Special Session Summary Context Effects in Consumer Judgments

Andrew A. Mitchell, University of Toronto
[ to cite ]:
Andrew A. Mitchell (2002) ,"Special Session Summary Context Effects in Consumer Judgments", in NA - Advances in Consumer Research Volume 29, eds. Susan M. Broniarczyk and Kent Nakamoto, Valdosta, GA : Association for Consumer Research, Pages: 455-457.

Advances in Consumer Research Volume 29, 2002     Pages 455-457

SPECIAL SESSION SUMMARY

CONTEXT EFFECTS IN CONSUMER JUDGMENTS

Andrew A. Mitchell, University of Toronto

Research in a number of areas of psychology finds that human behaviour is very context dependent. For instance, research in cognitive psychology finds that the similarity between encoding and retrieval conditions facilitates memory retrieval. As expected, strong context effects also have been found in consumer behavior (e.g. Myers-Levy and Sternthal 1993; Lynch et al. 1993). These context effects can be due to different usage situations, differences in the products seen prior to judgment, or to the activation of different information in memory. In the proposed session, we focus on the latter two reasons for context effect.

Although a number of studies have identified context effects in consumer judgments and a number of different causes of context effects have been proposed, including correction bias (Martin 1986), target ambiguity (Herr 1989), categorization processes (Herr 1986), range-frequency effects (Parducci 1965) and the formation of standards of comparisons (Schwarz and Bless 1992), we do not have a good understanding of which of these cognitive mechanisms may cause these effects and if there are different cognitive mechanisms, under what situations each operates.

The objective of this session is to obtain a better understanding of the cognitive mechanisms which cause these effects. In order to obtain this understanding, the three papers in this session (1) examine different causes of context effects, (2) examine both within and between product category context effects, nd (3) argue for different cognitive mechanisms which cause the effects.

The first paper, by Raghunathan and Irwin examines a situation where subjects see a sequence of products and then evaluate a target product which is from the same or a different product category. Their results indicate that contrast effects occur when the target product is from the same product category, however, assimilation effects occur when the target product is from a different product category. They present evidence that standard models of context effects, such as range-frequency effects, cannot explain these results. Instead, they argue the assimilation effects are due to mood, a new explanation of assimilation effects, while the contrast effects are due to the context products acting as standards of comparison.

The second paper, by Jung and Lee examines the effect of involvement on context effects when the target product is from the same or a different product category. In their studies, subjects first see an ad for an actual product and then evaluate the target product. They find contrast effects under high involvement, but assimilation effects under low involvement when the target product is from the same product category, and contrast effects under both high and low involvement when the product category is different. These results seem to differ from those obtained by Raghunathan and Irwin. Jung and Lee argue that the contrast effects found under high involvement when the target is from the same product category are due to correction bias, while the contrast effects found when the target product is different are due to the context products acting as standards of comparison within the category of "luxury things", even though they are from another product category. This finding and explanation differs from previous findings and explanations of cross category context effects (Stapel et al. 1998).

Finally, the paper by Dacin and Mitchell focuses on the cognitive mechanism underlying brand evaluation formation based on unambiguous information about the brand. These mechanisms are based on Norm Theory (Kahneman and Miller 1984) and the Inclusion/Exclusion Model (Schwarz and Bless 1992) and cause memory based context effects in brand evaluation. They argue that when forming brand evaluations, individuals generate similar brands from memory and these generated alternatives are used to form standards of comparisons. These standards of comparisons allow individuals to form evaluations of perceived attribute levels of the target brand. These attribute evaluations are then summed to form the brand evaluation. These processes require few cognitive resources and appear to occur below conscious awareness. The results of a series of experiments provide strong support for the model.

Since the brands generated from memory during brand evaluation formation are partially dependent on the accessibility of each brand, contextual factors such as usage situations, in-store displays and media advertising will affect which brands are generated and the resulting evaluation of the brand being evaluated.

REFERENCES

Bargh, John A. and Paula Pietromonaco (1982) "Automatic Information Processing and Social Perception: The Influence of Trait Information Presented Outside of Conscious Awareness on Impression Formation," Journal of Personality and Social Psychology 43:437-449.

Herr, Paul M.(1986) "Consequences of Priming: Judgment and Behavior", Journal of Personality and Social Psychology 51:1106-1115.

Herr, Paul M. (1989) "Priming Price: Prior Knowledge and Context Effects", Journal of Consumer Research 16:67-75.

Kahneman, Daniel and Dale T. Miller (1984) "Norm Theor: Comparing Reality to its Alternatives", Psychological Review 93:136-153.

Lynch, John G., Dipankar Chakravarti and Ansree Mitra (1991) "Contrast Effects in Consumer Judgments: Changing Mental Representations or Anchoring Rating Scales", Journal of Consumer Research 18:284-297.

Martin, Leonard L. (1986) "Set/Reset: Use and Disuse of Concepts in Impression Formation", Journal of Personality and Social Psychology 51:493-504.

Martin, Leonard L., John J. Seta and Rick A. Crelia (1990) "Assimilation and Contrast as a Function of People’s Willingness and Ability to Expend Effort in Forming an Impression", Journal of Personality and Social Psychology 59:27-37.

Myers-Levy, Joan and Brian Sternthal (1993) "A two Factor Explanation of Assimilation and Contrast Effects", Journal of Marketing Research 30:359-368.

Parducci, Allen (1965) "Category Judgment: A Range-Frequency Model", Psychological Review 72:407-418.

Schwarz, Norbert and Herbert Bless (1992) "Constructing Reality and its Alternatives: An Inclusion/Exclusion Model of Assimilation Effects in Social Judgment", in The Construction of Social Judgments. Leonard L. Martin and Abraham Tessor (eds.), Hillsdale, N.J.: Erlbaum, pp. 217-245.

Stapel, Diederick, Willem Koomen and Aart S. Velthuijsen (1998) "Assimilation or Contrast: Comparison Relevance, Distinctiveness and Impact of Accessible Information on Consumer Judgments", Journal of Consumer Psychology 7:1-24.

Stapel, Diederick, Willem Koomen and Joop Van Der Pligt (1997) "Categories of Category Accessibility: The Impact of Trait Concept Verses Exemplar Priming on Person Judgment", Journal of Experimental Social Psychology 33:47-76.

 

"THE EFFECT OF SEQUENCE, DOMAIN MATCH, AND SET SIZE OF CONTEXT PRODUCTS ON TARGET PRODUCT EVALUATIONS AND PURCHASE INTENTIONS"

Rajagopal Raghunathan and Julie R. Irwin, University of Texas

We often view a series of goods, services or ideas before deciding whether (and what) to purchase. We may browse among the product offerings ourselves (as through a catalog), or may be shown items (by a salesperson or a broker). Along the way, we are likely to have varying affective responses to products; these responses form a context within which happiness with a future product experience resides. The choice of what and how many items to include, as well as how to arrange them has become a focus of managerial interest, especially in retailing (e.g., Business Week 1999; Gladwell 1996; Underhill 1999) and in the design of web pages (e.g., Menon and Kahn 1999).

In this presentation, we show the results of three manipulations of the context (the browsing experience) on happiness with a target product. The contexts were series of either automobiles or vacation spots. They varied in how pleasing they were (e.g., visiting Detroit versus visiting Paris), whether they improved or worsened in quality, how many of them there were, and whether they were from the same domain as the target product.

We found that subjects contrasted the context with the target product when the products matched: seeing many nice vacation spots made an average vacation spot seem less pleasing. They assimilated the context when there was domain mismatch: seeing many nice vacation spots made an average automobile more pleasing.

Our primary contribution in this project was the addition of context size and domain match manipulation to our basic finding. If the context size was large (e.g., people browsed through 24 vacation sites), then assimilation was enhanced. We explain, and provide evidence, that this interction is mediated by mood: seeing more pleasant items enhances mood more, and this enhanced pleasure is associated with the target. On the other hand, context size did not affect contrast. We explain and provide evidence that contrast should be affected only by the average of the items, not by how many items are included.

Finally, in the last and focal experiment, we combined all of these effects and manipulated domain match using instructions. Instead of presenting a set of vacation spots and then a car (or vacation spot), we presented a set of vacation spots and then a picture of a room labeled either "a room in an apartment that you could rent near the university campus" or "a room in a vacation rental where you could stay while on vacation". This simple instructional manipulation induced all of our previous results. There was contrast with the vacation spots when labeled as a rental (domain match), assimilation when labeled as a vacation rental (domain mismatch). Furthermore, contrast did not depend on context size, but assimilation did.

 

"COMPARISON VERSUS CORRECTION PROCESSES: EFFECTS OF CONSTRUCT ACCESSIBILITY ON JUDGMENT"

Susan H. Jung and Angela Y. Lee, Northwestern University

While advertisers often try to impose restrictions on the number of ads from competing brands appearing in close proximity to their own ads, they are less concerned about the effect of advertising of products from different categories. Thus, an interesting question is: Do exposures to ads from a different category affect consumers’ judgment of a target brand? The present research examines the effects of exposures to product information from the same versus different category on the judgment of a target brand. We also attempt to understand the process by which these effects occur.

Numerous studies have shown that the priming of a construct in memory may affect participants’ attitude by shifting it towards the accessible construct in memory, resulting in an assimilation effect; or away from the accessible construct, resulting in a contrast effect. Whereas assimilation effects on judgment are thought to occur automatically, and have been observed without conscious awareness or intention (e.g., Bargh and Pietromonaco 1982), contrast effects are thought to require cognitive resources (e.g., Martin, Seta and Crelia 1990). More recent findings, however, suggest that contrast effects may be the result of two different processes: 1) The prime may be used as a standard of comparison for judging the target resulting in a comparison contrast effect (e.g., Herr 1989). 2) Alternatively, the prime may be perceived as having a biasing influence on the judgment of the target, thus participants correct their judgment away from the prime, resulting in a correction contrast (Martin et al. 1990). It has been suggested that a comparison contrast is mediated by the automatic and unconscious use of the prime that serves as some standard, while a correction contrast is produced through more effortful processes.

The primes are usually an exemplar that belongs to the same category as the target, a characteristic of the target, or an affective state (e.g., positive mood). Few studies have used exemplars from a different category as a prime (see Stapel, Koomen and Velthuijsen 1998 for an exception). The objective of the present research is to examine the type of context effects that may arise when an exemplar from either the same or a different category than the target is used as a prime. Specifically, we investigate if the contrast effect observed may be the result of a comparison contrast or a correction contrast.

Study 1 was conducted to examine the priming effects of an extreme same-category exemplar on judgment of a fictitious target under high- and low-involvement conditions. Participants reviewed two ads, a prime, followed by the target. The results show that participants’ evaluation of an ambiguous target (a fictitious hotel in Florida) assimilated towards the exemplar (The Four Seasons Hotel) in the low-involvment condition, but contrasted away from the exemplar under the high-involvement condition. The finding that a contrast occurred only when participants expended more cognitive resources suggests that the contrast effect observed is the result of a correction process.

Next, we examined the priming effects of an extreme different-category exemplar on the target, using similar procedure as in Study 1. The results of Study 2 show that participants’ evaluation of the target (a Florida hotel) contrasted away from the exemplar (Porsche) under both high- and low-involvement condition. This suggests that the different-category exemplar was used as a standard of comparison, and the contrast effect observed is the result of an automatic process that does not require cognitive resources.

The results of the two studies suggest that ads from either the same or different product categories may have an effect on consumers’ attitude toward a subsequent product. Further, ads of competing brands in the same category are more likely to prompt consumers to correct their attitude towards the product as compared to ads from a different product category. A third study (Study 3) is currently is currently underway to test the robustness of these findings. Study 3 is also designed to examine if these effects of same- versus different-category primes on judgment may be moderated by the extremity of the prime.

 

"A MODEL OF BRAND EVALUATION FORMATION WITH MEMORY BASED CONTEXT EFFECTS"

Peter A. Dacin, Queen’s University

Andrew A. Mitchell, University of Toronto

In the proposed model, based on Norm Theory (Kahneman and Miller 1986) and the Inclusion/Exclusion Model (Schwarz and Bless 1992), consumers generate related product exemplars from memory when forming a brand evaluation for a target brand. If the description of the target brand is unambiguous, the generated product exemplars are used to form standards of comparisons which are then used to evaluate the attribute levels of a target brand. These attribute levels are then summed to form a brand attitude. These processes require few cognitive resources and appear to occur below conscious awareness.

Since the products that are generated from memory during attitude formation are partially dependent on the accessibility of each product, contextual factors such as usage situations, in-store displays and media advertising will affect which products are generated and the resulting evaluation of a particular brand.

A series of experiments are used to test the different aspects of the model. The first experiment tests whether the generation of product exemplars produces contrast effects on brand evaluation, whether the generation of product exemplars is an activation as opposed to an inhibilitory process, and whether multiple product exemplars are generated. The results of the first experiment provide empirical support of these aspects of the model.

The second experiment tests whether the primed product exemplars affect both perceived attribute levels and attribute evaluations when the target brand is ambiguous, attribute evaluations when the target brand is unambiguous and whether the sum of the attribute evaluations predicts the brand attitudes which are formed.

The third experiment tests whether high accessible product exemplars affect brand attitude formation in both an immediate and a delay condition. This provides strong evidence that consumers generate product exemplars in forming brand attitudes.

Finally, the last experiment tests whether the generation of product exemplars and the formation of standards of comparisons are low-level cognitive processes that require little or no cognitive resources. The results of this study provide strong support for this aspect of the model.

In summary, the results of a series of experiments which test different aspects of the model provide strong support for the model of attitude formation.

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