Warranty and Other Extrinsic Cue Interaction Effects on Consumers' Confidence

ABSTRACT - Two experiments were performed to examine the manner in which warranty quality interacts with price and warrantor reputation in influencing consumers' confidence. Support was found for the contention that extrinsic cues are of significant value to consumers in situations where intrinsic cues have low predictive and confidence values.

Citation:

Terence A. Shimp and William O. Bearden (1980) ,"Warranty and Other Extrinsic Cue Interaction Effects on Consumers' Confidence", in NA - Advances in Consumer Research Volume 07, eds. Jerry C. Olson, Ann Abor, MI : Association for Consumer Research, Pages: 308-313.

Advances in Consumer Research Volume 7, 1980     Pages 308-313

WARRANTY AND OTHER EXTRINSIC CUE INTERACTION EFFECTS ON CONSUMERS' CONFIDENCE

Terence A. Shimp, University of South Carolina

William O. Bearden, University of South Carolina

ABSTRACT -

Two experiments were performed to examine the manner in which warranty quality interacts with price and warrantor reputation in influencing consumers' confidence. Support was found for the contention that extrinsic cues are of significant value to consumers in situations where intrinsic cues have low predictive and confidence values.

INTRODUCTION

Product warranties perform important functions for marketers, serving as a persuasive sales variable (Kendall and Russ 1975) and protecting sellers from unreasonable claims by product users (Udell and Anderson 1968). However, what role they perform for consumers is unclear. Lehmann and Ostlund (1972) observed several years ago that: "Despite the widespread concern over product warranties, little research has been conducted on how consumers behave toward warranties" (p.52). The state of the warranty literature has not changed much during the interim period. The limited research, most of which studied warranties only tangentially, has been restricted primarily to examining the relative importance of warranties vis-a-vis other product cues (e.g., McClure and Ryans 1968; Roselius 1971; Olson 1972; Perry and Perry 1976).

Questions concerning how consumers use warranties in reducing risk, in forming brand evaluations, and in making product-level and brand-specific purchasing decisions remain largely unanswered. Various views and some empirical evidence have appeared, however. Research by Olson (1972) and Feldman (1976) suggests that warranties represent assurances of product quality and value to consumers. Armstrong, Kendall, and Russ (1975) have shown that warranties increase consumers' specific self-confidence, while Perry and Perry (1976) contend that warranties reduce consumer feelings of risk. According to Darden and Rao (1977), warranties serve to increase consumer satisfaction by reducing dissonance.

The nexus of these views is that the general role of warranties is one of enhancing product attractiveness by reducing perceived risk. The specific circumstances under which warranties are particularly important are not clear, however. How warranties operate in conjunction with other extrinsic and intrinsic product cues to influence consumers' decision-making and purchase behavior is similarly obscure.

HYPOTHESIZED WARRANTY ROLE

Inherent risk (Bettman 1973) is encountered by consumers when undertaking purchasing decisions. This risk involves uncertainty about the outcome and consequences of purchase (Bauer 1960; Taylor 1974; Ross 1975) and various risk components are involved (Jacoby, Kaplan, and Szybillo 1974; Peter and Tarpey 1975). The amount of perceived risk is particularly great for new products having no established performance record. Interpersonal differences such as self-esteem, self-confidence, and product familiarity also influence the amount of perceived risk (Taylor 1974; Ross 1975; Schaninger 1976).

Various risk reduction strategies are used by consumers (Ross 1975), among which is reliance on extrinsic and intrinsic product cues as indicators of product quality and performance. Price, for example, serves as a valuable extrinsic cue when objective product features (i.e., intrinsic cues) are unavailable (Olson 1977). Similarly, a product warranty may provide consumers with an additional means of reducing performance risk: if a product does not perform up to expectations, the consumer has assurance, in principle at least, that redress is possible.

Just as the price-quality research has shown that price has its effect through interaction with other cues (Jacoby, Olson, and Haddock 1971; Monroe 1973; Olson 1977), the ability of a warranty to reduce consumers' perceived risk also is likely realized in conjunction with other extrinsic cues. Price of the product and reputation of the warrantor are two cues that are expected to be particularly important in interacting with the warranty to affect consumers' perceived risk and enhance confidence. [Risk reduction and confidence enhancement are tautological, as the latter is simply the inverse of the former. As a matter of preference, confidence enhancement (or just confidence) will be used throughout the paper.]

The role of price has been detailed elsewhere (e.g., Olson 1977) and need not be discussed here. However, the influence of warrantor reputation on confidence requires some explanation. The notion of warrantor or company reputation is essentially equivalent to the source credibility construct. It is well-established that source credibility is a critical component of communication effectiveness (e.g., Craig and McCann 1978; Sternthal, Dholokia, and Leavitt 1978). The purchase and use of a well-known brand with a reputation for quality is an effective risk-reduction strategy (Engel, Kollat, and Blackwell 1973). Indirect support for this proposition is provided in several studies (Levitt 1967; Montgomery 1975; Simon 1970).

When considered in isolation, it would be expected that consumer confidence would increase directly with improvements in warranty quality. For example, a consumer who perceives considerable risk in purchasing an appliance would be expected to be more confident if the product received full warranty coverage for, say, five years than if it had limited coverage for only one year. However, this simple, zero-order relationship is very likely modified by the price of the warranted product along with the reputability of the warrantor. [For simplicity, the following discussion will concentrate on warrantor's reputation only. However, the role of price in interaction with warranty quality on consumers' confidence is expected to be similar to that for reputation.]

In particular, for a product warranted by a company with a well-established reputation, the relation between warranty quality and consumer confidence should be monotonically increasing. On the other hand, for a company perceived by consumers as low in reputability, the relationship is expected to be nonmonotonic, with confidence leveling off or even decreasing as the warranty quality increases (see Figure 1). The presumed reason for this is the initiation of a "too good to be true" perception. That is, the consumer's credulity is exceeded when a low reputability firm promotes an outstanding warranty. As such, improvements in warranty quality may be relatively ineffective and possibly even boomerang as consumers mistrust the warranty claim; the result being a diminution rather than enhancement of consumers' confidence.

FIGURE 1

HYPOTHESIZED INTERACTION BETWEEN WARRANTY QUALITY AND REPUTATION OF WARRANTOR

The research discussed below was performed to test this presumed interaction between warranty quality and warrantor reputation. The price level of tile experimental products was manipulated also, and was expected to interact with warranty quality in the same fashion postulated for warrantor reputation.

METHODS

Overview

Two experiments were performed with the second one serving as a systematic replication of the first (Carlsmith, Ellsworth, and Aronson 1976). Separate samples of college students were used for each experiment. The exploratory nature of these investigations and concern for testing theoretical relations justified the emphasis on achieving internal validity, while relegating external validity considerations to secondary importance (Cook and Campbell 1976).

Successful conduct of the experiments required that the experimental products be ones involving high inherent risk so that the warranties would represent important cues. Olson (1972) has suggested that warranties, as well as other extrinsic cues, are particularly important when a product's intrinsic cues have low confidence and low predictive values (Cox 1962).

Two hypothetical products, representing technologically feasible innovations, were selected to satisfy these desiderata. The product used in the initial experiment was described to respondents as a plastic automobile tire having advantages over conventional rubber tires by being longer-lasting and safer while providing a more comfortable ride. The other product, a multi-screen television, was selected to be more innovative and more risky. Respondents were informed that its two screens would enable simultaneous viewing of two different programs. For both products, the typical, non-expert consumer would most likely utilize extrinsic rather than intrinsic cues as a means of indicating product quality and reducing risk.

Another major concern in designing the experiments was assuring experimental realism (Carlsmith, Ellsworth, and Aronson 1976). This was particularly important since the experiments were performed under non-naturalistic conditions in classrooms. To accomplish the desired impact, respondents in each experiment were told that they were participating in a new product concept test being conducted by a marketing research firm. Debriefing feedback and responses to a questionnaire item concerning subjects' perceptions of how worthwhile this concept test would be to the manufacturer revealed that the experimental ruse did accomplish experimental realism in both experiments.

Experimental Design

In both the plastic tire and multi-screen TV experiments, a 3 X 3 X 2 between-subjects design was utilized, involving three levels of warranty quality (low, moderate, and high), three price levels (price absent, low price, high price), and two warrantor reputation levels (low or high). In the plastic tire experiment, the random assignment of 198 students to 18 treatment combinations resulted in 11 students per cell. There were 12 students per cell in the multi-screen TV experiment from randomly assigning 216 students to the 18 treatment combinations.

Manipulations

Plastic Tires.  The "low" warranty quality treatment labeled the warranty as "limited"; indicated that in case of malfunction the amount of refund and additional charges for the replacement tire would be prorated based on extent of tire wear; and specified the length of warranty coverage as up to 5000 miles or for a period of 6 months, whichever occurred first. The "moderate" warranty quality level was identical to the low level with exception of coverage length, which in this case was specified to cover malfunctions up to 20,000 miles or for a period of 2 years, whichever occurred first. The "high" level was identified as a "full" warranty, providing either a full refund or a free replacement tire (buyer's choice) in case of malfunction, and covering a period of 5 years or 50,000 miles, whichever lasted longer.

The three levels of price were $40 (low), $80 (high), and a "price absent" condition. This latter condition was performed following Jacoby, Olson, and Haddock's (1971) lead and based on Olson's (1977) suggestion that present-absent manipulations of the price cue are of greater theoretical interest than demonstrating effects due to specific price levels.

The manipulation of warrantor's reputation consisted of low and high levels. The "high reputability" level characterized the plastic tire innovator as an industry leader in rubber tire sales, as having vast experience in manufacturing tires, as marketing its tires worldwide and so forth. The "low reputability" level described the innovator as having no prior experience in tire manufacturing and as having previously been a manufacturer of plastic products such as seat covers and floor mats used by the automobile industry. This manipulation was intended to stimulate a perception such as: "How can a manufacturer of seat covers and floor mats produce safe, reliable, and dependable tires?" The names of actual manufacturers were not used to avoid any confounding that may have resulted from a company's name evoking cognitive responses irrelevant to the experimental manipulations.

Multi-Screen TVs.  The same general types of manipulations were used in the television experiment as for plastic tires. The low, moderate, and high warranty levels involved periods of 6 months, 2 years, and 5 years, respectively, with varying degrees of coverage associated with each level. The three price levels of the multi-screen television (specified as a 25-inch color set) were $450 (low), $900 (high), and no price information given. The "high" reputability manipulation described the company as a manufacturer of televisions since the 1940s; as a world-wide marketer and leader in the sale of televisions and other electronic products; and as having the most sophisticated engineering laboratories in the electronic products industry. The hypothetical company representing the "low" reputability level was characterized as having been a manufacturer of electronic products such as radio tubes until its engineers "hit upon" the idea of a multi-screen television.

Criterion Variable

Confidence, while not having received uniform usage in prior research (e.g., Howard and Sheth 1969; Howard 1974, 1977), has been used generally in the literature as a construct to reflect consumer's perceived degree of certainty or expected satisfaction regarding brand purchases.

As related to the present research, confidence reflects the consumer's perception concerning the product's performance capability, its quality, and its ability to satisfy product-relevant needs. An eleven-item scale was used to operationalize this criterion variable. Nine of the items involved measures of perceived quality, satisfaction, and performance risk. The remaining two items measured respondents' perception of the degree of financial risk associated with purchasing the experimental product. A number of recent research undertakings have documented the influence of consumer concerns about performance and financial uncertainties on product acceptance, and provide sufficient rationale for including risk dimensions as part of the overall confidence construct (e.g., Bettman 1973, 1975; Jacoby and Kaplan 1972; Peter and Ryan 1976; Peter and Tarpey 1975; Roselius 1971; Ross 1975; Taylor 1974).

Covariates

To enhance the precision of the experiments, four variables that are theoretically related to the criterion variable were treated as covariates:

Product Familiarity.  Perceived risk in buying a product is greater for those with little product experience than it is for those with extensive experience (Bettman 1973; Schaninger 1976; Taylor 1974). In a study of direct relevance to the present effort, Raju (1977) found that students' familiarity with stereos was significantly related to confidence in their abilities to choose the "right" brand. A three-item scale was used in the present research to operationalize product familiarity.

Specific Self-confidence.  Cox and Bauer (1964) and Bell (1967) found that product specific self-confidence influences consumers' persuasibility. Others have suggested that self-confidence is related to risk-taking (e.g., Taylor 1974). In similar manner, it is expected that one's specific self-confidence would influence innovativeness tendency (Midgley and Dowling 1978) and one's perceived degree of certainty in evaluating product quality and satisfaction from using the product, all of which are related to the confidence construct used in this research. A three-item scale operationalized this covariate.

Personal Competence.  The notion of personal competence involves a person's feeling of control and mastery of self and environment (Campbell, Converse, Miller, and Stokes 1960). It is related theoretically to ego-strength and internal control. Individuals who are high in personal competence would be expected to rely on their own judgment, and thus use extrinsic cues to a lesser extent than individuals low in personal competence. Campbell et al.'s (1960) eight-item scale provided the operational measure.

Sex.  This covariate was included because men have traditionally been socialized to show more interest in technical products of the type employed in this research.

Procedures and Instrument

Respondents received a test booklet containing one of the 18 product descriptions, and were given approximately five minutes to study the description. Detailed instructions and measurement scales followed. Individual items for the measured variables were dispersed throughout the questionnaire to reduce carryover effects. The bipolar adjective response scales were randomly reversed in direction to reduce acquiescence bias.

RESULTS

Manipulation Checks

Subjective-state questions were included in the test instrument to assess respondents' reactions to the warranty quality and warrantor reputation treatments to which they were exposed. [An unfortunate oversight resulted in failure to include similar questions for the price manipulation.] The experimental results are meaningful, of course, only if the manipulations influenced respondents in the intended manner. A three-item scale and a four-item scale provided the operational manipulation checks for warranty quality and warrantor reputation, respectively.

Statistical results indicated clearly that the warranty quality and warrantor reputation manipulations worked as intended in both experiments. In the plastic tire experiment, the warranty quality means (and standard deviations) for the high, moderate, and low treatment levels were 22.95 (3.66), 15.65 (6.13), and 12.77 (5.85). The F-value was 64.05 (p<.001, 2/197 df), and all paired-comparisons were significantly different. The warranty quality means (and standard deviations) in the multiscreen TV experiment for the high, moderate, and low levels were 23.13 (3.48), 13.28 (4.22), and 10.56 (5.07), yielding an F-value of 169.86 (p<.001, 2/215 df), with all paired-comparisons significantly different in the expected direction.

The warrantor reputation means (and standard deviations) in the plastic tire experiment for the high and low reputation levels, respectively, were 30.36 (5.22) and 20.54 (6.02). These were highly significantly different (t = 12.27, p<.001, 196 df). For the television experiment, the means for the high and low reputability groups were 29.11 (5.15) and 18.80 (5.34), and were also significantly different (t = 14.46, p<.001, 214 df).

Reliability and Validity Data

Reliability data for the measurement scales are presented in Table 1. Included is the criterion variable, confidence, along with measurement scales and reliability data for the covariates, manipulation checks, and two additional scales (intention and information adequacy) which were used for validation purposes. Test-retest and internal consistency (Cronbach's alpha) reliability tests were performed on all scales. The results in Table 1 reflect high stability and internal consistency across all scales, with minor exception of the information adequacy scale.

TABLE 1

RELIABILITY DATA FOR MEASUREMENT SCALES

The validity of the criterion variable, confidence, was assessed by means of a theory-relevant or nomological validation procedure. Based on Howard and Sheth's (1969) theory, confidence performs a central equilibrating role, and is related in predictable fashion to both buying intentions and to information needs. The greater the consumer's confidence, the more positive the intention to purchase and the less the perceived need for additional information.

The confidence-intention correlations were .71 and .25, respectively, for the plastic tire and multi-screen TV experiments, while the confidence-information adequacy correlations for the plastic tire and TV experiments were .59 and .40, respectively. Ail correlations were significant at p<.001. Although Bennett and Harrell (1975) have claimed that the confidence-intention relationship is basically tautological, these results do offer some support for the validity of the confidence construct employed in the present experiments.

Plastic Tire Experiment Results

Results of a three-way analysis of covariance are presented in Table 2, part A. [Tables of means and standard deviations for both experiments are not presented due to space limitations, but are available upon request.] The covariates were included to enhance the precision of the analysis and collectively accounted for approximately 16 percent of the explained variance. Personal competence was the only statistically significant covariate.

A strong warrantor reputation (R) main effect was obtained (F = 7.53, df = 1/176, p = .007). The mean for the high warrantor reputability level was 55.79, while that for the low level was 49.99. A significant main effect for warranty quality (W) also resulted (F = 3.74, df = 2/174, p = .025). The means for the high, moderate, and low levels were 57.74, 50.80, and 50.12, respectively. Paired-comparisons indicated that the mean confidence ratings for the moderate and low levels were not significantly different, but that the mean rating for the high level was significantly greater than both. The main effect for the price factor did not achieve a conventionally accepted level of significance.

The reputation X warranty quality (R X W) and price X warranty quality (P X W) interactions were the effects of primary relevance to this experiment. Figure 1 demonstrates the hypothesized R X W interaction. Though not shown in Figure 1, a similar P X W interaction was expected also. Table 2 reveals that the anticipated interactions failed to achieve statistical significance. Because of their a priori significance, however, and because both interactions approach significance, they are presented in Figure 2. Comparison of the R X W interaction in Figure 2 with the interaction hypothesized in Figure 1 reveals that the experimental data did not fit the anticipated pattern. The price X warranty quality interaction (P X W), shown also in Figure 2, did not exactly fit the anticipated pattern, but a tendency in the right direction is shown: the hypothesized leveling-off partially materialized for the price interaction while it did not for the R X W interaction.

FIGURE 2

TWO-WAY INTERACTIONS FOR PLASTIC TIRE EXPERIMENT: R X W AND P X W

In summary, the plastic tire experiment results revealed significant main effects due to warrantor reputation (R) and warranty quality (W), but neither the R X W nor the P X W interaction effects exhibited the exact patterns hypothesized.

Multi-Screen TV Experiment Results

Analysis of covariance results are presented in Table 2, part B. None of the covariates was significant. As found in the plastic tire experiment, significant main effects were obtained for both warrantor reputation (R) and warranty quality (W). The mean confidence ratings for the high and low reputability treatments were 55.82 and 50.20, respectively, yielding an F = 9.41 (df = 1/194, p = .003). The warranty quality treatment means were 56.72, 51.38, and 50.93 for the high, moderate, and low treatments, respectively (F = 3.74, df = 2/194, p = .025). Results from the paired-comparison tests indicated that the mean confidence ratings for the high warranty group were significantly greater than those for the other two groups, which, in turn, were not significantly different from each other.

None of the interaction effects approached significance. However, to provide a visual comparison of the results from the two experiments, the reputation X warranty quality and price X warranty quality interactions are displayed in Figure 3.

TABLE 2

ANALYSIS OF COVARIANCE RESULTS

FIGURE 3

TWO-WAY INTERACTIONS FOR MULTI-SCREEN TV EXPERIMENT: R X W AND P X W

DISCUSSION

The results emerging from these experiments provide additional insight into the cue utilization process. Support is provided for Olson's (1972) contention that extrinsic cues are of significant value to consumers in situations where intrinsic cues have low predictive and confidence values. The experimental procedures were designed to simulate such a situation by using rather novel and presumably risky products and presenting them in a concept test format.

The significant main effect obtained in both experiments for the warranty quality manipulation supports the hypothesis that warranties perform an important role by enhancing consumers' confidence in the performance and satisfaction expected from a product. The important role that a company's reputation plays in enhancing consumers confidence is demonstrated also by these findings. In both experiments, the most significant effect on confidence scores was due to the reputation of the hypothetical companies that were supposedly testing these product concepts. The absence of a significant main effect for the price treatment provides support for other research which has found that price is not the preeminent cue when other cues are available to consumers (e.g., Jacoby, Olson, and Haddock 1971).

Associated with these studies are also several problems and unanswered questions. The inability to account for some of the findings is a major concern. For example, no explanation is available to explain why in the plastic tire experiment the confidence scores were lower for the high reputation-low warranty quality combination. The fact that the hypothesized warrantor reputation X warranty quality and price X warranty quality interactions were not manifest, as hypothesized, represents another unanswered question. Is it that these interactions do not exist in reality, or is it that the experiments did not involve respondents sufficiently to allow the interactions to materialize?

Another potential limitation of these experiments stems from the use of specific levels of warranty quality and warrantor reputation rather than using more straightforward, present or absent manipulations. While these experimental results indicate that confidence scores do indeed depend on the specific levels of warranty quality and warrantor reputation, future studies should address the more basic issue of whether the mere presence of a warranty or reputation information has a significant impact on consumers' confidence. Clearly, additional research is needed to obtain a more complete understanding of how warranties are perceived and used by consumers.

REFERENCES

Armstrong, Gary M., Kendall, C. L., and Russ, Frederick A. (1975), "Applications of Consumer Information Processing Research to Public Policy Issues," Communications Research, 2, 232-245,

Bauer, Raymond A. (1960),"Consumer Behavior as Risk Taking,'' in Dynamic Marketing for a Changing World, ed., Robert S. Hancock, Chicago: American Marketing Association.

Bell, Gerald D. (1967), "Self-Confidence and Persuasion in Car-Buying," Journal of Marketing Research, 4, 46-52.

Bennett, Peter D., and Harrell, Gilbert D. (1975), "The Role of Confidence in Understanding and Predicting Buyers' Attitudes and Purchase Intentions," Journal of Consumer Research, 2, 110-117.

Bettman, James R. (1973), "Perceived Risk and Its Components: A Model and Empirical Test," Journal of Marketing Research, 10, 184-190.

Bettman, James R. (1975), "Information Integration in Consumer Risk Perception: A Comparison of Two Models of Component Conceptualization," Journal of Applied Psychology, 60, 381-385.

Campbell, A., Converse, P. E., Miller, W. E., and D. E. Stokes (1960), The American Voter, New York: John Wiley & Sons, Inc.

Carlsmith, J. Merrill, Ellsworth, Phobe C., and Aronson, Elliot (1976), Methods of Research in Social Psychology, Reading, Mass: Addison-Wesley Publishing Company.

Cook, Thomas D., and Campbell, Donald T. (1976), "The Design and Conduct of Quasi-Experiments and True Experiments in Field Settings," in Handbook of Industrial and Organizational Psychology, ed., Marvin D. Dunnette, Chicago: Rand McNally College Publishing Company.

Cox, Donald F. (1962), "The Measurement of Information Value: A Study in Consumer Decision-Making," in Emerging Concepts in Marketing, ed., W. S. Decker, Chicago: American Marketing Association, 413-21.

Cox, Donald F., and Bauer, Raymond A. (1964), "Self-Confidence and Persuasibility in Women," Public Opinion Quarterly, 28, 453-66.

Craig, C. Samuel, and McCann, John C. (1978), "Assessing Communication Effects on Energy Conservation," Journal of Consumer Research, 5, 82-9.

Darden, William R., and Rao, C. P. (1977), "Satisfaction with Repairs Under Warranty and Perceived Importance of Warranties for Applications," in Consumer Satisfaction, Dissatisfaction, and Complaining Behavior, ed., Ralph L. Day, Urbana: Division of Research, University of Indiana, 167-170.

Engel, James F., Kollat, David T., and Blackwell, Roger D. (1973), Consumer Behavior, 2nd Edition, New York: Holt, Rinehart, and Winston, Inc.

Feldman, Lawrence P. (1976), "New Legislation and the Prospects for Real Warranty Reform," Journal of Marketing, 40, 41-47.

Howard, John A. (1974), "Confidence As a Validated Construct,'' in Models of Buyer Behavior: Conceptual, Quantitative, and Empirical, ed., Jagdish N. Sheth, New York: Harper and Row Publishers, 160-168.

Howard, John A. (1977), Consumer Behavior: Application and Theory, New York: McGraw-Hill Book Company.

Howard, John A. and Sheth, Jagdish N. (1969), The Theory of Buyer Behavior, New York: John Wiley and Sons, Inc.

Jacoby, Jacob, Olson, Jerry C., and Haddock, Rafael A. (1971), "Price, Brand Name, and Product Composition as Determinants of Perceived Quality," Journal of Applied Psychology, 55, 570-579.

Jacoby, Jacob and Kaplan, Leon B. (1972), "The Components of Perceived Risk," in Proceedings, Association for Consumer Research, Third Annual Conference, 382-393.

Jacoby, Jacob and Kaplan, Leon B., and Szybillo, George J. (1974), "Components of Perceived Risk in Product Purchase: A Cross-Validation," Journal of Applied Psychology, 49, 287-291.

Kendall, C. L., and Russ, Frederick A. (1975), "Warranty and Complaint Policies: An Opportunity for Marketing Management," Journal of Marketing, 39, 36-43.

Lehmann, Donald R., and Ostlund, Lyman E. (1972), "Consumer Perceptions of Product Warranties: An Exploratory Study," in Proceedings, 3rd Annual Conference of Association for Consumer Research, ed., M. Venkatesan, Chicago: Association for Consumer Research, 51-65.

Levitt, Theodore (1967), "Communications and Industrial Selling," Journal of Marketing, 31, 15-21.

McClure, Peter J., and Ryans, John K., Jr. (1968), "The Difference Between Retailers and Consumers' Perceptions," Journal of Marketing Research, 5, 55-60.

Midgley, David F., and Dowling, Grahame R. (1978), "Innovativeness: The Concept and Its Measurement," Journal of Consumer Research, 4, 229-42.

Monroe, Kent B. (1973), "Buyers' Subjective Perceptions of Price," Journal of Marketing Research, 10, 70-80.

Montgomery, David B. (1975), "New Product Distribution: An Analysis of Supermarket Buyer Decisions," Journal of Marketing Research, 12, 255-64.

Olson, Jerry C. (1972), "Cue Utilization in the Quality Perception Process," Proceedings, Association for Consumer Research, Third Annual Conference, 167-79.

Olson, Jerry C. (1977), "Price As an Informational Cue: Effects on Product Evaluations," in Consumer and Industrial Buying Behavior, eds., Arch G. Woodside, Jagdish N. Sheth, and Peter D. Bennett, New York: American Elsevier, 267-286.

Perry, Michael, and Perry, Amon (1976), "Service Contract Compared to Warranty as a Means to Reduce Consumers' Risk," Journal of Retailing, 52, 33-40.

Peter, Paul J., and Tarpey, Lawrence X. (1975), "A Comparative Analysis of Three Decision Strategies," Journal of Consumer Research, 2, 29-37.

Peter, Paul J. and Michael J. Ryan (1976), "An Investigation of Perceived Risk at the Brand Level," Journal of Marketing Research, 13, 184-88.

Raju, P.S. (1977), "Product Familiarity, Brand Name, and Price Influences on Product Evaluation," Proceedings, Association for Consumer Research, Seventh Annual Conference, 64-71.

Roselius, Ted (1971), "Consumer Rankings of Risk Reduction Methods," Journal of Marketing, 35, 56-61.

Ross, Ivan (1975), "Perceived Risk and Consumer Behavior: A Critical Review," in Advances in Consumer Research, Vol. 2, ed., M. J. Schlinger, Chicago: Association for Consumer Research, 1-19.

Schaninger, Charles M. (1976), "Perceived Risk and Personality,'' Journal of Consumer Research, 3 (September), 95-100.

Simon, Marjii F. (1970), "Influence of Brand Names on Attitudes," Journal of Advertising Research, 10, 28-30.

Sternthal, Brian, Dholokia, Ruby, and Leavitt, Clark (1978), "The Persuasive Effect of Source Credibility: Tests of Cognitive Response," Journal of Consumer Research, 4, 252-260.

Taylor, James R. (1974), "The Role of Risk in Consumer Behavior," Journal of Marketing, 38, 54-60.

Udell, J. G. and Anderson, E. E. (1968), "The Product Warranty as an Element of Competitive Strategy," Journal of Marketing, 32, 1-8.

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