Idiosyncratic matching and choice: When less is more

https://doi.org/10.1016/j.obhdp.2006.01.006Get rights and content

Abstract

In three studies, reliance on “goodness-of-fit” criteria exerted an influence on assessments of pay-TV packages that challenged normative standards relating to the principle of dominance. Studies 1 and 2 showed that the addition of one or more non-favored channels to a package of favored ones resulted in lower consumer interest, and the subtraction of one or more non-favored channels from a mixed package resulted in greater consumer interest, even when the relevant offering price remained unchanged. Study 3 demonstrated the specific relevance of perceived idiosyncrasy to this phenomenon. Thus, participants responded more positively to a given price reduction for relinquishing a channel they did not like when they believed that their preference ordering was idiosyncratic rather than one shared by their peers. These findings, buttressed by additional data collection to rule out some alternative interpretations and permit within-subject as well as between-subject comparisons, extend the theoretical and practical implications of previous research on the non-normative use of personal “goodness-of-fit” criteria in evaluating the attractiveness of products and programs.

Introduction

Two issues confront the consumer who is considering a purchase. The first is whether any available decision option satisfies the consumer’s threshold requirements in terms of potential benefits versus potential costs and risks, expenditures of time and effort, and loss of other opportunities. When more than one available option appears to exceed that threshold, a second issue arises—that is, how well does each of those options fit the consumer’s idiosyncratic tastes, resources, and tolerance for uncertainty. Normative models imply a process that involves evaluation and weighting of the various features and dimensions of the choices, and a calculation of overall expected utility for the relevant options in light of the decision-maker’s underlying preferences (Edwards, 1954, Savage, 1954). But everyday experience, as well as a considerable body of research, suggests that the process of evaluating options proceeds in a less normative fashion.

Evaluations of options and even underlying preferences often prove to be unstable; and the decisions that get made may reflect the influence of various cues and decision-making heuristics (Bazerman et al., 1992, Bettman et al., 1998, Hsee, 1996, March, 1978, Nowlis and Simonson, 1997, Payne et al., 1992, Shafir et al., 1993) rather than any normatively prescribed assessment and comparison of options. For instance, the consumer who is uncertain about the benefits to be derived versus the cost to be borne from a particular purchase may give great weight to evidence that the price of the object or opportunity in question has been reduced from some “regular” price or reference point, and is thus, in purely qualitative terms, a “good deal” or “bargain” (Shindler, 1989, Shindler, 1998). Indeed, both seller and buyer may engage in a familiar bargaining ritual whereby the seller stipulates a high asking price in the expectation that the buyer will make a lower counter-offer, and the “reduced price” ultimately agreed upon satisfies buyer and seller alike.

The consumer who is having difficulty weighing the relative attractiveness of competing options may also seek to determine whether, holding price constant, one option dominates some of the alternatives in terms of available features, or, holding available features constant, one option dominates some of the alternatives in terms of price (Payne, Bettman, & Johnson, 1993). Where risk is attached to one or more of the options, the consumer may adopt a “mini-max” or worst-case scenario strategy (Von Neumann and Morgenstern, 1944, Williams, 1986), or otherwise neglect the criterion of maximum expected utility by exhibiting “loss aversion” (Kahneman and Tversky, 1979, Kahneman and Tversky, 1984, Tversky and Kahneman, 1986).

The specific decision-making criterion or “heuristic” investigated in the present studies involves “goodness-of-fit” (Simonson et al., 1994, Thaler, 1985), or more precisely, “idiosyncratic matching” (Kivetz & Simonson, 2003). The latter term was coined by Kivetz and Simonson in reference to the “goodness-of-fit” between a given offering and the decision-maker’s personal preferences and/or opportunities for utilization. But as we shall note, these investigators did not actually deal with the issue of idiosyncrasy. More specifically, they did not seek to determine whether, holding constant considerations of goodness-of-fit, beliefs about the idiosyncrasy or atypicality of one’s particular preference ordering exert an independent influence on the attractiveness of one’s decision options. Our specific hypotheses, by contrast, explicitly attach weight to such “social comparison” considerations. In particular, we argue that consumers who are uncertain about whether a given option is worth the price may be influenced not only by whether its features match their personal preference ordering, but also by whether it matches their own idiosyncratic tastes better than it matches those of others.

In many instances, of course, reliance on this decision-making criterion may serve the consumer well. If one’s peers have been willing to pay the asking price for a product or service, and one knows that one is more likely to use and derive pleasure from that product or service than those peers, one is relatively likely to find the purchase satisfactory. Conversely, if one knows one is likely to make less use and derive less pleasure from a product or service than one’s peers, one is relatively unlikely to find the purchase satisfactory. But in certain instances, we shall argue, the same criterion may lead consumers to decisions that violate normative standards regarding dominance and thus fail to maximize utility.

The research to be reported here will test three specific predictions about the use, and counter-normative consequences, of such “idiosyncratic matching” in the responses of young Israelis who were asked to consider various cable TV offerings. We shall outline each of these predictions briefly, and then discuss some relevant prior research before turning to the details of our studies.

  • (1)

    Adding or including channels that the potential viewer has little interest in watching to an offering—even if the price of the offering is not increased—will make the offering less attractive to that viewer.

  • (2)

    Subtracting or excluding channels that the viewer has little interest in watching from a package offering—even if the price of that offering is not decreased—will make the offering more attractive to that viewer.

  • (3)

    Offers that are congruent with the viewers’ preference ordering will be deemed more attractive by those viewers when they believe that their ordering is relatively idiosyncratic than when they believe it is shared by most others facing the same offer.

The third, and most novel prediction listed above, represents an extension of the basic social comparison theory notion (Festinger, 1954, Goethals, 1986, Latane, 1966) that in the face of ambiguity, people make assessments relevant to the self through a process that involves comparison of self to socially relevant others. More specifically, we are suggesting that when objective information about the absolute quality of a given option is lacking or ambiguous, people essentially ask themselves not only how well that option matches their own preference ordering, but also how well it matches their preference ordering relative to that of their peers. They then use this comparison in assessing the attractiveness of the relevant purchasing option accordingly.

One relevant line of evidence for the use of a more general goodness-of-fit criterion was provided a decade ago by Simonson et al. (1994). These investigators showed the negative effects on product attractiveness of adding product features or opportunities that a significant segment of the consumer market would not value or utilize. Thus, informing consumers that purchasing a cake mix will afford them the opportunity to purchase a collector’s plate in which they have little interest, or that the calculator they are considering includes an esoteric function they are unlikely ever to use, makes them see those products as less attractive than seemingly comparable alternatives that lack the relevant “bonuses.” In a sense, the poor fit between the offering and the tastes and needs of many consumers provides those consumers with various “arguments” against the purchase, including some arguments, as the authors note, that are quite rational or normative. Indeed, all things being equal, it is not at all unreasonable to assume that if merchandiser X is offering a premium to buyers of a given product, and merchandiser Y is not offering such a premium to buyers of a seemingly similar product, then the quality, desirability, and market value of the latter product is apt to be greater than that of the former product.

More recent work by Kivetz and Simonson, 2002, Kivetz and Simonson, 2003 on responses to “loyalty programs” (programs that offer some reward for reaching some particular number or total of expenditures) is particularly provocative both because the authors explicitly show the relevance of individual differences in the circumstances of consumers that make a given offer a better “fit” for some consumers than others, and because they considered normative versus non-normative explanations for the phenomenon in question. Their basic finding involved the tendency for a particular group of individuals to express greater interest in programs with relatively difficult to satisfy requirements than programs with relatively easy to satisfy requirements. The group in question consisted of those whose particular circumstances and/or tastes would make it easier for them to meet the more stringent requirements than it would be for their peers. Thus, for example, people told to assume that they lived close to a particular gas station at which they regularly filled their tanks expressed greater interest in a program that called for 20 purchases in order to receive a free car vacuum cleaner than did those given the same information about residence but told that only 10 purchases were required in order to receive the same loyalty reward. (Research participants told that they lived far from the gas station, not surprisingly, showed more interest in the program when the reward could be earned with the smaller number of purchases.)

The Kivetz and Simonson (2003) study that is most provocative in terms of our present theses, perhaps, is one that considered the attractiveness of loyalty program for diners. This study showed that a sample of diners who said that they greatly liked sushi showed (marginally) greater interest in a program that required 12 sandwich purchases plus 10 sushi purchases in order to receive the relevant loyalty reward (a movie ticket good at any local theater and a prepaid $10 phone card) than they did in a program that required only the 12 sandwich purchases. (Not surprisingly, participants who were not “sushi lovers” overwhelmingly showed greater interest in the program when the requirement included only the 12 sandwich purchases). In other words, the addition of a requirement that fits the idiosyncratic tastes of the consumer made the program more attractive—even though the additional requirement made it more difficult and expensive to receive the reward, and even though the nature and value of the loyalty reward was unambiguous.1

The loyalty program studies by Kivetz and Simonson (2003) offered findings consistent with the use of a kind of idiosyncratic “matching” but, as the authors noted, their studies are subject to alternative (normative) explanations. Thus, it could be argued that the difficulty and expense of gaining the vacuum cleaner might be regarded by the consumer as an indication of its quality and value. It could also be argued that potential diners who were sushi lovers might welcome the addition of a sushi consumption requirement because it gave them an additional justification for engaging in a gastronomic activity they enjoyed—even though they might not opt for such a heightened requirement if given an explicit choice.

One goal of the present research is to provide evidence for the use of a goodness-of-fit or matching criterion that does not readily lend itself to such alternative interpretations. Our first two studies were designed to provide such evidence, using paradigms that do not involve any of the special properties and complexities of premium offers or loyalty programs. The specific prediction to be tested is that people will show greater interest in purchasing a cable TV deal offering them a single channel or package of channels that they like than a deal offering them the same channel or package of channels plus an additional channel or set of channels that they do not like and would be unlikely to watch. This predicted pattern of between-condition differences in ratings, we suggest, is one that would be difficult to justify on any normative grounds involving “signals” about objective or “market” value.

The present research will also allow us to test two additional predictions that have not been addressed in previous research. One of these predictions is that subtracting an unattractive option or set of options (with no reduction in price) from a larger set of options will increase the attractiveness of the relevant package. Again, the predicted pattern of differences in mean ratings would be difficult to justify on normative grounds. The other prediction relates directly to the novel feature of our present theorizing, which involves the perceived idiosyncratic versus non-idiosyncratic nature of the individual’s preference ordering. Specifically, we contend that a given match between the individual’s preference ordering and the purchase opportunity at hand will be more attractive when the individual feels that his or her ordering is in fact idiosyncratic or unusual rather than one that would be shared by most other potential consumers. The research designs and procedures used to test these additional predictions, as later will become apparent, further clarify the nature of the relevant decision-making heuristic. They will also help us to rule out alternative interpretations of our findings—an objective that will be furthered by collection of additional data in Study 2 dealing with participants’ expressed preferences when asked directly to choose between larger and smaller packages at the same price.

Section snippets

Study 1: Effects of increasing or decreasing matching via addition or subtraction of a non-desired option

Our first study tests the two predictions offered earlier about the effect of including or permitting the exclusion of viewing options that the potential Pay TV consumer does not value or intend to utilize. In the case of both predictions, the comparison ultimately involves the relative attractiveness of an offering of a “favored” channel versus the attractiveness of the same “favored” channel plus a “non-favored” channel at the identical price.

Study 2: Further tests of the idiosyncratic matching hypothesis: 6 channels versus 9 channels

Study 2 was designed in part to show that the phenomena demonstrated in Study 1 did not reflect a simple tendency for many participants to attach a negative utility to the channel that they had the option to add to, or subtract from the offer presented to them. (For example, female respondents might have been expressing their objections to violent sports and/or seeking to avoid having such programming available to their spouses or children. Less plausibly, perhaps, male respondents might have

Study 3: Responses to apparently “Idiosyncratic” versus “Non-Idiosyncratic” matches

In our final and most crucial study, as in Study 1, some participants were asked about their interest in purchasing a two-channel cable TV package consisting of one channel that they were apt to like and one they were apt not to like, while other participants were asked about their interest in purchasing only the channel they were apt to like, for the same (purportedly reduced) price. What varied in Study 3 was the supposedly idiosyncratic versus non-idiosyncratic nature of the match between

General discussion

The series of studies reported here was designed to illustrate the non-normative use of a decision-making heuristic that employs “matching” or “goodness-of-fit” criteria in general, and idiosyncratic matching criteria in particular. Thus, we sought ultimately to demonstrate that beyond assessing the “goodness-of-fit” between the features of a given option and their own preference orderings, decision-makers engage in a kind of social comparison whereby they assess how well those features match

Acknowledgments

This research reported here benefited from preliminary discussions with Itamar Simonson and Ran Kivetz. We are also indebted to Dale Griffin and Chip Heath, and our anonymous reviewers for their valuable comments on an earlier draft of this paper.

References (36)

  • C.K. Hsee

    The evaluability hypothesis: An explanation for preference reversals between joint and separate evaluations of alternatives

    Organizational Behavior and Human Decision Processes

    (1996)
  • R.E. Nisbett et al.

    The dilution effect: Nondiagnostic information

    Cognitive Psychology

    (1981)
  • E. Shafir et al.

    Reasons-based choice

    Cognition

    (1993)
  • N.H. Anderson

    Averaging versus adding as a stimulus-combination rule in impression formation

    Journal of Experimental Psychology

    (1965)
  • M.H. Bazerman et al.

    Reversals of preference in allocation decisions: Judging an alternative versus Choosing among alternatives

    Administrative Science Quarterly

    (1992)
  • J.R. Bettman et al.

    Constructive consumer choice processes

    Journal of Consumer Research

    (1998)
  • L.A. Brenner et al.

    Remarks on support theory: Recent advances and future directions

  • R.B. Cialdini

    Influence: The psychology of persuasion

    (1993)
  • W. Edwards

    The theory of decision making

    Psychology Bulletin

    (1954)
  • L. Festinger

    A theory of social comparison processes

    Human Relations

    (1954)
  • G.R. Goethals

    Social comparison theory: Psychology from the lost and found

    Personality and Social Psychology Bulletin

    (1986)
  • D. Kahneman et al.

    Representativeness revisited. Attribute substitution in intuitive judgment

  • D. Kahneman et al.

    Judgment under uncertainty: Heuristics and biases

    (1982)
  • D. Kahneman et al.

    On the psychology of prediction

    Psychological Review

    (1973)
  • D. Kahneman et al.

    Prospect theory: An analysis of decision under risk

    Econometrica

    (1979)
  • D. Kahneman et al.

    Choices, values, and frames

    American Psychologist

    (1984)
  • R. Kivetz et al.

    Earning the right to indulge: Effort as a determinant of customer preferences toward frequency program rewards

    Journal of Marketing Research

    (2002)
  • R. Kivetz et al.

    The role of effort advantage in consumer response to loyalty programs: The idiosyncratic fit heuristic

    Journal of Marketing Research

    (2003)
  • Cited by (7)

    • Fate or fight: Exploring the hedonic costs of competition

      2012, Organizational Behavior and Human Decision Processes
      Citation Excerpt :

      The existing literature has documented much richer evidence for the choice-free-peace notion than for the opportunity-free-peace notion. A number of studies demonstrate that individuals can sometimes be happier without the freedom of making choices than with (e.g., Botti & Hsee, 2010; Botti & Iyengar, 2004; Botti & McGill, 2006, 2011; Botti, Orfali, & Iyengar, 2009; Gilbert & Ebert, 2002; Luce, Bettman, & Payne, 2001), and happier with fewer choice alternatives than with more choice alternatives (e.g., Botti & Iyengar, 2006; Iyengar & Lepper, 2000; Liberman & Ross, 2006; Schwartz, 2000; Schwartz et al., 2002). For example, Botti and Iyengar (2004) found that participants were happier with the chosen food item among several unattractive alternatives if the choice was made by others than if the choice was made by oneself.

    • Beating the market: The allure of unintended value

      2013, Journal of Marketing Research
    View all citing articles on Scopus
    View full text