Cue Utilization in the Quality Perception Process

Jerry C. Olson, The Pennsylvania State University [Assistant Professor of Marketing, College of Business Administration.]
Jacob Jacoby, Purdue University [Associate Professor of Psychology.]
[ to cite ]:
Jerry C. Olson and Jacob Jacoby (1972) ,"Cue Utilization in the Quality Perception Process", in SV - Proceedings of the Third Annual Conference of the Association for Consumer Research, eds. M. Venkatesan, Chicago, IL : Association for Consumer Research, Pages: 167-179.

Proceedings of the Third Annual Conference of the Association for Consumer Research, 1972      Pages 167-179

CUE UTILIZATION IN THE QUALITY PERCEPTION PROCESS

Jerry C. Olson, The Pennsylvania State University [Assistant Professor of Marketing, College of Business Administration.]

Jacob Jacoby, Purdue University [Associate Professor of Psychology.]

[The authors gratefully acknowledge the assistance of Bruce Anderson in conducting this study.]

From an information theoretic perspective, products and brands are conceived to consist of an array of cues (e.g., price, brand name, packaging, color, etc). Moreover, each cue provides a basis for developing various impressions of the product itself (cf. Cox, 1962; Jacoby, Olson, & Haddock, 1971). One impression of considerable importance in consumer behavior is the perceived quality of the product or brand. In addition to being of interest in its own right, quality perception seems to be strongly related to actual purchasing behavior, especially brand loyalty (cf. Jacoby, 1971). [For example, Jacoby and Hollander (1972) obtained correlations of .775 (for orange juice), .763 (for bathroom tissue), .823 (for cooking oil), and .803 (for ketchup) for 97 housewives who responded (on 9-point semantic differential scales) to the following two questions: (a) "The brands of some products vary greatly in quality while the brands for other products hardly vary at all. About HOW LARGE do you feel the DIFFERENCES in QUALITY are BETWEEN BRANDS of bathroom tissue? Circle the number on the scale below that comes closest to your feeling on the matter." and (b) "When you do your shopping, how IMPORTANT is it for you to get the brand of (orange juice) that you USUALLY BUY? Circle the number on the scale below that comes closest to your feeling on the matter."]

One cue to perceived product quality that has generated considerable recent experimental interest is price (Andrews & Valenzi, 1971; Enis & Stafford, 1969; Gardner, 1971; Jacoby, Olson, & Haddock, 1971; McConnell, 1968; Olson, 1972; Peterson, 1970; Rao, 1971; Szybillo & Jacoby, 1972; Valenzi & Andrews, 1971; and White, 1966). Other cues examined experimentally include brand image (Allison & Uhl, 1964; Jacoby, Olson, & Haddock, 1971), store image (Enis & Stafford, 1969; Szybillo & Jacoby, 1972), and actual composition differences (Jacoby, Olson, & Haddock, 1971; Szybillo & Jacoby, 1972; Valenzi & Andrews, 1971).

Examination of these studies leads to the conclusion they are basically exploratory and atheoretical. Although they are more realistic and sophisticated than the earlier studies of perceived product quality (cf. Laird, 1932; Leavitt, 1954; Tull, Boring, & Gonsior, 1964) in that they include more than a single quality cue, they nonetheless lack a theoretical framework and overall direction. It is the thesis of the present paper, to be developed later, that this lack of model and direction greatly hinders the study of cue utilization in the quality perception process.

One reason for the current state of affairs may be that all of the quality perception studies have utilized experimental designs which, although particularly useful for establishing cause and effect relationships, tend to drastically limit the number of cues that can be examined in combination. Witness the fact that the above-cited experiments have studied only one or two, occasionally three, and at the most four, cues simultaneously. Perhaps such limitations are understandable given the difficulty of interpreting higher order interactions when they occur.

Since only a few of the many possible cues can be incorporated into any one experiment, the critical question becomes which cues to examine. In answering, most researchers seem to have chosen quality cues for their investigations on the basis of intuition or specific interest in a particular cue (e.g., price). However, if one's purpose is to assess the relative effects of those cues which most strongly influence quality perceptions (hopefully with the intention of developing a comprehensive understanding of quality cue utilization), then it becomes imperative that the most salient cues first be identified so that they can be incorporated into one's experiment.

Accordingly, the present exploratory research attempted to answer the following question: Of those product attributes (or cues) that consumers take into consideration in deciding on a brand purchase, which cues are felt to have the greatest influence upon perceptions of brand quality? Since certain cues may be influential in some product categories but not in others, the present investigation sought to provide information concerning the strength of several potential quality cues across several product categories.

METHOD

Subjects

Sixty-nine housewives living in seven different socio-economic and geographic segments of Lafayette, Indiana, served as subjects.

Instrument

Data were collected using a self-administered questionnaire. Comprehensive lists of product-related attributes or potential quality cues were generated separately for five products selected to cover a wide range of prices and usage characteristics. In all, there were 15 product attributes for living room rug, 14 for hair dryer, 12 for ground coffee, 15 for shampoo, and 13 for aspirin tablets. Several of these attributes were product-specific (e.g., "type of fiber" for living room rugs); others were applicable to several products (e.g., "packaging" for aspirin, coffee, and shampoo); while others were common to all five products (e.g., "price" and "brand name"). Each product and its accompanying randomized list of product attributes were reproduced as a separate page in a six-page questionnaire. Excluding the first page of introductions, the order of the pages within the questionnaire was randomly determined for each subject.

Task

The housewives were asked to imagine that they were about to purchase each of the five products. For each product, the subject first checked those factors that she would take into consideration when making a decision about which brand of the product to purchase. Second, each subject rank ordered those factors she would consider in terms of their usefulness and importance as indicators of (or cues to) product quality.

RESULTS

The data were analyzed separately for each of the five products. Two indices were computed for each product attribute: (a) the number of subjects who stated that they considered the attribute in making a purchase decision, and (b) the average ranking of each attribute (based on those subjects considering it) in terms of its perceived usefulness and importance as an indicator of or cue to product quality.

Parenthetically, these data are relevant to an issue regarding the number of important, desired product attributes used as evaluative criteria by consumers in choosing among alternative brands during the purchase decision process (cf. Engel, Kollat, & Blackwell, 1968, pp. 430-431). In the present research, the first cue index (i.e., frequency of consideration of a cue in a purchase decision) can be considered as a rough measure of "cue use as an evaluative criterion." The average number of product attributes these consumers reported considering in each product category was: hair dryer = 5.97; living room rug = 7.17; ground coffee = 4.64; shampoo = 5.13; and aspirin tablets = 4.51. These data suggest that, for most consumers, several (rather than one or two) product attributes are considered during the evaluation process. Given this likelihood of multiple cue use, it becomes even more crucial to determine relative cue importance. It must be noted, however, that the data cited above are based upon self-reports rather than actual cue use and, therefore, may be inaccurate estimates of the "true" number of product attributes used as evaluative criteria.

Table 1 presents the four cues in each product category with the highest average ranks in terms of perceived importance or accuracy as indicators of product quality. One can easily detect a lack of commonality across product categories in terms of the specific attributes thought to be accurate indicators of product quality. This is especially evident for the most important cue in each product category. Accurate cues to quality seem to be highly product specific.

A brief digression is warranted at this point. Thus far, the results of this research provide yet another example of a common conclusion for consumer behavior research in general, and quality research in particular. That is, one's findings are likely to be specific to the type of product and/or consumer investigated. In such cases, generalizations beyond the product or consumers examined are of dubious validity. One strategy in such a situation is to develop concepts or constructs at a more abstract level than the observable data, which hopefully can explain confusing or conflicting empirical results and may enable one to generalize beyond the immediate situation.

After some thought regarding the lack of commonality in Table 1, such a potentially useful conceptual dimension was hypothesized. Note in Table 1 that the first or second most important cue to perceived quality generally was one which could be described as "intrinsic to the product"--i.e., a product attribute which cannot be changed or experimentally manipulated without also changing the physical characteristics of the product itself. In contrast, extrinsic cues are product-related attributes which are not a part of the physical product. For example, "nature of the fiber" (for living room rug), "taste" (for coffee), and "special ingredients" (for aspirin) are all integral, intrinsic components of the physical products. Conversely, product-related factors such as "guarantees" (for living room N g), "manufacturer's reputation" (for hair dryer), and "brand name" (for ground coffee) are extrinsic cues in that they are not a part of the physical product. In contrast to intrinsic cues, if extrinsic cues are experimentally manipulated the physical characteristics of the product need not necessarily change. The present study found that intrinsic, rather than extrinsic, cues were generally perceived to be the most accurate indicators of brand quality.

Another interesting finding (see Table 1) is that price does not appear among the best four cues to perceived quality for any of the five products. However, Table 2, which presents, for each product, the four attributes with the highest frequency of self-reported consideration in a brand purchase decision, reveals that price was a relatively important factor to consider when making a brand purchase in each of the five product categories. Thus, although these housewives felt that price was an important factor to be considered in making purchase decisions among brands, they did not feel price to be an especially accurate indicator of brand quality. Note that virtually the same statement can be made regarding the product attribute/quality cue of "brand name''

TABLE 1

FOUR MOST INFLUENTIAL CUES TO PERCEIVED BRAND QUALITY

TABLE 2

FOUR MOST IMPORTANT FACTORS TO CONSIDER WHEN PURCHASING A BRAND IN FIVE PRODUCT CATEGORIES

Finally, although there was a lack of agreement across products in terms of both important quality cues and considered product attributes, some general correspondence between the two factors was evident within product categories. Moreover, this consistency extended beyond the most important attributes throughout the entire list. Table 3 presents Spearman rank order correlations between product attribute consideration in a purchase decision and attribute importance as a cue to quality, computed across all attributes for each of the five product categories. Clearly, the relationships are fairly strong, indicating a general tendency for consumers to use as cues to product quality those product attributes which are important considerations in purchase decisions, or vice versa. For the researcher, this relationship is a useful aid in identifying important quality cues, since salient cues are also likely to be product-related factors important in the general purchase decision-making process.

TABLE 3

RELATIONSHIP BETWEEN PRODUCT ATTRIBUTE CONSIDERATION IN A PURCHASE DECISION AND IMPORTANCE AS A CUE TO PRODUCT QUALITY

CONCLUSIONS AND DISCUSSION

Several tentative conclusions can be drawn from the present exploratory investigation. First, consumers may use several (e.g., 4-7) product attributes as evaluative criteria in brand-choice decisions rather than only 1 or 2. This finding suggests that future researchers should include several quality cues in combination in their experimental designs. Interestingly, prior research indicates that cues in combination may not have simple linear effects upon quality judgments, but rather may evidence non-additive effects through complex interactions between cues. For example, Olson (1972) determined that of the 11 quality perception studies in which cue interaction could occur, 9 investigations found such effects (e.g., Andrews & Valenzi, 1971; Enis & Stafford, 1969; Jacoby, Olson, & Haddock, 1971).

Second, given the finding that intrinsic cues were perceived to be somewhat more accurate indicators of product quality than were extrinsic cues, a tentative conclusion is that intrinsic cues have a more powerful effect upon judgments of quality than do extrinsic cues. This suggests that certain frequently studied extrinsic cues such as price, brand name, and store image are not expected to exert relatively strong effects on quality perceptions--unless, of course, more potent cues are omitted from the study. Support for this conclusion is provided by work based upon and subsequent to the present study (Olson, 1972; Szybillo & Jacoby, 1972), as well as earlier experiments which included both intrinsic and extrinsic cues (Andrews & Valenzi, 1971; Jacoby, Olson, & Haddock, 1971). All of these investigations found that intrinsic cues (e.g., physical product differences such as taste) had stronger effects upon quality perceptions than did extrinsic cues (e.g., price, store image, brand name). Moreover, a reexamination of the experimental literature to date reveals that in most of the instances in which extrinsic cues had strong effects on quality perceptions, the more powerful intrinsic cues had been omitted from the study (e.g., Enis & Stafford, 1969; McConnell, 1968; Peterson, 1970. In addition, certain data indicate that the "true" state of quality cue influence may be even more complex than described above. Results obtained by Jacoby, Olson, and Haddock (1971) suggest that the effect of extrinsic cues on perceived quality may be evidenced primarily through interactions with intrinsic cues rather than through main effects. For these reasons, therefore, it is suggested that future attempts to experimentally assess the impact of specific extrinsic cues on perceived quality should also incorporate intrinsic cues in the design (especially the kinds of cues obtained through actual product usage--e.g., taste, appearance, fit, etc), if the results are to be externally valid and maximally meaningful.

A third conclusion, related to the second and partially supportive of it, is that, while price (an extrinsic cue) is an important factor to consider in a purchase decision, the price cue generally is not perceived to be as good an indicator of quality as are certain intrinsic cues (or other extrinsic cues). Support for this statement may be found in recent investigations which have found that price, when combined with other cues, has an insignificant impact on quality judgments (e.g., Jacoby, Olson, & Haddock, 1971; Szybillo & Jacoby. 1972: Valenzi & Andrews. 1971).

The final conclusion (actually a combination of several) is based upon both the present data and a critical review of the previous quality perception literature (see Olson, 1972). After a substantial amount of experimental effort (19 studies), we still know very little about the answers to the following questions: (a) Which product attributes, from among the many available, are chosen by the consumer as surrogate indicators of product quality? tb) Why are these particular cues chosen for use and others are not? (c) Once cues are chosen, how are they used (i.e., combined) to form a quality judgment? (d) Why are the cues combined in that manner--i.e., what factors determine cue usage and cue impact upon the quality judgment? It seems that our present lack of understanding is at least partially due to the atheoretical, exploratory basis upon which most investigations, including the present one, have been designed (see Cox, 1962, for a notable exception).

An obvious conclusion, therefore, is that the topic of quality judgment formation requires a theoretical model or conceptual framework which defines and describes the critical, determining factors in the quality perception process. At least three major elements are involved in this process--product, consumer-types, and quality cues (product-related attributes)--each of which may require its own model/framework. An attempt has been made to develop each of these models (Olson, 1972); however, the remainder of this paper will briefly present only that model concerned with quality cues.

A MODEL OF CUE CHOICE AND USAGE

A basic assumption involved in the development of a model dealing with cue choice and usage is that, instead of simply examining whether specific cues affect a quality judgment for a certain product, a more fruitful research strategy is to begin at a more basic, abstract level and attempt to identify the factors which influence (a) whether or not a cue will be used in the judgment process, and (b) if so, the magnitude of cue impact on the final judgment. Such a conceptual model would provide a basis from which researchers could predict which cues affect quality perceptions and the relative strength of their effects.

Previously, few researchers have developed abstract constructs -with which to describe and categorize quality cues ln terms of dimensions relevant to cue usage and impact. Exceptions are Cox (1962) and Tull, Boring, and Gonsior (1964), although their work has not been followed up by other researchers. Two dimensions of the present framework were originally suggested by Cox and are modified and extended here.

The model presented below consists of two major dimensions (cue predictive value and cue confidence value) and, in addition, a dichotomous classificatory factor (cue intrinsicness/extrinsicness). Any potential quality cue can be described in terms of these three variables.

Predictive Value

Cue predictive value (PV) is defined as the degree to which an individual consumer associates a cue with product quality. That is, cue PV is the extent to which the consumer perceives or believes that the cue is related to or is indicative of product quality. To use a quantitative analogy, PV is similar to a probability coefficient which can take on values only between .00 (i.e., no probability that the cue is associated with product quality) and +.99 (l.e., near certainty that a cue is predictive of product quality).

It must be emphasized that PV is concerned only with the generic cue (i.e., price as a general cue, not specific levels of price) and the perceived predictive relationship between the generic cue and product quality. Although it is recognized that specific levels of a cue (e.g., high or low price) may indicate various degrees of quality (e.g., high or low quality), such considerations are not incorporated into the model at its present level of development. The functional relationship between levels of a cue and degrees of perceived quality is expected to be product specific and, therefore, is left to empirical determination.

Operationally, PV can be measured by having consumers rate the degree to which they believe a cue is indicative of product quality. For example, given the extrinsic cue of price:

How accurately do you think the PRICE of a brand of beer indicates the OVERALL QUALITY of that brand of beer?

FIGURE

Such ratings are obviously product specific and, therefore, PV ratings for the same cue may vary across products. Also, it should be emphasized that PV is measured at the level of the individual consumer and thus is not necessarily constant for a given cue and product across all consumers.

Confidence Value

The second major dimension of the conceptual cue framework is the confidence value (CV) of the cue, defined as the degree to which a consumer is confident in his ability to accurately perceive and judge that cue. Cue CV is the individual consumer's self-confidence in his ability to distinguish the cue and make accurate evaluations and judgments concerning it.

Confidence value can be operationalized by asking the consumer to rate his confidence in his judgment or evaluation of a given cue. For example, Riven the intrinsic cue of flavor:

How confident are you in your ability to perceive and evaluate differences in the beer samples in terms of their FLAVOR?

FIGURE

Like PV, the CV of a cue is rated relative to a specific product by an individual consumer. For instance, a consumer may perceive a high CV for the cue of "taste" relative to wine, but a low CV for the "taste cue" relative to beer. In addition, the CV of a particular cue relative to a single product will vary across consumers as a function of differential learning experiences, cue knowledge, and cue familiarity. Furthermore, the average consumer is likely to perceive different CV's for the various cues related to a single product. Unless the consumer is an "expert" concerning that product, he is not likely to have high CV for all possible quality cues.

Awareness

An implicit aspect of the two-dimensional cue framework is awareness. Since the model has a distinctively cognitive orientation, it must be noted that the model applies only to those cues of which the consumer is consciously aware. That is, even though cues perhaps influence quality perception in a sub- or unconscious manner, the present model does not deal with such processes. Thus, in order to apply the model it is first necessary to identify which cues -to perceived quality, of the many potential cues, an individual is aware of.

Intrinsic-Extrinsic

The intrinsic-extrinsic dimension (as described earlier) of the quality cue model is a dichotomous classificatory variable which has no direct effect on the process of cue utilization, but rather may be useful as a basis for determining the relative order in which individual cues from a set will enter the judgment process.

Combining PV and CV

The dimensions of PV and CV are both necessary to meaningfully describe and categorize quality cues in terms of cue utilization. It is assumed that these two dimensions alone (as mediated by the third classificatory variable-intrinsicness/extrinsicness) are sufficient to predict and account for quality cue choice and magnitude of cue impact upon quality perceptions. Based upon the ease with which one can imagine specific quality cues which are high on one dimension and low on the other, and vice versa, it seems reasonable, on a conceptual basis and pending empirical evidence, to treat PV and CV as orthogonal dimensions.

Hypotheses

Several working hypotheses have been derived from the conceptual model of cue utilization (see Olson, 1972), and others may be formulated based on future research. Following are three of the major hypotheses:

1. It is predicted that cue PV and CV in combination do not have an independent, additive effect on the probability of cue utilization or on the magnitude of cue effect. Rather, cue PV and CV are hypothesized to have an interactive effect on both aspects of the quality perception process. The general shape of this interaction is presented in Figure l. Stated verbally it reads, only when a cue has both high PV and high CV does it tend to be used in the quality judgment process and have a strong effect on the final judgment.

2. It is also hypothesized that, for most products, consumers generally believe intrinsic cues are more accurate indicators of product quality (i.e., have high Pt than are extrinsic cues. Therefore, given that both intrinsic and extrinsic quality cues are available to the consumer and are perceived by him, it is predicted that intrinsic cues are used more often and, when used, have a greater effect upon quality perception than do extrinsic cues.

3. Finally, as a more specific extension of the previous hypothesis, it is predicted that the use of extrinsic (indirect) cues to product quality is related to the PV and CV of those intrinsic cues cognitively available to the consumer. Specifically, the tendency to use extrinsic cues in the formation of quality judgments is hypothesized to be an interactive function (see Figure 2) of the general PV and CV levels of the available intrinsic cues. Stated verbally it reads, extrinsic cues have a greater tendency to be used when available intrinsic cues have low PV, low CV, or both, and a lesser tendency to be used when intrinsic cues have high PV and CV.

FIGURE 1

PREDICTED PV X CV INTERACTION EFFECT ON CUE USAGE AND CUE IMPACT.

FIGURE 2

TENDENCY TO USE EXTRINSIC CUES AS A FUNCTION OF THE PV ANT CV OF AVAILABLE INTRINSIC CUES.

SUMMARY

This paper has presented an exploratory study intended to provide information regarding general relative importance of specific cues to product quality. Based upon the results of this study and a critical review of the perceived quality literature, a serious lack of theoretical and conceptual direction was noted in the quality research area. In an attempt to provide such a direction, we have briefly presented a conceptual model dealing with the process of quality cue utilization. Several working hypotheses derived from this framework were also briefly discussed. One can hope that researchers interested in developing an understanding of the process of quality perception will adopt a more abstract, conceptual approach (either the one presented here or one of their own construction) in future investigations of perceived product quality.

REFERENCES

Allison, R. I., & Uhl, K. P. Influence of beer brand identification on taste perception. Journal of Marketing Research, 1964, 1, 36-39.

Andrews, I. R., & Valenzi, E. R. Combining price, brand name, and store cues to form an impression of product quality. Proceedings of 79th Annual Convention of the American Psychological Association, 1971, 6, 649-650.

Cox, D. F. The measurement of information value: A study in consumer decision-making. In W. S. Decker (Ed.), Emerging concepts in marketing. Chicago: American Marketing Association. 1962. 413-421.

Engel, J. F., Kollat, D. T., & Blackwell, R. D. Consumer behavior. New York: Holt, Rinehart, & Winston, 1968.

Enis, B. M., & Stafford, J. E. Consumers perception of product quality as a function of various informational inputs. In P. McDonald (Ed.), Marketing involvement in society and the economy. Chicago: American Marketing Association, 1969, 340-344.

Gardner, D. M. Is there a generalized price-quality relationship? Journal of Marketing Research, 1971, 8, 241-243.

Jacoby, J. A model of multi-brand loyalty. Journal of Advertising Research, 1971, 11, (3). 25-31.

Jacoby, J., & Hollander, S. W. Using dollar-metric scaling procedures to measure multi-brand loyalty. Unpublished manuscript, 1972.

Jacoby, J., Olson, J. C., & Haddock, R. A. Price, brand name, and product composition characteristics as determinants of perceived quality. Journal of Applied Psychology, 1971, 55, 570-579.

Laird, D. A. How the consumer estimates quality by subconscious sensory impressions. Journal of Applied Psychology, 1932, 16, 241-246.

Leavitt, A. J. A note on some experimental findings about the meaning of price. Journal of Business, 1954, 27, 205-210.

McConnell, J. D. The effect of pricing in an experimental setting. Journal of Applied Psychology, 1968, 52, 331-334.

Olson, J. C. Product quality perception: A model of quality cue utilization and an empirical test. An unpublished doctoral dissertation, Purdue University, 1972.

Peterson, R. A. The price-quality relationship: Experimental evidence. Journal of Marketing Research, 1970, 7, 525-528.

Rao, V. R. Salience of price in the perception of product quality: A multidimensional measurement approach. In F. C. Allvine (Ed.), Marketing in motion/Relevance in marketing. Chicago: American Marketing Association, 1971, 571-577.

Szybillo, G. J., & Jacoby, J. The relative effects of price, store image and composition differences on product evaluation. Purdue Papers in Consumer Psychology, Department of Psychological Sciences, Purdue University, 1972, Paper No. 123.

Tull, D. S., Boring, R. A., & Gonsior, M. H. A note on the relationship of price and perceived quality. Journal of Business, 1964, 37, 186-191.

Valenzi, E. R., & Andrews, I. R. Effect of price information on product quality ratings. Journal of Applied Psychology, 1971, 55, 87-91.

White, I. S. The perception of value in products. In J. Newman, On knowing the consumer. New York: John Wiley, 1966.

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