damian Läge (University of Zurich)
Ideal point models are designed to visualize the preferential relations of consumers to a set of brands or products in form of a geometrical configuration. This allows an intuitive interpretation of the preference structure of a market. The application of multidimensional ideal point models using the unfolding technique looks promising, but usual problems with degenerated solutions have so far inhibited frequent use of such models with market research data. We present a new procedure for creating ideal point models without degeneracies which is based on sequential fitting of single ideal points for persons or brands using the nonmetric multidimensional scaling algorithm RobuScal and a robust Procrustean transformation. The resulting ideal point models excel those of traditional internal and external unfolding procedures regarding both goodness of fit and interpretability. A study on preferences in lifestyle-relevant markets (car brands, fashion brands, fashion styles for women and men, holiday offers, leisure activities, beverages and living room furnishing) shows a successful application of this new method and demonstrates the potential of good interpretable ideal point models.
« back