Special Session Summary Putting Context Effects in Context: the Role of Information About the Choice Environment

Kyeong Sam Min, The Ohio State University
[ to cite ]:
Kyeong Sam Min (2003) ,"Special Session Summary Putting Context Effects in Context: the Role of Information About the Choice Environment", in NA - Advances in Consumer Research Volume 30, eds. Punam Anand Keller and Dennis W. Rook, Valdosta, GA : Association for Consumer Research, Pages: 196-199.

Advances in Consumer Research Volume 30, 2003     Pages 196-199

SPECIAL SESSION SUMMARY

PUTTING CONTEXT EFFECTS IN CONTEXT: THE ROLE OF INFORMATION ABOUT THE CHOICE ENVIRONMENT

Kyeong Sam Min, The Ohio State University

Past research on context effects has considered the consumer to be a cognitive miser prone to rely upon simple heuristics (e.g., Huber, Payne and Puto 1982; Simonson and Tversky 1992). However, we view the consumer as an active information processor who encodes and interprets information about the choice environment as well as the set of alternatives considered. Each of the three papers examines how information about the choice environment affects the consumer’s pliability to context effects.

Understanding the interaction between the consumer and the choice environment is crucial to explaining and predicting the consumer’s reactions to rapidly changing retail settings. For example, retailers are experimenting with various product displays: some are focusing on varietyBthe width of the choice setBwhile others are offering fewer varieties but a greater quantity of each product. Manufacturers and retailers may include inferior products, or "strategically dominated alternatives" in the choice set. What is the effect on choice when the choice set is widened, and is the effect the same for all attributes? How do consumers react when they are told that an alternative is unavailable, and does the timing of this information matter? Does it matter if consumers are aware that the choice context may have been manipulated to influence their choices?

The first paper, by Yeung and Soman, explored how the consumer’s familiarity with the product attributes influences choice when the range of product alternatives in the choice set is widened. It shows that widening the range of encountered options increases the choice of the cheaper (lower quality) product relative to the expensive (higher quality) product. Yeung and Soman show that the asymmetric range effect accounts for this findingBa harder-to-evaluate attribute (e.g., quality) is more susceptible to the range effect than an easier-to-evaluate attribute (e.g., price).

The second paper, by Min and West, examined how the timing of the consumer’s receiving information regarding a product’s availability affects consumer decision making. Min and West show that consumers switch a product type when they experience psychological reactance. In particular, they find that consumer response depends on when consumers are notified regarding unavailability and whether the unavailable product alternative is their favorite. Consistent with reactance theory, they show detrimental impacts of psychological reactance on choice and satisfaction.

The third paper, by Hamilton, investigated how the consumer’s knowledge about the selection of alternatives in the choice context affects susceptibility to context effects. While context effects have been repeatedly demonstrated to influence choice, product displays in experimental settings are not usually perceived to be persuasion attempts (Friestad and Wright 1994). Hamilton shows that the perceived motives of the person creating the choice context and whether this person is a friend or stranger affect the chooser’s perceptions of the alternatives, and that reactions to these factors differ systematically for attraction effects and compromise effects.

Itamar Simonson provided a summary of the research, brought together the theoretical, methodological, and practical issues addressed, and led the audience in a discussion.

 

THE IMPACT OF ASYMMETRIC RANGE EFFECTS ON PREFERENCE REVERSALS

Catherine Wing-man Yeung, The Hong Kong University of Science & Technology

Dilip Soman, The Hong Kong University of Science & Technology

In a choice between two products, we show that widening the range of encountered options at the time of choice increases the choice of the cheaper (lower quality) product relative to the expensive (higher quality) product. We propose an asymmetric range effects account for this finding. Literature on the range effects suggests that people use the range of encountered stimuli to set a lower and upper bound of stimuli, and that a given difference between two stimuli along this range appears perceptually smaller as this range increases. We argue that this effect would be greater for a harder-to-evaluate attribute than for an easier-to-evaluate attribute. Therefore, when we make tradeoff between a harder-to-evaluate and an easier-to-evaluate attribute, varying the two attribute ranges concurrently would induce range effects along the two attributes asymmetrically.

We first study the asymmetric range effects by considering price-quality tradeoffs where quality has lower evaluability (i.e., is harder to evaluate independently) than price. For instance, it is difficult for consumers to evaluate a speaker based on the attribute that "it has a power output of 300W", because they may not know how to interpret the meaning of "300W". However, it is easier for them to judge whether a $300 stereo speaker is cheap or expensive, since they know how much does "$300" worth to them.

To illustrate how the asymmetric range effect influences price-quality tradeoffs, let us consider a consumer who is choosing between a more expensive option, x, and a cheaper option, y. The perceived tradeoff between price and quality can be conceptualized as the following:

EQUATION

Suppose the tradeoff is made in the context of other products that differ in price and quality. We propose that widening the range of the context products (and therefore, the ranges of price and quality concurrently) would reduce the size of perceived quality difference to a greater extent than the size of perceived price difference. Consequently, the perceived price-quality tradeoff would become less favorable, shifting preference towards the cheaper option. Stated formally,

H1: Widening the context range would result in a shift of preference from the more expensive option to the cheaper option.

More generally, we suggest that the shift of preferences towards the cheaper option as ranges increase can be reversed by making price more difficult to evaluate and quality easier to evaluate.

H2: In a tradeoff between a more evaluable and a less evaluable attribute, widening the context range would result in a shift of preference towards the option that having a greater quantity of the more evaluable attribute.

We ran a series of four experiments to examine the two hypotheses. In all experiments, we examine two scenarios: one in which consumers encounter a wide range context and one in which consumers encounter a narrow range context. Context range (wide vs. narrow) was manipulated by the range of assortment that participants were given information about. A wide range assortment included products that varied greatly in both price and quality, whereas a narrow range assortment included products that are relatively similar to each other in both price and quality.

In the first two experiments, we asked participants to judge between a more expensive option and a cheaper option. Participants in the "wide range condition" received price and quality information on six products, which belonged to five different varieties of breakfast food (experiment 1) or beverages (experiment 2). Participants in the "narrow range condition" also received price and quality information on six products, but all of them belong to the same varietyBcereal (experiment 1) or coffee (experiment 2). The result shows that, participants’ preferences shifted towards the cheaper option when the width of context range increased. Thus, hypothesis 1 was supported.

In the third experiment, we kept product variety constant across range contexts, and manipulated only price and quality ranges. We employed PDA as the domain of the experiment. The wide (vs. narrow) range assortment contained PDAs that varied more extensively in price and quality. We asked participants to consider a more expensive and a cheaper PDA, and measured their relative preferences, perceptions of price difference, and perceptions of quality difference between the two options. The result shows that, participants’ preferences shifted towards the cheaper option as the width of context range increased. Moreover, the perceived quality difference (rather than perceived price difference) mediates the effects of assortment range on preference shifts.

We tested hypothesis 2 using a fourth experiment, which had a 2 (price evaluability: high vs. low) x 2 (quality evaluability: high vs. low) x 2 (range: wide vs. narrow) design. We gave participants information about the price and quality of six PDAs. This information was made complicated so that participants would find it hard to evaluate. However, participants in the "high price evaluability" conditions received aids to evaluate the price information. Similarly, participants in the "high quality evaluability" conditions received aids to evaluate the quality information. The result shows that, when price and quality differed in evaluability, widening the context range shifted preferences towards the option that having a greater quantity of the more evaluable attribute. However, when price and quality did not differ in evaluability, participants’ preferences did not change as a function of assortment range.

 

CONSUMER RESPONSE TO PRODUCT UNAVAILABILITY

Kyeong Sam Min, The Ohio State University

Patricia M. West, The Ohio State University

As retailers and e-tailers strive to woo consumers by offering a vast assortment of products, individuals have been increasingly faced with choice constraints when their preferred product is sold out, or temporarily unavailable. How does the timing of information regarding product unavailability influence consumer decision making? Does it matter whether the unavailable product is the consumer’s favorite product or not?

Prior research examining consumer response to product unavailability has focused on the point in time that individuals are informed of product unavailability before they make their choice (e.g., Pettibone and Wedell 2000; Simonson and Tversky 1992). Research to date suggests that when consumers are notified about unavailability prior to making a choice, the likelihood of their selecting the product most similar to the unavailable alternative is increased, relative to a control group for whom the unavailable product has been removed altogether.

However, restricting the consumer’s freedom of choice is likely to have adverse effects, including lowered choice consistency accompanied by feelings of frustration, which cause consumers to assert their freedom of choice by taking an action in the opposite direction than the choice the store had expected them to make (e.g., Brehm 1966; Clee and Wicklund 1980). When the consumer is informed about product unavailability after she has already made a choice, and thus experiences a larger amount of psychological reactance, she may be more likely to select a product that is slightly different from her initial choice. The detrimental effect of psychological reactance on consumer decision making can occur only when the unavailable product is the consumer’s favorite due to its salience (e.g., Broniarczyk, Hoyer, and McAlister 1998; Fitzsimons 2000). Thus, we predict that the choice share of the alternative most similar to the unavailable product will decrease when the consumer is notified about product unavailability after versus before making a choice, only when the unavailable product is the consumer’s preferred product. Similarly, we also expect that decision satisfaction will be qualified by the timing of information and preference for the unavailable product.

To test these predictions, we ran an experiment (N=384) with a 2 (timing of information) x 2 (preference for the unavailable product) x 4 (product category) between-subjects design. The timing of information was manipulated by notifying participants of the presence of an unavailable product either before or after they made a choice. Preference for an unavailable product was determined by whether it was the consumer’s preferred product type. The four product categories tested included restaurants, movies, soft drinks, and cookies.

Subjects’ preference ratings for each product type (e.g., movie genre, restaurant cuisine) were collected one week prior to their participation in the study. In the "before" condition, participants were presented with three alternatives for consideration, including a high quality but unavailable product, and asked to choose between the two remaining moderate quality products. In contrast, in the "after" condition, participants were given the opportunity to select among all three available alternatives. Participants who selected one of the two moderate quality products as their initial choice were never exposed to a choice constraint. Those who selected the high quality product were informed that the product was unavailable. These individuals were then required to choose again from the remaining two moderate quality products. After making each choice, participants were asked to evaluate their decsion satisfaction as well as their feelings toward product unavailability.

Consistent with our prediction, we observed that under the high preference condition, the choice share of a similar alternative was lower when the consumer was informed about product unavailability after (M=59.5%) versus before (M=89.2%) making a choice (c2=8.49, p<.02). In contrast, under the low preference condition, the choice share of a similar alternative did not change when the timing of information was varied (p>.1).

In terms of decision satisfaction, we did not find any significant change in decision satisfaction between "before" and "after" conditions, regardless of preference for the unavailable product (p>.1). Rather, we observed a marginally significant overall main effect of the timing of information on decision satisfaction (t=-1.68, p<.10). Decision satisfaction was lower when participants were in the "after" (M=3.63) rather than "before" condition (M=4.08). In the follow up analysis, we found that decision satisfaction was higher when participants never encountered a choice constraint (M=4.79) than when they were notified of a choice constraint prior to choice (M=4.08; t=2.16, p<.08) or when they were notified of a choice constraint after making a choice (M=3.63, t=3.45, p<.002). These findings indicate that decision satisfaction is influenced not only by whether the consumer is faced with a choice constraint, but also by whether the notification of product unavailability is made before or after a product is selected.

Our empirical results strongly support the predicted role of psychological reactance in consumer decision making. We show that psychological reactance is an important explanatory mechanism for a constrained consumer choice. In addition, our findings offer direct managerial implications for retailers faced with the decision dilemma of how to communicate with their customers about product unavailability. When a preferred product is unavailable, it is best to notify customers of the choice constraint in advance of their making a selection. However, when a less preferred product is unavailable, it is best not to notify the customers of the choice constraint and wait to see if they express an interest in the unavailable product. The late notification strategy has the advantage of offering the customer a wider assortment to choose from, which increases customer satisfaction and consistent choice, while eliminating any frustration or anger associated with having one’s freedom of choice constrained.

 

REACTIONS TO CONTEXT EFFECT IN INTERPERSONAL INTERACTIONS

Rebecca W. Hamilton, University of Maryland

Although they systematically influence people’s choices, displays of information about products in experimental settings are not usually perceived as persuasion attempts (Friestad and Wright 1994). In fact, if people believed these displays of information were constructed to influence their choices, it might be more difficult to obtain effects like the compromise effect or the attraction effect. Persuasion research has shown that in many cases, people who are aware that persuasion tactics are being used to influence them are more resistant to persuasion. For example, forewarning people of the persuasive intent of a message typically results in higher resistance to persuasion (Petty and Wegener 1998) because people don’t like to be told what to think or how to feel (Hass and Grady 1975; Petty and Cacioppo 1979). Thus, a fundamental question is whether context effects will lose their effectiveness if people are aware of them.

Recent research suggests that people respond flexibly to persuasion attempts (Campbell and Kirmani 2000; Wegener and Petty 1997). People may react negatively when they believe a salesperson has an ulterior motive for persuasion (Campbell and Kirmani 2000), yet yield to a recognized persuasion attempt by a friend suggesting restaurants for a shared dinner. Based on this flexibility, we hypothesized that two factors would influence people’s reactions o context effects if they recognized that the choice context was being manipulated: the perceived motives of the person creating the choice context and the relationship between the person creating the choice context and the person making the choice.

To examine these factors, we ran a study with 65 dyads comprised of friends who had signed up for our session together. Each dyad made two choices together and each member of the dyad made two choices with a stranger. Subjects were randomly assigned to the role of menu creator, in which they created the choice context for another person, or chooser, in which they made choices from sets of three alternatives. Dyads were assigned to either the Ulterior Motive condition, in which the menu creator was individually rewarded for influencing the chooser, or the Group Motive condition, in which the goal was to make the best choice for the group.

Subjects were led through a scenario in which they purchased appliances for a hypothetical shared apartment with another person. To increase experimental control, choosers were given experimentally manipulated menus rather than the menus that had been created for them by the menu creators. Each chooser was given two attraction effect menus and two compromise effect menus. These effects were chosen because the attraction effect constrains the chooser more than the compromise effect, making it a more aggressive influence tactic (Barry and Shapiro 1992). Choosers selected one alternative from each of two menus that they believed had been created by their friend, and one alternative from each of two menus that they believed had been created by a stranger. After making their choices, choosers were asked to describe why the menu creator had constructed the menu as she or he had, reconstruct the set of alternatives that had been available to the menu creator, rate their own preferences and the menu creator’s preferences for each of the alternatives, and rate their satisfaction with their choices.

Susceptibility to context effects varied by relationship and by menu type. When subjects received attraction effect menus, they were significantly more likely than chance to choose the dominant alternative (M=68% vs. 33%; Z=5.66, p<.01), especially when they believed the menu had been created by a friend (M=75%) rather than by a stranger (M=61%; Z=1.88, p<.10). In contrast, when subjects received compromise effect menus, they were marginally more likely to chose the compromise option when they believed the menu was from a stranger (M=61%) than when they believed the menu was from a friend (M=44%; Z=2.06, p<.05).

Subjects’ willingness to choose as they believed the menu creator wanted them to choose also varied by menu type and motive. Subjects were marginally more likely to choose as they believed the menu creator wanted them to choose when they received attraction effect menus (M=93%) than when they received compromise effect menus (M=83%; Z=2.20, p<.05). This effect is particularly interesting because subjects perceived a stronger influence attempt when they received attraction effect menus than when they received compromise menus, F(1,60)=5.76, p<.05. When subjects received attraction effect menus, they were equally likely to choose as expected in both motive conditions, but when subjects received compromise effect menus, they were significantly more likely to choose as expected in the Ulterior Motive condition (M=93%) than in the Group Motive condition (M=75%; Z=2.15, p<.05).

While our results indicate that reactions to context effects are influenced by the relationship between the menu creator and chooser and the perceived motive of the menu creator, the effects of these variables differ systematically by menu type. While subjects were more willing to choose the dominant alternative from an attraction effect menu when they believed the menu had been created by a friend rather than a stranger, they were more willing to choose compromise alternatives when they believed the menus had been created by a stranger. While motive did not affect subjects’ willingness to choose as expected from attraction effect menus, subjects were more willing to choose as expected from compromise effect menus when they thought the menu creator had an ulterior motive for persuading them. We discuss these findings and propose future research designed to help us understand the degree to which these findings generalize across context effects and across interpersonal interactions such as salesperson/ customer interactions and group decision making.

REFERENCES

Barry, Bruce and Debra L. Shapiro (1992), "Influence Tactics in Combination: The Interactive Effects of Soft Versus Hard Tactics and Rational Exchange," Journal of Applied Social Psychology, 22(18), 1429-1441.

Brehm, Jack W. (1966), A Theory of Psychological Reactance, New York, NY: Academic Press.

Broniarczyk, Susan M., Wayne D. Hoyer, and Leigh McAlister (1998), "Consumers’ Perceptions of the Assortment Offered in a Grocery Category: The Impact of Item Reduction," Journal of Marketing Research, 35 (May), 166-176.

Campbell, Margaret C. and Amna Kirmani (2000), "Consumers’ Use of Persuasion Knowledge: The Effects of Accessibility and Cognitive Capacity on Perceptions of an Influence Agent," Journal of Consumer Research, 27 (June), 69-83.

Clee, Mona A. and Robert A. Wicklund (1980), "Consumer Behavior and Psychological Reactance," Journal of Consumer Research, 6 (March), 389-405.

Farquhar, Peter H. and Anthony R. Pratkanis (1993), "Decision Structuring with Phantom Alternatives," Management Science, 39 (October), 1214-1226.

Fitzsimons, Gaven J. (2000), "Consumer Response to Stock-Outs," Journal of Consumer Research, 27 (September), 249-266

Friestad, Marian and Peter Wright (1994), "The Persuasion Knowledge ModelBHow People Cope with Persuasion Attempts," Journal of Consumer Research, 21 (1), 1-31.

Hass, R. Glen and Kathleen Grady (1975). "Temporal Delay, Type of Forewarning, and Resistance to Influence," Journal of Experimental Social Psychology, 11(5), 459-469.

Hsee, Christopher K., Geroge F. Loewenstein, Sally Blount, and Max H. Bazerman (1999), "Preference Reversals Between Joint and Separate Evaluations of Options: A Review and Theoretical Analysis," Psychological Bulletin, 125(5), 576-590.

Huber, Joel, John W. Payne, and Christopher Puto (1982), "Adding Asymmetrically Dominated Alternatives: Violations of Regularity and the Similarity Hypothesis," Journal of Consumer Research, 9 (June), 90-98.

Pettibone, Jonathan C. and Douglas H. Wedell (2000), "Examining Models of Nondominated Decoy Effects across Judgment and Choice," Organizational Behavior and Human Decision Processes, 81, 300-328.

Petty, Richard E. and Duane T. Wegener (1998), "Attitude Change: Multiple Roles for Persuasion Variables," in The Handbook of Social Psychology, Gilbert, Daniel T., Fiske, Susan T. and Lindzey, G. (eds.), New York: McGraw-Hill, vol. 1, 323-390.

Petty, Richard E. and John T. Cacioppo (1979), "Effects of Forewarning of Persuasive Intent and Involvement on Cognitive Responses and Persuasion," Personality & Social Psychology Bulletin, 5(2), 173-176.

Simonson, Itamar and Amos Tversky (1992), "Choice in Context: Tradeoff Contrast and Extremeness Aversion," Journal of Marketing Research, 29 (August), 281-295.

Wegener, Duane T. and Richard E. Petty (1997), "The Flexible Correction Model: The Role of Naive Theories of Bias in Bias Correction," Advances in Experimental Social Psychology, 29, 141-208.

----------------------------------------