Impact of Information on Preference and Perception

Jim McCullough, University of Arizona
Doug MacLachlan, University of Washington
Reza Moinpour, University of Washington
ABSTRACT - Information about stimuli is provided to subjects in an attempt to alter perception and preference. Perceptual and hedonic structures appear to be resistant to change-inducing information even when perceptions of individual brands are altered. The results support previous findings of stability in perceptual structure and suggest that emphasis on attributes known to be salient to consumers may be a more effective promotional approach than attempting to alter the importance consumers attach to different product characteristics.
[ to cite ]:
Jim McCullough, Doug MacLachlan, and Reza Moinpour (1982) ,"Impact of Information on Preference and Perception", in NA - Advances in Consumer Research Volume 09, eds. Andrew Mitchell, Ann Abor, MI : Association for Consumer Research, Pages: 402-405.

Advances in Consumer Research Volume 9, 1982      Pages 402-405

IMPACT OF INFORMATION ON PREFERENCE AND PERCEPTION

Jim McCullough, University of Arizona

Doug MacLachlan, University of Washington

Reza Moinpour, University of Washington

ABSTRACT -

Information about stimuli is provided to subjects in an attempt to alter perception and preference. Perceptual and hedonic structures appear to be resistant to change-inducing information even when perceptions of individual brands are altered. The results support previous findings of stability in perceptual structure and suggest that emphasis on attributes known to be salient to consumers may be a more effective promotional approach than attempting to alter the importance consumers attach to different product characteristics.

INTRODUCTION

Promotional activity in marketing is undertaken to influence consumer behavior in the marketplace. A common objective of these actions is change of consumer preference for particular products through alteration of consumer perception and preference structures. This objective can be accomplished through provision of information about specific products or through discussion of the important characteristics of the specific product class. In order to determine which of these approaches allows the consumer to use information provided in promotional campaigns most effectively, it is necessary to understand how preferences are related to perceptions and how information is used to effect change in preference and perception.

Previous studies have suggested that information can be used by consumers to produce change in a variety of ways (MacLachlan et al., 1977); in addition they have indicated the potential of longitudinal analysis of preference and perception for measuring the effects of marketing activity (Moinpour et al., 1976, McCullough et al., 1979, 1981). This study continues the investigation of the effect of persuasive communication on perception ant preference using longitudinal analysis. In this context, it also examines the effect of variation in information format on perception and preference.

CHANGES IN PERCEPTION AND PREFERENCE

Persuasive communication as it is most commonly used in marketing is directed at changing attitudes about specific brands in a product class, usually with the objective of generating a more favorable attitude toward a particular product. The changes that might be induced through this type of communication can be characterized as either spatial, those changes influencing specific brands ant their interrelationships within a consistent perceptual or hedonic framework, or structural, those changes resulting from a change in the underlying framework. The mixed results reported by Moinpour, et al. (1976) and MacLachlan et al. (1977) indicated that spatial change might be easier to effect than structural change but this has not been conclusively demonstrated.

Spatial changes occur when the perception of one product changes relative to another product within a perceptual spacE. When perceptual position changes relative to an ideal point, a change in preference occurs. Spatial change is usually effected by providing information about individual stimuli, or brands, in terms of product characteristics previously identified. This type of change is typically measured by attribute scores in the basic multiattribute attitude motel. Multidimensional scaling techniques can also provide means of measuring changes relative to the positions of the stimuli within a perceptual space when the underlying dimensions of the space remain unaltered

Structural change is conceptually more complex. A change in the perceptual structure occurs when the subject either changes the dimensions of the perceptual space (i.e., uses different attributes to discriminate between stimuli), or changes the importance assigned to the underlying dimensions. This type of change is recorded in multiattribute attitude motels when the subject selects new attributes or alters the weights assigned to previously used attributes. Individual differences scaling (INDSCAL) can also be used to identify the change in dimension weights of a perceptual space when the individuals share the same perceptual framework.

Preference and perception are closely linked in these situations. That is, the dimension; used to discriminate between products generally are the same ones used to determine preference. In most studies of perception and preference, importance of perceptual dimension is termed salience while the corresponding hedonic importance (weight) is termed utility.

EXPERIMENTAL METHOD

Five groups of approximately 20 subjects (students at a southwestern university) participated in two data collection sessions one week apart during which they were told that they were testing the usefulness of new marketing research techniques. At each session, subjects were asked to rate all possible pairs of 10 toothpaste brands on nine-point dissimilarity scales and to rank the brands in order of preference.

Four groups served as experimental groups and one group was held as a control. During the second testing session the four experimental groups were given information concerning the brands prior to completing the dissimilarity and preference judgments. To insure that the messages provided had been observed, each subject was asked to record the most favorably rated brand based on the information received.

Previous work had shown the objective characteristics of abrasiveness and fluoride content to be closely related to the perceived attributes of whitening ant decay prevention --the two underlying dimensions of the perceptual space for toothpaste. As a consequence, two groups received information concerning abrasiveness (see Table 1) ant the other two groups received information concerning fluorite content (see Table 2). Within each of these pairs of groups one group received only objective, unprocessed information, while the other additionally received text material highlighting the dimension and the position of Gleem II. The latter message was not based in fact and was presented to have maximum impact on Gleem II. Thus, in a two by two design, groups of subjects in two experimental conditions received objective information concerning abrasiveness and fluoride content; subgroups in each condition were then presented either objective (general) or highlighted (emphasizing importance of the dimension and the high rating of Gleem II) text messages.

It was expected that subjects receiving the objective messages (abrasiveness or fluoride content) would change the structure of the perceptual hedonic spaces by shifting the salience or utility for related dimensions and that the highlighted message would result in a spatial change in the Position of Gleem II. [Although previous studies have shown abrasiveness/whitening and fluorite content/decay prevention to be salient, it has also been indicated that the factual information is not readily available or at least not equally processed with regard to these dimensions. In fact, the average subject appears to be more knowledgeable about the characteristic fluorite content than abrasiveness ant consequently more capable in attempting to map fluoride content into decay prevention than abrasiveness into whitening dimension. Therefore, any cue providing basic, factual information should be expected to result in a change in the perceptual structure.]

TABLE 1

ABRASIVENESS MESSAGE INFORMATION

OBJECTIVE TEXT

The Council on Dental Therapeutics of the American Dental Association has released the following results of a study of the abrasiveness/whitening of the leading brands of toothpaste.

HIGHLIGHTED TEXT

Recent American Dental Association studies have revealed that slightly abrasive toothpastes do not cause serious damage to teeth but greatly increase whitening power. The table below reports the results of a study by the Council on Dental Therapeutics of the Abrasiveness/Whitening of the leading brands of toothpaste.

Gleem II has the highest whitening power and whitens teeth better than any other toothpaste currently on the market.

TABLE 2

FLUORIDE CONTENT MESSAGE INFORMATION

OBJECTIVE TEXT

The Council on Dental Therapeutics of the American Dental Association has released the following results of a study of the fluoride contents of the leading brands of toothpaste.

HIGHLIGHTED TEXT

Recent American Dental Association studies have revealed that the amount of fluoride in a toothpaste influences the brand's effectiveness in preventing tooth decay. The Table below reports the results of a study by the Council on Dental Therapeutics of the fluoride contents of the leading brands of toothpaste.

Gleem II has the highest level of fluoride ant is more effective in preventing tooth decay than any other toothpaste currently on the market.

ANALYSIS

Pairwise dissimilarities were analyzed using the INDSCAL algorithm which produces group stimulus space coordinates ant individual dimension saliences. The group stimulus space coordinates ant preference rankings for each individual were analyzed using the LINMAP mixed mote option to produce individual dimensional weights or utilities.

The INDSCAL Model

The INDSCAL model assumes all subjects utilize the same attributes to discriminate between products but allows individual subjects to modify that space by applying different dimensional saliences. The group space coordinates Yjt for brand j and dimension t are transformed into individual coordinates Yijt for individual i by the application of subject's dimension weights (saliences) With the algorithm developed by Carroll and Chang (1970) simultaneously derives Xjt, ant Wit values from matrices of brand dissimilarities for all individuals usin2 the motel:

EQUATION    (1)

The LINMAP Model

The LINMAP model is one of many "utility" generating algorithms. It has the advantage of modeling decreasing marginal utility and ideal points for some attributes and constant marginal utility (vector model) for other attributes. It posits the following functional form for utility of brand k at time e:

EQUATION    (2)

where j is the set of all attributes having finite points ant i2 is the set of all attributes with infinite ideal points; Vj is the attribute "importance" (i.e., the value of the attribute in determining preference); Xjk is the value of brand k on dimension j; ant Ij is the subject's ideal point on the jth dimension (Srinivasan ant Shocker, 1973).

Results

The magnitudes of the changes in salience and utility resulting from the treatments are shown in Tables 3 ant 4. Although there were changes in individual saliences, no significant changes occurred in any group mean values for either salience or utility as a result of-the treatments. The treatment group results were not significantly different from the results for the control group previously reported by McCullough et al. (1981). Changes in preference are reported in Table 5. Some changes in mean preference scores did occur in the experimental groups, particularly regarding Gleem II, the brand receiving the most extreme information, ant Ipana, a brand generally unfamiliar ant negatively viewed by the subjects ant a brand for which perceptions are not well developed.

The displacement of stimuli in perceptual spaces is shown in Table 6. Although these movements are more difficult to interpret than shifts in salience and utility, they appear to correspond to the degree of discrepancy between the initial perception ant the information provided.

TABLE 3

MEAN DIMENSION SALIENCES

TABLE 4

MEAN DIMENSION UTILITIES

TABLE 5

CHANGES IN MEAN PREFERENCE RANKINGS

TABLE 6

INFORMATION-INDUCED SHIFTS IN BRAND POSITIONS

STUDY LIMITATIONS AND SUGGESTIONS FOR FURTHER RESEARCH

As part of a continuing investigation, one objective of this study was to examine the effect of variation in information format on perceptual structure. In this attempt, two possible limitations of the study may have contributed to the apparent ambiguity in results:

a) Given that the abrasiveness and fluoride content of toothpaste have been shown to be highly salient dimensions in previous studies, perhaps it is not surprising that attempts to make them more salient failed.

b) Additionally, since the message on abrasiveness may have presented information contrary to generally accepted positionS the study may have inadvertently confounded the measurement of salience and perceptual positions on this dimension.

Future studies should work with other product classes or manipulate less salient dimensions of the toothpaste product such as taste or texture.

CONCLUSIONS

Information about brands in a product class provided to consumers with the intention of altering their preference and perception appear to affect perceptions of individual brands within the perceptual space without changing the underlying structure of the space. Similarly, changes in preference seem to be brand specific rather than reflecting changes in the utility structure. These results are consistent with the previous studies indicating that perceptual spaces are very resistant to perturbation either through direct manipulation or through provision of additional information to subjects.

Although stronger messages, long term promotional campaigns, and environmental change might eventually impact the structure of the perceptual and hedonic spaces, these results indicate that promotion aimed at changing how the consumer views the world may be less effective than material aimed at altering the perception of an individual brand within an established framework. It appears that concentrating on product attributes with high salience or utility should be more effective than attempts to alter the importance assigned to the underlying dimensions of the product space.

REFERENCES

Carroll, J. D. and Chang, J. J. (1970), "Analysis of Individual Differences in Multidimensiona1 Scaling Via N-way Generalization of 'Eckart-Young' decomposition," Psychometrika, 35, 283-319.

MacLachlan, D. L., Moinpour, R. and McCullough, J. M. (1977), "Experimenting with Perceptual Change Strategies," Multivariate Behaviors1 Research, 12, 429-446.

McCullough, J. M., MacLachlan, D. L. and Moinpour, R. (September 1979), "Linking Preference and Perception: A Longitudinal Study," Proceedings: American Psychology Association, Division 23.

McCullough, J. M. (1981), "Temporal Links Between Preference and Perception," Advances in Consumer Research, Vol. VII, 178-181.

Moinpour, R., McCullough, J. M. and MacLachlan, D. L. (1976), "Time Changes in Perception: A Longitudina1 Application of Multidimensional Scaling," Journal of Marketing Research, R. 245-253.

Srinivasan, V., and Shocker, A. D. (1973), "Estimating the Weights for Multiple Attributes in a Composite Criterion Using Pairwise Judgments," Psychometrika, 18, 473-493.

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