PAJAK RINGKAS

The effects of price-comparison advertising on buyers' perceptions of acquisition value, transaction value, and behavioral intentions

2012-01-11T08:28:00.000+07:00

Grewal, Dhruv; Monroe, Kent B; Krishnan, R. Journal of Marketing 62. 2 (Apr 1998): 46-59.

A study expands and integrates prior price-perceived value models within the context of price comparison advertising. More specifically, the conceptual model explicates the effects of advertised selling and reference prices on buyer's internal reference prices, perceptions of quality, acquisition value, transaction value, and purchase and search intentions. Two experiments are used to test the hypothesis. Results indicate that the experiments support the hypothesis that buyer's internal reference prices are influenced by both advertised selling and reference prices as well as the buyer's perception of the product's quality. In addition, the effect of advertised selling price on buyers' acquisition value was mediated by their perceptions on transaction value, and the effects of perceived transaction value on buyers' behavioral interns were mediated by their acquisition value perceptions.
Headnote
The authors expand and integrate prior price-perceived value models within the context of price comparison advertising. More specifically, the conceptual model explicates the effects of advertised selling and reference prices on buyers' internal reference prices, perceptions ofquality, acquisition value, transaction value, and purchase and search intentions. Two experimental studies test the conceptual model. The results across these two studies, both individually and combined, support the hypothesis that buyers' internal reference price sare influenced by both advertised selling and reference prices as well as the buyers' perception of the product's quality. The authors also find that the effect of advertised selling price on buyers' acquisition value was mediated by their perceptions oftransaction value. In addition, the effects of perceived transaction value on buyers' behavioral intentions were mediated by their acquisition value perceptions. The authors suggest directions for further research and implications for managers.
To compete successfully in a value-conscious environment, sellers must stress the value oftheir offerings. One value-based strategy involves emphasizing the value of acquiring the product (i.e., acquisition value) (Monroe and Chapman 1987). Sellers can increase acquisition value perceptions by enhancing buyers' perceptions of the product's quality or benefits relative to the selling price (Bolton and Drew 1991; Dodds, Monroe, and Grewal 1991; Monroe and Krishnan 1985; Zeithaml 1988). Thus, firms might opt for one of three value-based positioning strategies-high quality, low price, or some balance of quality to price.
Sellers also can compare a lower selling price to a higher advertised reference price (e.g., was $200, now $150) to enhance buyers' value perceptions. This value-oriented strategy is aimed at enhancing buyers' deal perceptions (or transaction value). Stressing the price bargain the buyer would be getting by undertaking the transaction can effectively promote the offering (i.e., increasing the salience of the reduction in the selling price). Unfortunately, previous research has not examined the effects of price-comparison or reference-price advertising on buyers' perceptions of acquisition value, transaction value, or behavioral intentions. There is need for researchers to understand why this price tactic (i.e., price-comparison advertising) seemingly works. From such information, effective value-oriented promotions could be developed. Furthermore, price promotions have become so widespread that several State's Attorneys' General offices (e.g., Maryland, New York, Colorado) have become concerned that some sellers use "value pricing" to deceive buyers (Grewal and Compeau 1992; Kaufmann, Smith, and Ortemeyer 1994).
In this article, we provide an understanding ofhow price-comparison advertising could influence buyers' perceptions of value and establish a framework for addressing the deception issue. The conceptual argument suggests that advertised reference prices in these deal-oriented advertisements can enhance buyers' internal reference prices (Lichtenstein and Bearden 1989; Urbany, Bearden, and Weilbaker 1988). These enhanced internal reference prices, when compared with the lower selling price, result in higher transaction value perceptions. The increase in perceived transaction value enhances purchases and reduces search behavior for lower prices. If sellers intentionally increase the advertised reference prices above normal retail prices, that is, inflate advertised reference prices, the resulting inflated perceptions of transaction value would be deceptive. Harm to both buyers and competitors could result from the effect of the inflated transaction value on buyers' search and purchase behaviors.
In addition, past research has not addressed the conceptual distinction between acquisition and transaction value, nor whether it is empirically feasible to make this distinction. The research reported in this article addresses this important limitation. We develop a conceptual model that outlines how advertised pricing tactics (i.e., comparison price-advertising) affect buyers' purchase decisions. Then we present two tests of the conceptual model and discuss the results and their implications.
The Conceptual Model
In price-comparison advertising, a higher advertised comparison price(commonly termed advertised reference price) is compared with a lower advertised selling price. Buyers' judgments of these advertised prices depend not only on the pricesper se, but also on the contextual cues presented within the advertisement, situational influences surrounding buyers, and buyers' internal reference prices (Rajenderan and Tellis 1994). The proposed model has two exogenous constructs (advertised reference price and advertised selling price) and six endogenous constructs (buyers' perceptions of product quality, their internal reference price, perceived transaction value, perceived acquisition value, willingness to buy, and search intentions) (see Figure 1). Each of these constructs and the relationships between them are explained subsequently.
Perceived Quality
Several past studies have examined the effects of information cues, such as price, on buyers' perceptions of quality (see reviews by Monroe and Krishnan 1985; Rao and Monroe 1989; Zeithaml 1988). Perceived quality is defined as a buyer's estimate ofa product's cumulative excellence (Zeithaml 1988). The general consensus of these studies is that price is less likely to have a significant effect on buyers' perceptions ofquality in the presence of other attributes and when buyers are familiar with the product or product category (Rao and Monroe 1988, 1989). In addition, there is evidence in the domain of comparative price advertising that these advertised prices (both the reference price and the selling price) do not have an effect on buyers' perceptions ofquality (see Grewal 1989; Urbany and Bearden 1990).
Consequently, subjects exposed to a comparative price offer for a well-known brand (Dodds, Monroe, and Grewal 1991), in the presence of several cues (including a picture of the product) (Grewal 1989; Rao and Monroe 1989), and having familiarity and knowledge of the product category (Rao and Monroe 1988) are not likely to use these advertised prices to shift their perceptions of quality. Therefore, in the proposed model, we do not expect the advertised selling price and the advertised reference price to affect buyers' perceptions of quality. However, we test for these paths.
Internal Reference Price
The concept of internal reference price, while operationally elusive, is an important cornerstone for behavioral price research. In this research, an internal reference price is defined as a price (or price scale) in buyers' memories that serves as a basis for judging or comparing actual prices (Monroe 1973; Monroe, Grewal, and Compeau 1991).
Della Bitta, Monroe, and McGinnis (1981) use adaptation-level theory to argue that buyers' internal reference prices are influenced by the key focal cues in an advertisement: the advertised selling price and the advertised reference price. Furthermore, adaptation-level theory suggests that these internal reference prices are influenced by residual cues (e.g., previously acquired information that has been assimilated to form perceptions and/or expectations of the quality of products in a product category or a specific brand). Buyers forming an initial level of perceived quality for the product and/or brand depend on information in the advertisement and on previously acquired information (Herr 1989). Using this level of perceived quality for the product and/or brand and the advertised prices (sales and reference) as a basis, buyers develop internal reference prices (or pricescales) to be used during subsequent judgments of value. The possibility of a perceived quality to the price mapping phenomenon has been illustrated by Monroe (1973). Consequently, advertised reference price, advertised selling price, and perceived product quality positively influence buyers' internal reference prices (Lichtenstein and Bearden 1989; Urbany, Bearden, and Weilbaker 1988). This conceptualization leads to the following three paths in the conceptual model:
HI: There is a positive relationship between buyers' perceptions of quality and their internal reference price.
H2: There is a positive relationship between advertised selling price and buyers' internal reference price.
H3: There is a positive relationship between advertised reference price and buyers' internal reference price. Buyers' internal reference prices adapt to the stimuli prices presented in the advertisement. That is, buyers either adjust their internal reference price or accept the advertised reference price to make judgments about the product's value and the value of the deal. Our conceptual model on how comparison price advertising influences buyers' perceptions of value explicitly recognizes this adaptive nature of buyers' internal reference prices.
Perceived Acquisition Value
Past acquisition value-based models (e.g., Dodds, Monroe, and Grewal 1991; Zeithaml 1988) have defined this concept as the perceived net gains associated with the products or services acquired. That is, the perceived acquisition value of the product will be positively influenced by the benefits buyers believe they are getting by acquiring and using the product and negatively influenced by the money given up to acquire the product (i.e., the selling price). One important element of this "get" component is product quality or buyers' perceptions of product quality. 1
Several researchers have conceptualized acquisition value in this manner, though they used different terms such as "bargain value" (Keon 1980), "perceived value" (Dodds, Monroe, and Grewal 1991; Lichtenstein and Bearden 1989; Monroe and Krishnan 1985; Urbany, Bearden, and Weilbaker 1988), "perceived worth" (Szybillo and Jacoby 1974), "acquisitionutility" (Thaler 1985), and "value consciousness" (Lichtenstein, Netemeyer, and Burton 1990; Lichtenstein, Ridgeway, and Netemeyer 1993). Therefore, defining perceived acquisition valueas the buyers' net gain (or tradeoff) from acquiring the product or service represents "a more global and enduring kind of value which takes into account both price and quality" (Urbany and Bearden 1990, p. 4). This conceptualization leads to the following paths in the conceptual model:
H4: There is a positive relationship between buyers' perceptions of quality and their perceived acquisition value. H5: There is a negative relationship between the advertised selling price and buyers' perceptions of acquisition value.
Perceived Transaction Value
Buyer sexposed to price-comparison advertisements and similar price promotions are presented with an expressed deal or bargain in terms of a selling price that is explicitly reduced in magnitude. They are likely to assess the merits or value of such a deal by comparing the selling price to their internal reference prices (Monroe and Chapman 1987; Thaler 1985). For example, in a recent interpretive study of buyers' reactions to price-deals, one shopper indicated the following: "Sometimes if I get a good deal at the discount rack, I feel good about that and I'll stroll through the other parts [of the mall or store] and not feel guilty if I buy more expensive, originally priced items" (Grewal and Compeau 993, p. 11).
Therefore, a buyer, on examining the financial terms of the price offer, might perceive additional value beyond that provided by acquisition value. Thus, perceived transaction value is the perception of psychological satisfaction or pleasure obtained from taking advantage of the financial terms of the price deal (Lichtenstein, Netemeyer, and Burton 1990; Monroe and Chapman 1987; Thaler 1985; Urbany and Bearden 1989). This conceptualization leads to the following paths in the conceptual model:
H6: There is a positive relationship between buyers' internal reference price and their perceived transaction value.
H7: There is a negative relationship between the advertised selling priceand buyers' perceptions of transaction value.
Effects of Acquisition and Transaction Value on Willingness to Buy and Search Intentions
Willingness to buy is defined as the likelihood that the buyer intends to purchase the product (Dodds, Monroe, and Grewal 1991). All things being equal, willingness to buy is positively related to overall perceptions of acquisition and transaction value (Della Bitta, Monroe, and McGinnis 1981; Monroe and Chapman 1987; Urbany and Dickson 1990; Zeithaml 1988). Buyers' willingness to buy is positively linked to their perceptions of acquisition and transaction value:
H8: There is a positive relationship between buyers' perceptions of acquisition value and their willingness to buy. H9: There is a positive relationship between buyers' perceptions of transaction value and their willingness to buy. Search intention is defined as a buyer's willingness to search for additional price information. Stigler (1961) suggests that because of variations in price in the marketplace buyers generally are uncertain what the lowest available price is. To reduce this uncertainty, buyers must seek information from sellers. Willingness to search for price information is contingent on buyers' trading off the perceived benefits (e.g., money saved) relative to the costs of the search (e.g., time, money, effort spent in conducting the search) (Marmorstein, Grewal, and Fishe 1992). Previous research shows that when buyers are exposed to an advertised regular price coupled with a lower sale price, their willingness to conduct additional search declines because of an increase in their perceptions of value (Della Bitta, Monroe, and McGinnis 1981; Urbany, Bearden, and Weilbaker 1988). Therefore, buyers' intentions to search is linked negatively to their perceptions of acquisition and transaction value:
H10: There is a negative relationship between buyers' perceptions of acquisition value and their intentions to search.
H11: There is a negative relationship between buyers' perceptions of transaction value and their intentions to search.
An Alternative Model
A key issue that must be addressed is whether perceived transaction value and perceived acquisition value are interrelated. Similar to Thaler's (1985) conceptualization and Monroe and Chapman's (1987) model, the proposed model (Figure 1) assumes that buyers' perceptions of acquisition and transaction value are independent of each other. However, there are several potential reasons that suggest these two value dimensions are not independent of each other. Conceptually and operationally, the interrelationship between these two value dimensions has not been addressed by previous research (cf. Urbany and Bearden 1990).
It is reasonable to propose that a price promotion that leads to positive perceived transaction value (i.e., greater psychological pleasure associated with obtaining favorable financial terms) would in turn influence buyers' perceptions of the value of acquiring the product or receiving the service (i.e., greater net gain by reducing the financial outlay). It is proposed that positive perceived transaction value enhances buyers' evaluations of the value of acquiring the product. Buyers' perceptions of transaction value are situation specific, and though their assessments of acquisition value are more holistic evaluations of the product's value, it is likely that their transaction value influences their perceptions of acquisition value and not vice versa.
The two constructs have an overlapping antecedent construct, the advertised selling price. Drawing on equity notions from satisfaction research (see Bolton and Drew 1991; Oliver and Swan 1989), transaction value could be considered akin to the fairness construct (i.e., the equity/pleasure associated with getting a fair price) (see also Huppertz, Arenson, and Evans 1978), and acquisition value similar to the overall satisfaction construct (Grewal 1995). Research in this post-purchase domain suggests that buyers' assessment of equity affects their overall evaluations. The parallel in thepre-purchase domain is that perceived transaction value affects perceived acquisition value.2 This conceptualization leads to the following additional path: H12: There is a positive relationship between buyers' perceptions oftransaction value and their perceived acquisition value.
Hl2 suggests that sellers also might influence buyers' evaluations of the value of the product (i.e., acquisition value) indirectly through the effects of comparative advertised reference prices on buyers' perceived transaction value. This suggestion also leads to the possibility that the effect of advertised selling price on buyers' acquisition value might be mediated by their perceptions of transaction value. It also leads to the possibility that the effect of perceived transaction value on behavioral intentions could be mediated by perceived acquisition value.
On the basis of these proposed revisions, an alternative model (Figure 2) also is tested. The alternative model hypothesizes a link between perceived transaction value and perceived acquisition value. The effects ofadvertised selling price on acquisition valueare expected to be mediated by transaction value (i.e., no significant direct effect of advertised selling price on acquisition value). The revised model also hypothesizes that the effects of perceived transaction value on behavioral intentions are mediated by perceived acquisition value (i.e., no significant direct effects of perceived transaction value on purchase and search intentions). In addition, the no-effect paths pertaining to advertised selling price and advertised reference price on buyers' perceptions ofquality have been dropped.
Research Method Research Plan
These hypotheses (and models) were tested using causal modeling. Two studies were conducted. Both studies used a 2 x 2 between-subjects experimental design, that is, two selling price levels ($249.95 and $349.95) and two advertised reference price levels ($400 and $500).3 In both studies, the subjects were shown a booklet containing an advertisement for a bicycle and a questionnaire. The advertisement used a known brand name (Raleigh USA). The subjects then responded to a set of questions that assessed the various latent constructs.
In Study 1, a laboratory experiment, the subjects were 361 undergraduate students at a western state university. The mean age of respondents in Study I was 23 years (range 18-49), 52.9% of the respondents were male, and 76.1% owned a bicycle. In Study 2, an experimental survey assessing the replicability and the boundary conditions of Study l's results, the subjects were staff employees at the same university. Of the 600 employees surveyed, 328 responded, a 54.6% response rate. To motivate a response, $125 in prizes were awarded by lottery. The mean age of respondents in Study 2 was 41 years (range 24-62), median family income was $40,000-$50,000, 37.6% of the respondents were male, and 71.6% owned a bicycle.
Pretests
A series of pretests were conducted. The first provided information about the subjects' knowledge, involvement, and acceptable price range for bicycles. The results indicate that a bicycle was a personally relevant product and thus was selected as the test product. Because Rao and Monroe (1988) found that product knowledge affected buyers' subjective product evaluations, another objective ofthis pretest was to determine a product about which subjects were knowledgeable. The results indicate that subjects exhibited high levels of product familiarity with this specific product (i.e., a Raleigh USA bicycle). The results and a market survey of prices suggest that $400 was around the average market price and $500 was an above-average market price (i.e., an inflated advertised reference price) for this bicycle. Three additional pretests were used to develop the scales for the various latent constructs, particularly acquisition value and transaction value.4
Measures
The scales used to measure the latent constructs are provided in Table 1. The constructs of buyers' perceptions of quality, acquisition value, transaction value, willingness to buy, and search intention were assessed using seven-point category rating scales. Buyers' internal reference prices were dollar estimates provided by the subjects.
Perceived quality. Buyers' quality perceptions were measured using three Likert statements (Dodds, Monroe, and Grewal 1991; Rao and Monroe 1988) that assessed the product's quality, durability, and reliability.
Internal reference price. Buyers' internal reference price was assessed using two common measures: average market price estimate and fair price estimate. These two items were based onscales developed by Lichtenstein and Bearden ( 1989) and Urbany, Bearden, and Weilbaker (1988).
Perceived acquisition value. Buyers' acquisition value was measured using nine Likert statements that built on past scales of perceived value (e.g., Chapman and Monroe 1990; Dodds, Monroe, and Grewal 1991). However, theproposed measure of perceived acquisition value is more comprehensive than the three- or four- item scales previously used that focused on"good value for the money." We explicitly attempted to capture the trade-off between a product's benefits and the cost of its acquisition. For example, sample items included the following: "I feel that acquiring this bicycle meets both my high-quality and low-price requirements;" "I would value this bicycle as it would meet my needs for a reasonable price;" and "This bicycle would be a worthwhile acquisition because it would help me exercise at a reasonable price."
Perceived transaction value. Past research has had considerable problems measuring buyers' perceptions of transaction value and developing a scale that discriminates adequately from perceived acquisition value. On the basis of our pretests and research by Lichtenstein, Netemeyer, and Burton ( 1990), we measured perceived transaction value using three Likert statements. These statements seem to capture the essence of transaction value-the pleasure buyers get from finding and taking advantage of a price deal (e.g., "taking advantage ofa pricedeal like this makes me feel good"). Principal component analysis of the acquisition and transaction value scales demonstrates that the two scales discriminate in both studies (see Table 2).
Willingness to buy. A three-item scale, based onDodds, Monroe, and Grewal's (1991) study, measured buyers' willingness to buy. The specific items were anchored from "very low" to "very high."
Search intentions. Buyers' intentions to search for additional information (e.g., visit other stores to check their prices) were measured using three Likert statements that were based onprior research by Della Bitta, Monroe, and McGinnis (1981).
Analysis and Results
Plan for Data Analysis
The data from the two studies were analyzed in two stages. The measurement model was assessed to confirm that the scales were unidimensional and reliable. When the reliability of the measures had been established, the structural model was tested using LISREL-VII causal modeling procedures (Joreskog and Sorbom 1989). This testing determined the strength of individual relationships, the model's goodness of fit, and the various hypothesized paths. The two-step procedure followed here reduces the number of interpretational confounds. PRELIS was used to generate the input matrix.
Measurement Properties of the Scales
The scales used to measure the latent constructs in the model are provided in Table 1. Item reliability, variance extracted, and construct reliability also are shown. The assessment of the measurement properties of all six scales indicated that the factor loadings (lambdas) were high and significant (p < .001), which satisfies the criteria for convergent validity. Anderson (1987) suggests the following criterion for assessing discriminant validity between scales: The correlation between two latent constructs plus or minus two standard errors does not include one. All six scales met this criterion in both studies. Furthermore, Fornell and Larcker (1981) suggest that discriminant validity can be assessed by determining whether the variance extracted estimates for two constructs are greater than the square of the parameter estimate between them . The six measured constructs met this criterion in both studies. We also assessed the discriminant validity of the scales using confirmatory factor analysis procedures (Anderson and Gerbing 1988). The results of each pairwise construct comparison suggest that the two factor solution was better than the single factor solution (see the Appendix).
Fornell and Larcker (1981) also stress the importance of examining composite reliability and variance extracted. Bagozzi and Yi (1988) suggest two criteria: Composite reliability should be greater than or equal to .60, and variance extracted should be greater than or equal to .50. For both studies, all six composite reliabilities were greater than .75, and all six variances extracted were greater than .55 (see Table l). Finally, we analyzed the structural model using summated scales and obtained similar results, which suggests that the measurement-structure interaction was minimal.
Model Fit
The causal models in Figures I and 2 were assessed using a full-information method. The causal models were specified as shown in the figures. The PHI, PSI, TD, and TE matrices were diagonal and free. The theta-deltas associated with advertised selling price and advertised reference price were fixed. Errors were treated independently to avoid interpretational confounds. The lambda matrices (both X and Y) were full and fixed. Then the individual items associated with the exogeneous and endogeneous constructs were freed. However, one of the lambdas for each construct was set to 1.0 to properly define the measurement (see Joreskog and Sorbom 1989).
The overall fit of the structural model was determined initially by examining the Z2 statistics for each study, which were significant. A significant 2 statistic could indicate an inadequate fit, but this statistic is sensitive to sample size and model complexity; therefore, rejection of a model on the basis of this evidence alone is inappropriate (Bagozzi and Yi 1988; Bearden, Sharma, and Teel 1982; Marsh, Balla, and McDonald 1988). Accordingly, other measures of fit compensating for sample size also were applied: Bentler and Bonett's (1980) normed fit index (A), Tucker and Lewis's (1973) non-normed fit index (p), and Bentler's (1990) comparative fit index (CFI). Each ofthese indices showed an adequate fit: A was .88 (Study 1) and .89 (Study 2); p was .91 (Study 1) and .90 (Study 2); CFI was .93 and .91 for Study I and 2, respectively (see Table 3).
Hypotheses Tests
The standardized estimates for the various model paths and the associated t-values for the two studies are provided in Table 3. The structural path estimates for the model furnished in Table 3 should be read with the caveat that our data are causal only for the effects of advertised reference price and advertised sale price. For all other relationships, our data are correlational, and the causal direction is based on prior theory.
Advertised pricesand perceived quality. As expected, advertised selling priceand advertised reference pricedid not affect buyers' perceptions of quality significantly.
Influences oninternal reference price. These results support H1, H2, and H3, which indicates that buyers' internal reference prices are functions of perceived quality, advertised selling price, and advertised reference price. Previously, Urbany, Bearden, and Weilbaker ( 1988) demonstrated the capability of advertised reference prices to serve as anchors and to shift internal reference prices in their direction. Finally, Hyun (1993), using Korean subjects, demonstrated a positive relationship between advertised selling prices and subjects' internal reference prices. Thus, our results complement past research and are consistent with predictions based on adaptation-level theory.
Influences on perceived acquisition value. In both studies, perceived acquisition value is a positive function of subjects' perceptions of quality (H4 is supported). This relationship has previous empirical support (Dodds, Monroe, and Grewal 1991; Hyun 1993). Therefore, we provide empirical support in a price comparison context for the proposition that perceived acquisition value is influenced, in part, by buyers' perceptions of quality. (Substantively, as mentioned previously, the literature stresses that perceived quality is an important part of the"value equation.")
The hypothesized influence of the advertised selling price on perceptions of acquisition value was supported in both studies (H5 is supported). The overall available evidence presented here supports the theoretical proposition and the lay belief that selling price is a negative element of buyers' perceptions of acquisition value.
Influences on perceived transaction value. The model in Figure 1 suggests that perceived transaction value is a function of buyers' internal reference prices and the actual selling price. As is shown in Table 3, these relationships were supported by both studies (H6 and H7 are supported). There is a significant negative relationship between the actual selling price and subjects' perceptions of transaction value. Moreover, there is a positive relationship between subjects' internal reference prices and their perceptions of transaction value. Thus, the theoretical arguments for the influence of selling price and buyers' internal reference prices on buyers' perceptions of transaction value (i.e., the perceived merits of the offer) have empirical support.
Value perceptions and behavioral intentions. Della Bitta, Monroe, and McGinnis ( 1981 ) observe that buyers' positive perceptions of value were a necessary but insufficient condition to induce willingness to buy. Other research has found a positive relationship between perceptions of value and willingness to buy (Dodds, Monroe, and Grewal 1991; Hyun 1993). However, previous research efforts measured the more global construct-perceived value. In the current empirical effort, we decompose perceived value into two theoretical components: perceived acquisition value and perceived transaction value.
The results shown in Table 3 indicate a significant positive relationship between perceived acquisition value and willingness to buy (HK is supported). The direct relationship between perceived transaction value and willingness to buy, though positive, is weak overall and not statistically significant in the second study (partial support for Hg). Similarly, the negative relationship between perceived acquisition value and search intentions is significant in both studies (Hio is supported). The relationship between perceived transaction value and search intentions was not supported (no support for HI,).
Test of the Alternative Model, H12, and Mediating
Hypotheses
Alternative model. A key theoretical argument of the alternative model presented in Figure 2 is that perceived transaction value has a positive influence onperceived acquisition value. The empirical relationships provided in Table 4 strongly support this relationship in both studies (H12 is supported). Previous efforts to decompose perceived value into its theoretical acquisition value and transaction value components have had measurement flaws. As is demonstrated in Tables 1 and 2, the research reported here has overcome the inherent measurement difficulties presented by these two concepts.
The results of the alternative model also are presented in Table 4. The model hypotheses are supported in both studies. Furthermore, the revised model (with four fewer paths) fits the data as well as the complete model (e.g., the CFI statistic was the same for both models). These results further validate our propositions that the effect of selling price on perceived acquisition value is mediated by perceived transaction value and that the effects of perceived transaction value on purchase and search intentions are mediated by
Discussion
Conceptual Developments
As has been noted, price comparison advertising is a widely used price promotion tactic. Although research investigating issues on the relative effectiveness of this tactic spans nearly 20 years, we are still trying to understand how and why it works. Drawing on prior research (e.g., Monroe and Chapman 1987; Thaler 1985), we provide a theoretical argument for why such price promotional tactics (and other similar forms) influence buyer behavior.
Previous empirical efforts to decompose the concept of overall perceived value into the two independent constructs of perceived acquisition value and perceived transaction value encountered measurement problems (e.g., Chapman and Monroe 1990; Grewal 1989). Examining prior research efforts (e.g., Chapman and Monroe 1990; Grewal 1989) and comments onThaler's original conceptualization (Bearden et al. 1992) led to the revised model in Figure 2 and to the stronger measurement model presented here.
Prior research (i.e., Monroe and Chapman 1987; Thaler 1985) typically modeled these two value dimensions as independent of each other. There is the possibility that one of the two value components is actually an antecedent of the other. Because, in the absence of a price promotion, the basic perceived value model (see Dodds, Monroe, and Grewal 1991; Zeithaml 1988) postulates that buyers' perceptions of value are formed from a mental trade-off between perceived quality (or benefits) and price, it seems logical that buyers would perceive a price promotion as enhancing the overall value of an acquisition. Thus, the model in Figure 2 shows the conceptual adjustments, which suggests that perceived acquisition value is a function of perceived quality and perceived transaction value (i.e., assessment of the price offer). The advertised selling price affects perceived acquisition value by its effect on perceived transaction value. As simple as this conceptualization seems, it represents an important addition to the research literature on the relationship between price and buyers' perceptions of quality and value.
Although this research emphasizes exploration of the relative effects of price comparison advertising on buyers' perceptions of value, important insights also have been found about the internal reference price concept. It has been conceptually argued and empirically confirmed that buyers' internal reference price is influenced by both the seller's advertised (higher) reference price and the advertised (lower) selling price. Previous research demonstrates the influence of the advertised reference price on buyers' internal reference prices (Urbany, Bearden, and Weilbaker 1988). As conceptualized by adaptation-level theory, this research also demonstrates that both prices presented in a price comparis on advertisement influence buyers' internal reference prices. Moreover, it has been shown that buyers' internal reference prices are influenced by their relative assessments of product quality. This finding is consistent with the tenets of cognitive reference points and categorization theory (Herr 1989; Monroe, Grewal, and Compeau 1991).
The conceptual argument by Thaler (1985) and Monroe and Chapman (1987) that perceived transaction value is a function of the selling price and buyers' internal reference price has been confirmed empirically. Although Urbany and Bearden (1989) show the positive relationship between buyers' internal reference price and their perceptions of transaction value, this is the first published research effort to demonstrate the apparent implied mental comparison between the advertised selling price and buyers' internal reference prices.
Also of concern is whether perceived transaction value has a direct or indirect influence on willingness to buy or intentions to search. Previous research offers convincing evidence on the direct relationship between perceived (acquisition) value and measures of behavioral intentions. Our empirical evidence supports the idea that perceived transaction value influences willingness to buy and intentions to search through its effect on perceived acquisition value. This particular finding strengthens the argument that, in comparative price advertising promotions, perceived transaction value enhances buyers' perceptions of acquisition value, and that these two components of perceived value are not independent constructs.
Measure Development
Operationally, this research develops and validates separate measures of perceived acquisition value and perceived transaction value (Table 1). One aspect of this measure development was to understand the psychological pleasure that buyers might experience when buying a product ondeal (i.e., obtaining a bargain). For further discussion on this point, see Schindler (1989). Another contribution of this research is the development of a unidimensional multi-item scale to measure buyers' internal reference price. Past research predominantly used single items that tapped into various aspects of the internal reference price scale (or continuum) (see Monroe, Grewal, and Compeau 1991). In addition, measures for perceived product quality and purchase and search intentions were further refined and validated (see Table 1).
Managerial Implications
A key managerial implication of this research is the demonstration of how advertised prices (both reference and sale) provided in price comparison advertisements affect buyers' internal reference prices (i.e., higher advertised prices lead to higher internal reference prices). In turn, these internal reference prices are linked to buyers' perceptions of value and behavioral intentions. Focusing ononly the final price (or sale price) to the exclusion of the contextual advertised (or display) reference price by advertisers and retailers might be a strategic mistake. This issue could be one of the reasons why the"everyday low price" strategy used by several retailers has not been successful.
The results of our studies, in conjunction with past research (e.g., Lichtenstein and Bearden 1989; Urbany, Bearden, and Weilbaker 1988), suggest that inflated advertised reference prices have the potential to be deceptive (in our studies, the $500 advertised reference price was considerably higher than the average market price). These higher advertised reference prices enhance buyers' perceptions of transaction value, acquisition value, and purchase intentions and reduce buyers likelihood of searching for a lower price. Furthermore, the potential for deception is likely to be more pronounced for buyers who have less price and/or product knowledge. The concerns voiced by several State's Attorneys' General offices that value pricing could be used to deceive consumers seem to have merit. Consequently, State's Attorneys' General offices and the Federal Trade Commission must monitor such practices. In cases in which retailers and manufacturers use fictitious or inflated advertised reference prices in their advertisements, appropriate action must be taken. Appropriate action by such agencies might include cease and desist orders, fines, and posted notices of how the retailer established the advertised reference price.
The current research findings support the notion that product quality perceptions enhance acquisition value and willingness to buy. Furthermore, past research shows that a high-quality position is important in developing brand equity and leads to higher market share and profitability in the long run (Curry 1985; Jacobson and Aaker 1987; Phillips, Chang, and Buzzell 1983). Our results (crosssectional study) in conjunction with those based on PIMS databases (i.e., longitudinal) suggest that developing and maintaining a high-quality position is important for shortterm adoption and long-term development of market share.
The lack of association between price and quality perceptions in our two studies supports past research findings that a high-quality position is not necessarily incompatible with a low cost (or price) position (Phillips, Chang, and Buzzell 1983). Many manufacturers try to maximize valu eto buyers by offering above-average quality at reasonable prices (Curry 1985). This positioning can be achieved through well-designed price promotions that emphasize the fairness or reasonableness of their selling prices and thereby enhance buyers' perceptions of transaction value. Our findings suggest that buyers' perceptions of transaction value enhances willingness to buy through their perceptions of acquisition value.
Our study results suggest that acquisition value has considerable influence on buyers' willingness to buy. It must be noted that customers balance the benefits of the purchase against the costs. Benefits can be functional, operational (e.g., durability, reliability), or personal (Shapiro and Jackson 1978). Costs include both financial (sale price) and nonfinancial aspects, such as time and effort (Zeithaml 1988). Today's information technology (e.g., through the Internet, consumer reports) enables buyers to compare benefits and prices with unprecedented ease and accuracy. Managers must understand the variables affecting the acquisition value of the product. They also must understand where their product fits on a continuum ranging from satisfying unique needs (e.g., CAT scanner) to satisfying undifferentiated needs (e.g., corn syrup) (Dolan 1995; Nagle and Holden 1995). Thus, manufacturers can position products that are unique using an acquisition value-enhancing strategy and those that are relatively undifferentiated from competitors using a transaction value-enhancing strategy.
The study results also suggest that the various value strategies (i.e., deal value versus product value) are important predictors of behaviors. In addition, past research suggests that the cost of serving buyers and the effectiveness of value strategies might vary across segments (Lichtenstein, Netemeyer, and Burton 1990; Shapiro et al. 1987). That is, some segments are sensitive toward price, whereas others are more benefit oriented (in our studies, the relative effect of acquisition value versus transaction value on willingness to buy was greater for the nonstudent sample). Therefore, value perceived by buyers will vary across segments. For some buyers, acquisition value might be more important than transaction value or vice versa. Managers should determine which value strategy is appropriate for their target segments and develop their positioning strategies appropriately.
Limitations and Avenues for Further Research
Further research should explore this study's limitations. For example, subjects were exposed to one type of semantic cue, the original price and selling price combination. Research using other product and semantic cues, such as "compare at, selling price," "MSLP, selling price," and "total value, selling price" would be worthwhile (see Grewal, Marmorstein, and Sharma 1996).
Another limitation of this research was Thaler's ( 1985) suggestion that buyers' acquisition utility (or value) should be equivalent to the comparison of their reservation price with the actual selling price. This research, following Monroe and Chapman's (1987) conceptualization, considered perceived acquisition value as a comparison between buyers' perceptions of quality and selling price. Although the approach used here is consistent with the extant marketing practitioner beliefs, the relative role of reservation price (or maximum acceptable price) should be explored as an alternative way to measure the get component of buyers' perceptions of acquisition value. In addition, research must explore whether other factors such as usage flexibility, usage convenience, and need congruence affect buyers' perceptions of acquisition value.
A related issue pertains to the functional form of perceived acquisition value. Monroe ( 1990) suggests a ratio (or proportional) model, but points out that this is only one means of illustrating the comparison. Associated research on price-quality and price-warranty trade-offs supports a subtractive model (e.g., Levin and Johnson 1984; White and Truly 1989). Our results also support a subtractive model.6 However, White and Truly (1989), using some important methodological variations, also show that some of their subjects apparently followed a proportional model of information integration. Therefore, an important research issue is whether the information integration implied by theformation of perceived acquisition value is represented best as a subtractive, ratio, or averaging model or as some other functional form. Finally, there is a need for research that involves examining whether the degree of believability of the advertised reference price, believability of the selling price, believeabilty of the overall price offer, and price consciousness influence buyers' perceptions of transaction value.
Another issue that warrants additional research addresses the distinction between acquisition value and overall value. Overall perceived value can be conceptualized with many distinct components (Forbes and Mehta 1978), which could include the value of the acquisition, the value associated with the start-up (e.g., a cellular phone with a recharger similar to an existing cellular phone will have greater value), and the value associated with reselling the product (e.g., some cars have greater resale value). Thus, research in business-to-business settings might need to distinguish, conceptually and operationally, acquisition value from overall value.
Research also must address how buyers form their internal reference prices. In this research, respondents' internal reference prices were operationalized through point estimates of their expected normal, average, and fair prices. However, Klein and Oglethorpe (1987) suggest that internal reference prices could be operationalized in a variety of other manners, including by expected prices, price last paid, or aspiration price. Further-more, internal reference prices also may be operationalized as a price range (Monroe, Grewal, and Compeau 1991; Urbany and Dickson 1990). Therefore, further research must address whether these different bases for, and ways of measuring, internal reference prices yield similar results.
Although this study examines the effectiveness of the framework in the context of price-comparison advertisements, further research is needed to test its effectiveness in explaining buyers' behavioral reactions to other price promotions (e.g., coupons, rebates) as well as price changes. The generalizability of the model should be examined by assessing the fit of the model for different samples of buyers. Value-conscious segments, deal-prone segments, and segments that do or do not believe the difference between the advertised selling price and the advertised reference price might vary. Furthermore, the predispositions of the subjects could affect their perceptions. It would be useful to look at the individual characteristics of subjects more closely and assess such variables as involvement, price consciousness, knowledge, and inclination to take risks.
Footnote
1 Thaler (1985) operationally defines this get component (or value equivalent) as the amount of money that would leave the person indifferent about receiving the money or the product as a gift. In economic theory, the value equivalent is similar to the reservation price (the maximum price the buyer is willing to pay), and therefore acquisition value could be considered comparable to consumer surplus (Monroe and Chapman 1987; Thaler 1985). However, operationally defining the value equivalent (get component) simply as a reservation price is a limited view, because it does not include the buyers' quality evaluation, except by indirect inference. Moreover, empirical evidence verifying that buyers use such reservation prices when assessing the value of a product is not available. For example, Bearden and collegues (1992) find no significant relationship between three price-estimate measures of the subjects' reservation price and their perceptions of acquisition value or their willingness to buy.
Footnote
2We thank an anonymous reviewer for this suggestion.
Footnote
3We do not predict an interaction term between advertised selling price and advertised reference price, in line with previous models, such as Monroe and Chapman's (1987). However, we did test for the interaction term. MANOVA analyses onboth data sets indicated that the interactions were not significant in either data set (p > .05). Furthermore, using LISREL procedures and modeling an interaction term, we did not find a significant effect of the interaction on internal reference price, acquisition value, and transaction value. Furthermore, the results with an interaction term suggest a worse fit.
Footnote
40ne of the pretests' (n = 400) results indicated that scales used by past research (e.g., Chapman and Monroe 1990) to assess overall perceived value did not adequately discriminate acquisition value from transaction value. The results suggest a two-factor solution. That is, acquisition value and transaction value load on separate factors, but overall value loads on both factors. Thus, this research explicitly focuses on acquisition and transaction value and not on overall value.
Footnote
5Three conditions must be met to establish mediation: ( I ) the independent variable affects the mediator: (2) the independent variable affects the dependent variable: and (3) when both the independent variable and the mediator are regressed on the dependent variable, the mediator is significant, whereas the effect of the independent variable is reduced. ANOVA and ANCOVA procedures suggested by Hastak and Olson (1989) (similar to Baron and Kenny [19861 procedures) were followed and supported the proposition that perceived transaction value mediates the effect of selling price onperceived acquisition value. ANOVA results indicated a significant effect of selling price on perceived transaction value (Study I: F([.344)= 12.IO, p < .OOI: Study 2: F(1,36)= 48.64, p < .()QI ) and perceived acquisition value (Study 1: F(I 344) = 7.57, p < OI; Study 2: F(1.316) = 25.87, lp < .OOI ). Furthermore, the effect of selling price on perceived acquisition value was nonsignificant when perceived transaction value was treated as a covariate (Study
Footnote
1: Fti 3)= 1.26, p > .05; Study ?: Fti,315)=.57, p > .OS). In addition, the covariate was significant (Study 1: Ft l 343) = 130.16, p < OOI: Study 2: FkI.315= 273.91, p < .OOI ). Procedures suggested by Baron and Kenny (1986) were followed to assess whether perceived acquisition value mediates the effect of perceived transaction value on buyers' willingness to buy. We find that perceived transaction value significantly enhanced perceived acquisition value (Study 1: tt3,rt;46) = 11.43, p < .OOI; Study 2: t(318)= 16.59, p < .OOI ) and willingness to buy (Study I: t(353= 7.25,jp < .001; Study 2; t(3i = 9.99, p < .001). Furthermore, when both transaction value and acquisition value were regressed on buyers' willingness to buy, acquisition value significantly affected willingness to buy (Study 1: t(343= 10.28, p < .OOI: Study 2: t(317)= 12.29, p < .OOI ), while the effect of transaction value was reduced and nonsignificant (Study 1: t(34= 1.41, p > .05; Study 2: t(317= 1.12; p > .OS). Thus, the regression results support the proposition that the effect of perceived transaction value on willingness to buy is mediated by perceived acquisition value. Similarly, we find that perceived
Footnote
transaction value significantly reduced search intentions only in Study 2 (Study I: th356) = -.66, p > .05; Study 2: t(322, = -3.34, p < .OOI ). Thus, mediation could only be tested in Study 2. When both transaction value and acquisition value were regressed on buyers` search intentions, acquisition value significantly affected search intentions (Study 2: t(316)= -3.74, p < .001), while the effect of transaction value was reduced and non-significant (Study 2: t(316) = .09, p > .OS). As was found for willingness to buy, acquisition value serves to mediate the relationship between perceived transaction value and search intentions. We also tested the mediation through nested models. We ran the proposed model (Figure 1 ) with an additional path (transaction valueto acquisition value). The results were X2 of630 (Study 1) and 841 (Study 2), both with df = 263. A nested model with the three fewer paths (no linkage between selling price and acquisition value, no linkages from acquisition value to willingness to buy and search intentions) had Z2 of 644.24 (Study I ) and 846.63 (Study 2), both with df = 266. Similar to the regression results, the nested model approach supports the mediation hypotheses for Study 2.
Footnote
6Acquisition value is conceptualized as a function of perceived quality of the product and the selling price. Monroe ( 1990) uses a ratio model but points out that this is only a way of illustrating the comparison. Associated literature on price-quality trade-offs have found greater support for a subtractive model (e.g., Levin and Johnson 1984; White and Truly 1989). We conducted two regression analyses:
Model 1: Perceived acquisition value as a function of perceived quality and advertised selling price.
Model 2: Perceived acquisition value as a function of perceived quality, advertised selling price, and an interaction term.
Footnote
The results of the regression analysis indicate the following: Model 1: Study I-F(2,301) = 78.67, adjusted R2 = .34; Study
2-F(2, 321) = 63.14, adjusted R2 = .28. Model 2: Study I-F(3,3oo) = 53. [, adjusted R2 = .347; Study 2-F(3.32o) = 42.66, adjusted R2 = .28.
Footnote
Interaction was not significant in either study. The results suggest that acquisition value might be represented best by a subtractive model. Results also suggest that transaction value is represented by a subtractive model. Model 1: Study 1-F(2.3o) = 36.77, adjusted R2 = .19; Study 2-F(233 1)= 12.73, adjusted R2 = .07. Model 2: Study II-F(3,303) = 24.451, adjusted R2 = .195; Study 2-F(3,330) = 8.47, adjusted R2 = .07. Interaction was not significant in either study.
References
REFERENCES
References
Anderson, James C. (1987), "An Approach for Confirmatory Measurement and Structural Equation Modeling of Organizational Properties," Management Science, 33 (4), 525-41. and David W. Gerbing (l988), "Structural Equation Modeling in Practice: A Review and Recommended Two-Step Approach," Psychological Bulletin, 103 (3), 411-23. Bagozzi, Richard P. and Youjae Yi (1988), "On the Evaluation of Structural Equation Models," Journal of the Academy of Marketing Science, 16 (Spring), 74-94.
Baron, Reuben M. and David A. Kenny (1986), "The Moderator Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations," Journal of Personality and Social Psychology, 51 (6), 1173-82.
References
Bearden, William O., Ajit Kaicker, Melinda Smith de Borrero, and Joel E. Urbany (1992), "Examining Alternative Operational Measures of Internal Reference Prices," in Advances in ConsumCer Research. Vol. 19, John F. Sherry, Jr. and Brian Sternthal, eds. Provo, UT: Association for Consumer Research, 629-35.
, Subhash Sharma, and Jesse E. Teel (1982), "Sample Size Effects on Chi-Square and Other Statistics Used in Evaluating Causal Models," Journal ofMarketing Research, 19 (November) 425 30.
Bentler, Peter M. (1990), "Comparative Fit Indexes in Structural Models," Psychological Bulletin, 107 (2), 238-46.
and Douglas G. Bonett (1980), "Significance Tests and Goodness of Fit in the Analysis of Covariance Structures," Psychological Bulletin, 88 (3), 588-606.
References
Bolton, Ruth N. and James H. Drew ( 1991), "A Multistage Model of Customers' Assessments of Service Quality and Value," Joun-al of Consumer Research, 17 (March), 375-84. Chapman, Joseph D. and Kent B. Monroe (1990), "The Framing Effects of Alternative Price Promotions on Buyers' Subjective
References
Product Evaluations," working paper, Department of Marketing, Ball State University.
Curry, David J. (1985), "Measuring Price and Quality Competition," Journal of Marketing, 49 (Spring), 106-17. Della Bitta, Albert J., Kent B. Monroe, and John M. McGinnis ( 1981), "Consumer Perceptions of Comparative Price Advertisements," Journal of Marketing Research, 18 (November), 416-27.
Dodds, William B., Kent B. Monroe, and Dhruv Grewal (1991), "The Effects of Price, Brand, and Store Information on Buyers' Product Evaluations," Journal of Marketing Research, 28 (August), 307-19.
References
Dolan, Robert J. (1995), "How Do You Know When the Price is Right'?" Harvard Business Review, 73 (September/October), 174-83.
Forbes, John L. and Nitin T. Mehta ( 1978), "Value-Based Strategies for Industrial Producs," Busines Horizon.sz 21 (October), 25-31. Fornell, Claes and David F. Larcker (1981), "Evaluating Structural Equation Models with Unobservable Variables and Measurement Error," Journal of Marketing Researrh, 18 (February), 39-50.
References
Grewal, Dhruv (1989), "The Effect of Intrinsic, Extrinsic Cues and Reference Prices on Buyers' Perceptions of Quality and Value," unpublished dissertation, Virginia Polytechnic Institute and State University.
(1995), "Product Quality Expectations: Towards an Understanding of Their Antecedents and Consequences," Journal of Business and Psychology, 9 (Spring), 22540. and Larry D. Compeau (1992), "Comparative Price Advertising: Informative or Deceptive?" Journal ofPublic Policy & Marketing, II (Spring), 5242.
and ( 1993), "Consumers' Interpretations ofSemantic Phrases in Comparative Price Advertisements: Some Prelimi
References
nary Evidence ona Public Policy Issue," in Advances in Consumer Research, Vol. 20, Leigh McAlister and Michael Rothschild, eds. Provo, UT: Association for Consumer Research, 479-80.
, Howard Marmorstein, and Arun Sharma (1996), "Communicating PriceInformation Through Semantic Cues: The Effects of Situation and Discount Size," Journal ofConsumer Research, 23 (September), 148-55.
Hastak, Manoj and Jerry C. Olson (1989), "Assessing the Role of Brand-Related Cognitive Responses as Mediators of Communication Effects on Cognitive Structure," Journal ofConsumer Research, 15 (March), 444-56.
Herr, Paul M. (1989), "Priming Price: Prior Knowledge and Context Effects," Journal ofConsumer Research, 16 (June), 67-75. Huppertz, John W., Sidney J. Arenson, and Richard H. Evans (1978), "An Application of Equity Theory to Buyer-Seller Exchange Situations," Journal ofMarketing Research, 15 (May) 250(60. Hyun, Soeun (1993), "The Effects of Contextual Cues onConsumers' Perceptions of Comparative Price Advertisements," doctoral dissertation, Department ofMarketing, Virginia Polytechnic Institute and State University.
References
Jacobson, Robert and David A. Aaker (1987), "TheStrategic Role ofProduct Quality," Journal ofMarketing, 51 (October), 3 -44. Joreskog, Karl G. and Dag Sorbom (1989), LISREL VII: Analysis ofLinear Structural Relationships by the Methods ofMaximum Likelihood. Mooresville, IN: Scientific Software, Inc. Kaufmann, Patrick J., N. Craig Smith, and Gwen Ortmeyer (1994), "Deception in Retailer High-Low Pricing: A Rule of Reason Approach," Journal of Retailing, 70 (Summer), I IS-38. Keon, John W. (1980), "The Bargain Value Model and a Comparison of Managerial Implications with the Linear Learning Model," Management Science, 26 (November), 1117-30. Klein, Noreen M. and Janet E. Oglethorpe (1987), "Reference Points in Consumer Decision Making,'` in Advances in Consumer Research, Vol. 14, Melanie Wallendorf and Paul F. Anderson, eds. Provo, UT: Association for Consumer Research, 183-87. Levin, Irvin P. and Richard D. Johnson (1984), "Estimating PriceQuality Tradeoffs Using Comparative Judgments," Journal of Consumer Research, II (June), 593-60.
References
Lichtenstein, Donald R. and William O. Bearden (1989), "Contextual Influences on Perceptions of Merchant-Supplied Reference Prices," Journal of Consumer Research, 15 (September), 55466. , Richard G. Netemeyer, and Scot Burton (1990), "Distinguishing Coupon Proneness from Value Consciousness: An Acquisition-Transaction Utility Theory Perspective," Journal ofMarketing, 54 (July), 54-67.
, Nancy M. Ridgeway, and Richard G. Netemeyer (1993), "Price Perceptions and Consumer Shopping Behavior: A Field Study," Journal of Marketing Research, 30 (May), 234-45. Marmorstein, Howard, Dhruv Grewal, and Raymond P. H. Fishe (1992), "The Value of Time Spent in Price Comparison Shopping: Survey and Experimental Evidence," Journal of Consumer Research, 19 (June), 5241.
References
Marsh, Herbert W., John R. Balla, and Roderick P. McDonald (1988), "Goodness-of-Fit Indexes in Confirmatory Factor Analysis: The Effects of Sample Size," Psychological Bulletin, 103 (3), 39110.
Monroe, Kent B. (1973), "Buyers' Subjective Perceptions of Price," Journal of Marketing Research, 10 (February), 70-80. (1990), Pricing: Making Profitable Decisions, 2d ed. New York: McGraw-Hill.
and Joseph D. Chapman (1987), "Framing Effects on Buyers Subjective Product Evaluations," in Advances in Consumer Research, Vol. 14, Melanie Wallendorf and Paul Anderson, eds. Provo, UT: Association for Consumer Research, 193-97. , Dhruv Grewal, and Larry D. Compeau (1991), "The Concept of Reference Prices: Theoretical Justifications and Research Issues," presented at the Association for Consumer Research Conference, Chicago (October).
References
and R. Krishnan (1985), "The Effect of Price on Subjective Product Evaluations," in Perceived Quality: How Consumers View Stores and Merchandise, Jacob Jacoby and Jerry C. Olson, eds. Lexington, MA: Lexington Books, 209-32. Nagle, Thomas T. and Reed K. Holden (1995), The Strategy and
Tactics of Pricing, 2d ed. Englewoods Cliffs, NJ: Prentice Hall. Oliver, R. L. and J. E. Swan ( 1989), "Consumer Perceptions of Interpersonal Equity and Satisfaction in Transactions: A Field Survey Approach," Journal of Marketing, 53 (April), 21-35. Phillips, Lynn W., Dae R. Chang, and Robert D. Buzzell (1983), "Product Quality, Cost Position and Business Performance: A Test ofSome Key Hypotheses," Journal of Marketing, 47 (Spring), 2643.
Rajendran, K. N. and Gerard J. Tellis ( 1994), "Contextual and Temporal Components of Reference Price," Journal of Marketing, 58 (January), 22-34.
References
Rao. Akshay R. and Kent B. Monroe (1988), "The Moderating Effect of Prior Knowledge onCue Utilization in Product Evaluations," Journal ofConsumer Research, IS (September), 253-64.
and (1989), "The Effect of Price, Brand Name, and Store Name on Buyers' Subjective Product Assessments: An Integrative Review," Journal ofMarketing Research, 26 (August), 351-57.
Schindler, Robert M. ( 1989), "The Excitement of Getting a Bargain: Some Hypotheses Concerning theOrigins and Effects of Smart Shopper Feelings," in Advances in Consumer Research, Vol. 16, Thomas K. Srull, ed. Provo, UT: Association for Consumer Research, 447-53.
Shapiro, Benson P. and Barbara B. Jackson ( 1978), "Industrial Pricing to Meet Customer Needs," Harvard Business Review, (November-December), 119-127.
, V. Kasturi Rangan, Rowland T. Moriarty, and Elliot B. Ross (1987), "Manage Customers for Profits (Not just Sales)," Harvard Business Review, 65 (September/October), 101-108.
References
Stigler, George J. (1961), "The Economics of Information," Journal of Political Economy, 69 (June), 213-25. Szybillo, George J. and Jacob Jacoby (1974), "Intrinsic Versus Extrinsic Cues as Determinants of Perceived Quality," Journal of Applied Psychology, 59 (February), 74-78. Thaler, Richard (1985), "Mental Accounting and Consumer
Choice," Marketing Science, 4 (Summer) 199-214. Tucker, Ledyard R. and Charles Lewis (1973), "A Reliability Coefficient for Maximum Likelihood Factor Analysis," Psychometrika, 38 (1), 1-10.
Urbany, Joel E. and William O. Bearden (1989), "Reference Price Effects on Perceptions of Perceived Offer Value, Normal Prices, and Transaction Utility," in Enhancing Knowledge Development in Marketing, Paul Bloom et al., eds. Chicago: American Marketing Association, 45-49.
and (1990), "The Effects ofAdvertised PriceInformation on Value Perceptions," unpublished manuscript, Department ofMarketing, University of South Carolina.
, and Dan C. Weilbaker (1988), "The Effect of Plausible and Exaggerated Reference Prices onConsumer Perceptionsand Price Search," Journal ofConsumer Research, 15 (June), 95-110.
References
and Peter R. Dickson (1990), "Consumer Knowledge of Normal Prices: An Exploratory Study and Framework," Working Paper Series, Report No. 90-112, Cambridge, MA: Marketing Science Institute.
White, J. Dennis and Elise L. Truly (1989), "Price-Quality Integration in Warranty Evaluation," Journal of Business Research, 19 (September), 109-25.
Zeithaml, Valarie A. (1988), "Consumer Perceptions of Price, Quality, and Value: A Means-End Model and Synthesis of Evidence," Journal ofMarketing, 52 (July), 2-22.
AuthorAffiliation
Dhruv Grewal is Associate Professor of Marketing, Department of Marketing, University ofMiami. Kent B. Monroe is J. M. Jones Professor of Marketing, University of Illinois. R. Krishnan is a professor, Department of Marketing, California Polytechnic Institute and State University. This article has benefited from comments and suggestions of James Littlefield, Julie Ozanne, Edward Fern, Dennis Hinkle, Michael Levy, Arun Sharma, Howard Marmorstein, Diana Grewal, Carolyn Costley, Banwari Mittal, Tamara Mangleburg, and A. Parasuraman. The authors also acknowledge the valuable feedback provided by the three anonymous JM reviewers. Financial support from a Cunningham Fellowship, Marketing Science Institute, and University of Miami Summer Research Grant is gratefully acknowledged.
Copyright American Marketing Association Apr 1998

The effects of price-comparison advertising on buyers' perceptions of acquisition value, transaction value, and behavioral intentions

2012-01-11T08:27:00.005+07:00

The dual credibility model: The influence of corporate and endorser credibility on attitudes and purchase intentions

2012-01-11T08:24:00.000+07:00

The dual credibility model: The influence of corporate and endorser credibility on attitudes and purchase intentions
Lafferty, Barbara A; Goldsmith, Ronald E; Newell, Stephen J. Journal of Marketing Theory and Practice 10. 3 (Summer 2002): 1-12.

A study proposes a theory of the combined influence of corporate and endorser credibility. Participants assessed the credibility of the companies and spokespersons as well as their attitudes toward the ads and brands, and their intent to purchase the advertised product. The covariance matrix was subjected to a path analysis. The model fit the data, and the findings corroborated prior research indicating that both types of source credibility have an impact on attitudes and purchase intentions albeit a differential one. The results suggest that the Dual Credibility Model partially predicts and explains advertising effectiveness for these dual sources of credibility.

Headnote
This study proposes a theory of the combined influence of corporate and endorser credibility. Participants assessed the credibility of the companies and spokespersons as well as their attitudes toward the ads and brands, and their intent to purchase the advertised product. The covariance matrix was subjected to a path analysis. The model fit the data, and the findings corroborated prior research indicating that both types of source credibility have an impact on attitudes and purchase intentions albeit a differential one. The results suggest that the Dual Credibility Model partially predicts and explains advertising effectiveness for these dual sources of credibility.

INTRODUCTION
In the quest to maximize the effectiveness of advertising, marketers and researchers have endavored to understand the relationship between consumers' attitudes toward an advertisement (Aad), their attitudes toward the advertised brand (AB), and how these attitudes influence their purchase intent (PI). The Dual Mediation Hypothesis (DMH) (Lutz, MacKenzie, and Belch 1983; MacKenzie, Lutz, and Belch 1986; MacKenzie and Lutz 1989; Homer 1990; Brown and Stayman 1992) has been shown to be robust in depicting the causal sequence of Aad, AB, and PI in an advertising context, thus providing marketers with a better understanding of the interrelationships among these variables. Since the development of this model, researchers and marketers have explored the determinants of consumers' attitudes toward advertisements and brands in greater detail in order to develop more persuasive advertising and marketing campaigns.
Source credibility has been proposed as an important antecedent to attitudes to ward an advertisement and ultimately, advertising effectiveness (e.g., Lutz et al. 1983). Originally, the concept of credibility meant the credibilityof the endorser or the spokesperson in the ad (e.g., Bergin 1962; Aronson, Turner, and Carlsmith 1963; Bochner and Insko 1966; Sternthal, Phillips, and Dholakia 1978; Ohanian 1990). Many studies support the generalization that perceived source (i.e., spokesperson) credibility influences attitudes and behavioral intentions (e.g., Sternthal, Dholakia, and Leavitt 1978; Harmon and Coney 1982; Wu and Shaffer 1987; Moore, Hausknecht, and Thamodaran 1988). Spokesperson credibility has been operationalized generally to include three dimensions: expertise, trustworthiness, and attractiveness (DeSarbo and Harshman 1985; Ohanian 1990). The inclusion of attractiveness was prompted by research suggesting that a physically attractive communicator is liked more and has a positive impact on opinion change and product evaluations (e.g., Joseph 1982).
More recently, corporate credibility also has been shown to play an important role in influencing attitudes and purchase intentions (Lafferty and Goldsmith 1999; Goldsmith et al. 2000a, 2000b). While this form of source credibility does appear to have an effect on attitudes toward thead, studies also suggest that it has a stronger effect on attitudes toward the brand (Lafferty and Goldsmith 1999; Goldsmith et al. 2000a, 2000b). In earlier studies, the influence of corporate credibilityon consumers' attitudes and purchase intentions had only been alluded to in the persuasion literature (e.g., Kalwani and Silk 1982; MacKenzie and Lutz 1989; Goldberg and Hartwick 1990) or in managerially focused discussions (Gregory 1991; Sobol et al. 1992). Previous studies have referred to corporate credibilityas advertiser credibility(Lutz et al. 1983; Lutz 1985; Mackenzie and Lutz 1989), advertiser reputation (Goldberg and Hartwick 1990) and only more recently, as a determinant of corporate reputation (Fombrun 1996; Keller 1998). Although it is acknowledged to be an important part of corporate image and reputation, corporate credibility has not been systematically studied in the context of its effect on attitudes and purchase intentions.
Corporate credibility forms part of a positive corporate image or reputation (Gregory 1991; Sobol et al. 1992; Fombrun 1996; Keller 1998). A corporate image is thetotality of impressions that a firm makes on the minds of consumers and is evoked by the corporate name or logo (Pope and Voges 1999). Corporate credibility, or the extent to which consumers, investors, and other constituents believe in a company's trustworthiness and expertise, makes up a portion of a corporation's image (Fombrun 1996). A corporation's credibility can be harmed when it is suspected of producing dangerous or inferior products, when it lies to consumers or others, or when it is reported to be in violation of legal or ethical norms (Frombrun 1996). The ethics literature indicates that when ethical perceptions are exceeded by company actions, the company is rewarded in terms of more positive attitudes and behavioral intentions (e.g., Creyer and Ross 1997). Corporate credibility plays a critical role in the firm's ability to secure loans, create partnerships, and market products (Gregory 1991; Haley 1996). Corporate credibilityis also important in shaping corporate identity (Stuart and Kerr 1999). Consumers who perceive a company as credible are more likely to evaluate thefirm's advertisements favorably and to buy the company's products (Keller 1998).
Goldsmith et al. (2000a) describe the causal relationships that exist for corporate credibilityand endorser credibilityin the context of the latent variables in the DMH model. The development of their proposed modelwas based on an earlier study suggesting the effect of both antecedents on Aad, AB, and PI (Lafferty and Goldsmith 1999). The purpose of this study is to empirically test the Dual Credibility Model(DCM) proposed by Goldsmith et al. (2000a) in an advertising context and suggest theoretical, managerial, and methodological implications from the findings.
THE DUAL CREDIBILITY MODEL
The Dual Credibility Model(see Figure 1) builds upon the latent variables in the DMH model, which specifies a direct causal sequence from Aad to AB to PI (see MacKenzie et al. 1986). The DMH specifies that ad cognitions represent an antecedent to Aad and have a direct effect on this variable. Within the context of ad cognitions, endorser credibilitycan be one of the variables consumers use to judge the advertisement and thus, its persuasiveness (e.g., Sternthal, Phillips, and Dholakia 1978; Harmon and Coney 1982; Moore et al. 1988). This, in turn, can lead to attitude formations toward theadvertisement. Seven hypothesized paths among thefive variables comprise the model.
The effects of endorser or source credibilityon advertising effectiveness have been well documented in the marketing and social psychology literature (e.g., Aronson, Turner, and Carlsmith 1963; Sternthal, Phillips, and Dholakia 1978; Harmon and Coney 1982;). According to Fishbein and Ajzen (1975), source credibilityaffects the probability that a message claim will be accepted. In general, a message delivered by a high credibilitysource will be accepted more readily and is more likely to lead to greater attitude change (Kelman and Hovland 1953; Johnson, Torcivia, and Poprick 1968; Miller and Baseheart 1969; Warren 1969; Schulman and Worrall 1970). In thecontext of Aad -> AB -> PI, the perceived credibilityof the endorser has been shown to have an influential effect on Aad in thestudy by Lafferty and Goldsmith (1999) and is one of the two exogenous variables in theproposed DCM (Goldsmith et al. 2000a). Thus, theDCM hypothesizes that
H1: Endorser credibilityis positively and directly related to attitude-toward-the-ad.
The findings by Lafferty and Goldsmith (1999) also showed that corporate credibility influences Aad directly. While the effects on Aad were weaker than the effects produced by the credible endorser, the results suggested that a direct path between the two variables did exist. MacKenzie and Lutz (1989) also indicated the likelihood of this relationship. Their modelshowed that advertiser credibility(synonymous with corporate credibility) had a strong positive effect on attitude toward the advertiser, which in turn, had a strong positive effect on Aad. The causal path from corporate credibilityto Aad was confirmed in the study by Goldsmith et al. (2000a), and thus corporate credibility represents thesecond exogenous variable in the DCM. Therefore, it is hypothesized that
H2: Corporate credibilityis positively and directly related to Aad.
Corporate credibility also has been shown to have a direct effect on AB (Lafferty and Goldsmith 1999). MacKenzie and Lutz (1989) indicated that advertiser credibilityis more a central processing cue. This would suggest that consumers' evaluations ofthe company's credibilitywould be incorporated into their assessment of the brand, prompting more cognitive processing. According to the Elaboration Likelihood Model (ELM), when consumers are more motivated to centrally process an ad, the brand relevant aspects increase and the peripheral cues decrease (Petty and Cacioppo 1983). The direct effect of corporate credibilityon AB was confirmed in the study by Goldsmith et al. (2000a). Thus, the DCM posits that
H3: Corporate credibilityis positively and directly related to AB.
Lafferty and Goldsmith (1999) also showed that corporate credibilityhad a direct effect on PI. This empirical evidence is consistent with studies by Newell (1993) and Davis (1994) who found a positive effect of corporate credibility on all three outcome variables, Aad, AB, and PI. Perceptions of the company's trustworthiness and expertise are part of the information consumers use to judge the company's products and determine whether they are inclined to buy them (Fombrun 1996). Firestone Tire Company exemplifies this phenomenon. The company suffered serious problems in 2000 from all the negative publicity it received regarding its Wilderness tires and their propensity to explode causing SUV's, particularly Ford Explorers, to flip and roll. Firestone stock fell 60 percent from August 2000 to January 2001 (Dawson 2001) and sales fell 40% in September and October 2000 (Davis 2000). The study by Goldsmith et al. (2000a) also supported the direct relationship between corporate credibility and PI. Therefore, it is hypothesized for the DCM that
H4: Corporate credibility is positively and directly related to Pl.
Substantial research has been done on the causal sequence of Aad, AB, and PI to explain advertising effectiveness (Heath and Gaeth 1994). The majority of findings suggest that the order of effects appears to be robust (MacKenzie et al. 1986; MacKenzie and Lutz, 1989; Homer 1990; Brown and Stayman 1992). This causal sequence is reflected as the latent variables in the DMH modeland is also the outcome variables in the DCM. Thus,
115: Aad is positively and directly related to AB. H6: AB is positively and directly related to PI.
Finally, there is evidence for a direct relationship between Aad and PI in certain circumstances (MacKenzie et al. 1986; Cox and Locander 1987; Biehal et al. 1992). This appears to be true when affective responses are evoked ( Petty, Cacioppo, and Goldman 1981; Gorn 1982; Batra and Ray 1985; Cox and Locander 1987;). Consumer may decide what they would buy without completely processing all brand information (see Biehal et al. 1992). The ads in this study contained no strong arguments and used minimal copy thus, providing limited information about the product and relied primarily on the influence of the endorser and on the information provided about the company to test the model. It is likely that the photograph of the endorser and the company description generated some feelings or affect either positive or negative owing to the lack of product information (Mitchell 1986; Stayman and Batra 1991). Thus, the evaluations stimulated by theads would more than likely be chiefly affective or emotional in nature and not largely analytical. This affective response to theads could influence the declarations of purchase intent directly without all the effects being mediated by the brand evaluations. Any cognitive processing that does occur to influence purchase intent would primarily be an outcome of attitude toward the brand. The specific selection of product and type of ad used for this study is similar to the stimuli used by Goldsmith et al. (2000a) and therefore is hypothesized to produce similar results. Thus:
H7: When an ad contains few and weak arguments, Aad is positively and directly related to PI.
In sum, theDCM posits both direct and indirect differential relationships between endorser and corporate credibilitywith important reactions to advertisements.
METHOD
Subjects
The data came from 315 undergraduate students who were taking an introductory marketing course and volunteered to participate in an advertising experiment. The data were collected in a manner similar to that used by Simonin and Ruth (1998), who randomly exposed small groups of subjects to different combinations of brands of microprocessors and automobiles and then tested a modelusing the aggregated data. Data were combined from two groups of subjects who were exposed to different combinations of endorsers and companies. The first group consisted of 161 women marketing students at a university in the U.S. Midwest. They were randomly assigned to one of four experimental treatments in a 2 (high/low corporate credibility) x 2 (high/low endorser credibility) between-subjects factorial design with female endorsers. The second group of subjects was 154 male students from the same Midwest University. They were randomly assigned to one of four experimental treatments in the same 2 x 2 design, but male endorsers were used in their ads. Men and women saw endorsers of the same sex because a pretest showed same sex endorsers to be viewed by the participants as more realistic. Women, especially, were not effective endorsers of athletic shoes for men. Thus, the only difference in the treatment between the two samples was the endorser used in the ad (see Stimuli section for details). In both cases, pretests were conducted to select suitable high and low credibility sources for the male and female endorsers. Both high credibility endorsers were athletes, and both low credibility endorsers were actors of similar weight and proportions. Because identical measures of all constructs were taken in both groups, cross validation procedures were employed to test the robustness of the findings.
Stimuli
Subjects saw one of two ads for Pride, a purportedly new brand of athletic shoes. This is a relevant product category that the student subjects were highly familiar with and with which most had purchasing experience. The study used a fictitious brand of athletic shoes to prevent prior knowledge of a brand from biasing the results of the study. The women subjects in the high endorser credibility condition viewed an ad featuring Florence Griffith-Joyner, the Olympic gold medallist, as its spokesperson. (The study was conducted prior to the untimely death of Ms. Griffith-Joyner.) Women in the low endorser credibility condition saw the same ad with Roseann Barr, the actress, as the spokesperson. These endorsers were used after a pretest suggested they represented opposite poles of endorser credibility for athletic shoes. For the men, discussions with students suggested that the golfer, Tiger Woods, would be a positive spokes person for running shoes; and Wayne Knight, the actor who played Newman on Seinfeld, was suggested as an appropriate substitute for Roseanne Barr. The questionnaire booklet was otherwise identical, and the same procedures were followed.
Providing descriptions of the fictitious company that manufactured the shoes influenced perceptions of high and low corporate credibility. Once again, a fictitious company was selected to prevent prior knowledge of a company unduly influencing the respondents. One corporate history presented a positive image while the other presented the company in a negative manner. This manipulation was adapted from a procedure used by Goldberg and Hartwick (1990) who successfully used it to manipulate the reputation of a company. Company descriptions were approximately equal in length and, except for the positive or negative tone, presented virtually the same information. Perceptions of the company's length of time in business, their degree of national and international success, the size of the company, and recognition of their social consciousness were manipulated to reflect a positive versus a negative (less positive) image. The only difference in content was the positive manipulation contained information on the company's community and environmental contributions while the negative manipulation contained information on poor quality control and alleged SEC violations. These elements are typical cues used by consumers to assess how credible a corporation might be.
Procedure
Four booklets containing different endorser/credibility combinations were prepared. The front page of each booklet contained general instructions for the study and the cover story. Subjects were told that the study assessed student reactions to magazine advertisements. They were instructed to read each page carefully and to look at the ad as if they were seeing it in a magazine (Yi 1990). The subjects were informed that when they finished, they should keep the booklet face down until it was collected.
The second page of the booklet contained the description of the corporation either negative or positive depending on the treatment (Goldberg and Hartwick 1990). The students were instructed to read the excerpt, which was said to be from a Wall Street Journal article. Following the description, they were instructed to proceed without turning back.
The next page showed the ad for Pride athletic shoes that contained a picture depicting one of the four endorsers. The layout for each ad as well as the headline and copy were identical in each of the four conditions to eliminate the possibility of any potential confounds. The copy consisted of a short quotation that was attributed to the endorser stressing the natural fit of the shoe. No other product attributes or information about benefits was provided. Thus, the only differences among the four treatments were the endorser and the credibilityof the corporation.
Measures
Six 7-point bi-polar adjective scales assessed endorser credibility. These scales were derived from those developed by Ohanian (1990) for measuring celebrity endorsers' perceived attractiveness, trustworthiness, and expertise. Two items, "attractive/unattractive" and "classy/not classy," measured attractiveness; two items with endpoints of "sincere/insincere" and "trustworthy/untrustworthy" measured trustworthiness; and two items with anchors of "expert/not an expert" and "experienced/inexperienced" measured expertise. Corporate credibility was measured via the Corporate Credibility Scale developed by Newell and Goldsmith (2001). This is an eight-item, 7-point Likert scale with a five-point, agree/disagree response format. Four items measure corporate honesty or trustworthiness (e.g., I trust the Pride Corporation."), and four items measure corporate expertise (e.g., "The Pride Corporation has great expertise."). These are two of the major dimensions of corporate credibility(Fombrun 1996; Haley 1996).
To measure attitude toward the ad, attitude toward the brand, and purchase intention (cf., Tripp et al. 1994) participants were first asked to use three 7-point, bi-polar adjective scales to rate their overall impression of the advertised product (AB). The scales were anchored with "good/bad," "favorable/ unfavorable," and "satisfactory/unsatisfactory" (Bruner and Hensel 1992). Following this measure of attitude-toward-the-- brand, the subjects were asked to rate their overall impression of the ad (Aad) on three 7-point bi-polar scales anchored by "good/bad," "favorable/unfavorable," and "pleasant/ unpleasant" (MacKenzie and Lutz, 1989). Next, the subjects were asked how likely it would be that they would consider buying that brand (PI) the next time they purchased athletic shoes. Three 7-point bi-polar scales were used, anchored by "very likely/very unlikely," "probable/improbable," and "possible/impossible" (Yi 1990).
The final questions in the booklet asked for demographic data, and one question asked subjects if they could identify the purpose of participating in the study. This was done to assure that no one guessed the hypotheses, which could bias their responses and invalidate their answers. The responses showed that no subject appeared to be aware of the intent of the study, so all questionnaires were used.
RESULTS
Preliminary Analyses
The items used to measure the manipulations and dependent variables were factor analyzed using the SPSS principal axis factor method followed by an oblique rotation to determine scale dimensionality and item convergent and discriminant validity. The six items that measured endorser credibility were factor analyzed. Only one factor with an eigenvalue greater than 1.0 was extracted. The six items were summed to form a measure of endorser credibility. We assessed internal consistency by computing Cronbach's coefficient alpha (alpha =.93).
Table 1 presents the factor analysis results for the corporate credibilityscale and the Aad, AB, and PI scales. For the corporate credibility, the factor analysis extracted two factors with eigenvalues greater than 1.0. Four items formed an "honesty" factor with the other four items making up an "expertise" factor (see Newell and Goldsmith 2001). We summed the eight items to form an overall measure of corporate credibility. We assessed internal consistency by computing coefficient alpha (alpha = .92).
The nine items measuring Aad, AB, and PI were also factor analyzed (see Table 1). The results showed a three-factor structure with the items loading on the"correct" factors. The three items measuring Aad were then factor analyzed and found to be unidimensional, with one factor accounting for 81 percent of the variance. Factor analysis was also performed on the three items assessing AB and on the three used to assess PI. Both sets also demonstrated unidimensionality, with one factor accounting for 90 percent of the variance in AB and one factor accounting for 89 percent of the variance in PI. Since each set of items was unidimensional, the items comprising each scale were summed to assess overall Aad, AB, and PI for Pride athletic shoes. These three scales had high internal consistency with coefficient alphas of .93, .96, and .96 for Aad, AB, and PI, respectively.
Thus, there is strong evidence that the items used to measure the variables in this study possessed convergent and discriminant validity as well as high internal consistency. The scales were all scored so that higher scores represented higher levels of the construct.
Pooling the Data
DISCUSSION
The purpose of this study was to empirically test the Dual Credibility Modelto determine if it could serve as a predictor of advertising effectiveness when evaluating dual sources of credibility. The findings are consistent with Goldsmith et al. (2000a) and argue for the usefulness of this modelin both theoretical and managerial contexts.
Theoretical Implications
Thefindings support Stem's (1994) argument that consumers do not perceive only a single communicator but may differentiate between the presenters and the companies they represent. While additional applications of this model need to be tested to confirm that the proposed paths are robust under varying conditions, this study supports the hypothesized influences of endorser and corporate credibility on Aad, AB, and PI. Similar to Goldsmith et al. (2000a), this study confirmed that endorser credibility had a greater effect on Aad (.24) than corporate credibility(.083), although both paths were significant. Moreover, this study confirms the hypothesis that corporate credibility has a direct effect on all three advertising-related variables, unlike endorser credibility, which mainly influences attitudes toward the ad. Credible endorsers seem to influence attitudes toward the brand and purchase intentions indirectly. This study also lends support to the hypothesized direct path from attitude toward the ad to purchase intent (Biehal et al. 1992). These findings represent an advance in our understanding of the complex interactions among the various features of advertising and consumer cognitions that lead ultimately to purchase. They focus attention on the impact of corporate credibility on consumer behavior in a domain hitherto preoccupied with spokesperson and endorsers.
Managerial Implications
This study indicates that corporate credibilityplays an important role in consumer evaluation of advertisements. In addition, a company's credibility seems to have a direct effect on perceptions of brands and of purchase intentions. Because these company perceptions may be central to the success or failure of promotions and sales, managers need to become much more aware of how consumers view the corporation (Fombrun 1996). Consequently, companies must constantly monitor perceptions of corporate credibility by consumers as well as other stakeholder groups. Annual, or even more frequent, surveys of consumers and their feelings about a corporation seem to be fundamental for effective decisionmaking. By tracking consumer perceptions of corporate credibility, managers can detect changes in this important marketing metric and take corrective action if needed. The measures for corporate credibility presented in this model can provide managers with a better idea of how well the company is viewed in the consumer's eyes.
Companies that face tough times may be particularly well served by the constant monitoring of consumer perceptions of their company's credibility. By understanding how perceptions have changed, and in particular how each component of the credibility construct has been altered, companies such as Exxon during the Valdez incident, Microsoft during their antitrust suit, and Anderson Consulting during and after its association with the Enron scandal, may have been better able to develop appropriate responses in order gain back lost trust and/or perceptions of expertise. For example, if a company found that after a recent crisis perceptions of trust among their distributors and feelings of competency and expertise among end consumers had significantly decreased, a two-pronged response could be developed. In this situation, a company might use sales representatives to help solidify their relationships with distributors by increasing visits to accounts and providing increased guarantees on products and greater reassurances to their customers about their commitment to serving them better. In addition, advertising and other media driven strategies for consumers could emphasize the long history of service and dependability and may use a highly respected spokesperson to present logical and compelling arguments about why the company is still the best choice. Overall, in times of crisis, those companies that have kept informed of consumers perceptions of company credibility may be better equipped to develop strategies to regain feelings of trust and/or perceptions of expertise among their target markets.
Coupling this measure of corporate credibility with the measure for the credibility of any endorser the company uses will allow managers to see the relative value of each credible source. This, in turn, should prove to be a useful measure for promotional and marketing budget allocations. This information can provide managers with a better understanding how consumers view their company, endorsers, and their advertisements. With this information managers can then develop more appropriate promotional strategies for both their products and their corporations.
Well thought-out strategies for maintaining a good reputation or improving a poor image are an essential step to developing effective advertisements. Specific marketing communications strategies to enhance a company's credibility may include developing appropriate corporate advertising that focuses on one or both of the dimensions of credibility. Another means to improve corporate reputation may be through public relations and sponsorships. In general, without a credible reputation, successful advertising is much more difficult to develop. Partnering with charitable causes in cause-brand alliances may also be a good way to enhance corporate reputation.
Methodological Implications
The findings promote the use of modeling to study these phenomena. Structural analysis provides researchers with the opportunity to test models using a single comprehensive method that examines multiple relationships simultaneously. Testing a model in this manner allowed us to test all the relationships among corporate and endorser credibility simultaneously with measures of reaction to advertising in order to corroborate experimental studies. Both methods (experimentation and causal modeling) should be used in tandem to provide a comprehensive explanation of marketing phenomena. Moreover, the results confirmed the reliability and validity of the measures of endorser and corporate credibility, thus promoting their use by other researchers in any studies of these constructs.
Limitations to Generalizability
The DCM was tested for a single product with a fictitious brand in a single ad exposure condition. Clearly, to establish any sort of generalizable conclusions for this model, more tests to extend these critical dimensions must be examined. Differences for the model may exist for durable versus nondurable products, for higher versus lower involvement products, and for authentic and more recognizable brands. Multiple exposures to the ad also may influence the various relationships. The endorsers selected also limited the study's generalizability. While pretests ensured that the endorsers selected were appropriate for the treatment conditions, other endorsers may change the strength of the relationships between variables or the very relationship. Certainly, the selection of the low credibility endorsers may be considered extreme, but they did represent realistic examples of low credibility spokespersons. In addition, the manipulation for corporate credibility presented two opposite corporate images. While this manipulation has been used in the past with success (Goldberg and Hartwick 1990; Lafferty and Goldsmith 1999), it may unduly influence the respondents' impressions of brand quality, producing a stronger effect on attitudes toward the brand. This is not necessarily an inappropriate representation of a company since this type of description, whether positive or negative, can exist in the real world.
While the DCM appears robust in this study, the weak link in the modelis the path from attitude toward the ad to purchase intent (M). This path exists under certain circumstances and possibly will not be robust when the content of the ad changes or with products that engender less affect. Additional study should be devoted to delineating the circumstances where this direct path pertains and where it does not.
Specifications of the potential limits of our findings, however, do not necessarily diminish their value. The DCM may represent the basic structure of response processes that exist when consumers are exposed to a new brand. This study also extends the generalizability of the findings of the DCM by Goldsmith et al. (2000a) since they used a real company, an institutional ad, and a single, relatively high credibility endorser to test the proposed model.
Future Research
The limitations provide opportunities for future research to test the viability of the DCM. Different products, different types of ads (institutional, comparative, reminder), different endorsers (CEO's versus celebrities), different companies (real versus fictitious), different media (broadcast versus print) all present opportunities to test the model under different conditions. Additional moderating variables also should be examined such as knowledge or involvement to see what impact these variables have on the DCM. These factors may influence the strength of the relationships among the variables in the model and are, therefore, candidates for discussion.
References
REFERENCES
References
Anderson, J. C. and Gerbing, D. W. (1988), "Structural Equation Modeling In Practice: A Review And Recommended Two-Step Approach," Psychological Bulletin, 103(3),411-423.
Aronson, E., Turner, J. A. and Carlsmith, J. M. (1963), "Communicator CredibilityAnd Communications Discrepancy As Determinants Of Attitude Change," Journal of Abnormal and Social Psychology, 67 (July), 31-36.
Batra, R. and Ray, M. L. (1985), "How Advertising Works At Contact," in L. Alwitt and A. Mitchell (eds) Psychological Processes and Advertising Effects: Theory, Research, and Application, Hillsdale, NJ: Lawrence Erlbaum, 13-43.
Bergin, A. E. (1962), "TheEffect Of Dissonant Persuasive Commications Upon Changes In Self-Referring Attitudes," Journal of Personality, 30 (MarchDecember), 423-438.
Biehal, Gabriel, Debra Stephens, and Eleonora Curlo (1992), "Attitude Toward theAd and Brand Choice," Journal of Advertising, 21 (3), 19-36.
Bochner, S. and Insko, C. A. (1966), "Communicator Discrepancy, Source Credibility, And Opinion Change," Journal of Personality and Social Psychology, 4 (December), 614-621.
References
Brown, S. P. and Stayman, D. M. (1992), "Antecedents And Consequences Of Attitude Toward TheAd: A MetaAnalysis," Journal of Consumer Research, 19 (June), 34-51.
Bruner, G. C. II and Hensel, P. J. (1992) Marketing Scales Handbook: A Compilation of Multi-Item Measures, Chicago, IL: American Marketing Association.
Cox, D. and Locander, W. (1987), "Product Novelty: Does It Moderate TheRelationship Between Ad Attitudes And Brand Attitudes," Journal of Advertising, 16 (3), 39-44.
References
Creyer, Elizabeth and William T. Ross (1997), "TheInfluence of Firm Behavior on Purchase Intentions: Do Consumers Really Care about Business Ethics?" Journal of Consumer Marketing, 14 (6), 421-428.
Davis, Bruce (2000), "Firestone Prepares Effort to Rebuild Damaged Brand,"Advertising Age 71 (December), 8. Davis, J. J. (1994), "Consumer Response To Corporate
Environmental Advertising," Journal of Consumer Marketing, 11 (2), 25-47.
Dawson, Chester (2001), "What Japanese CEO's can Learn from Bridgestone," Business Week (January), 50.
References
DeSarbo, W. S. and Harshman, R. A. (1985), "Celebrity-- Brand Congruence Analysis," in J. H. Leigh and C. R. Martin, Jr. (eds) Current Issues and Research in Advertising, Ann Arbor, MI: Division of Research Graduate School of Business Administration, TheUniversity of Michigan, 17-52.
Fishbein, Martin and Ajzen, Icek (1975), Belief Attitude, Intention, and Behavior: An Introduction to Theory and Research, Reading, MA: Addison-Wesley.
Fombrun, C. J. (1996), Reputation, Boston, MA: Harvard Business School Press.
References
Goldberg, M. E., and Hartwick, J. (1990), "TheEffects Of Advertiser Reputation And Extremity Of Advertising Claim On Advertising Effectiveness," Journal of Consumer Research, 17 (September), 172-179.
Goldsmith, Ronald E., Lafferty, Barbara A., and Newell, Stephen J. (2000a), "TheImpact Of Corporate CredibilityAnd Endorser Celebrity On Consumer Reaction To Advertisements And Brands," Journal of Advertising, (Fall), 43-54.
Goldsmith, Ronald E., Lafferty, Barbara A., and Newell, Stephen J. (2000b), "TheInfluence of Corporate Credibilityon Consumer Attitudes and Purchase Intent," Corporate Reputation Review, 3(4),304-318.
Gorn, G. J. (1982), "TheEffects Of Music In Advertising On Choice Behavior: A Classical Conditioning Approach," Journal of Marketing, 46 (Winter), 94101.
References
Gregory, J. (1991), Marketing Corporate Image, Lincolnwood, IL: NTC Business Books.
Harmon, R. and Coney, K. (1982), "ThePersuasive Effects Of Source CredibilityIn Buy And Lease Situations," Journal ofMarketing Research, 19 (May), 255-260.
Haley, E. (1996), "Exploring TheConstruct Of Organization As Source: Consumers' Understandings Of Organizational Sponsorship Of Advocacy Advertising," Journal of Advertising, 25 (Summer), 19-35.
Heath, T. B. and Gaeth, G. J. (1994), "Theory And Method In TheStudy Of Ad And Brand Attitudes: Toward A Systemic Model." In E. M. Clark, T. C. Brock, and D. W. Stewart, (eds). Attention, Attitude, and Affect in Response to Advertising. Hillsdale, NJ: Lawrence Erlbaum, 125-148.
References
Homer, P. M. (1990), "TheMediating Role Of Attitude Toward TheAd: Some Additional Evidence," Journal of Marketing Research, 27 (February), 7886.
Johnson, H., Torcivia, J., and Poprick, M. (1968), "Effects Of Source CredibilityOf TheRelationship Between Authoritarianism And Attitude Change," Journal of Personality and Social Psychology, 9, 179-183.
Joseph, W. B. (1982), "The CredibilityOf Physically Attractive Communicators: A Review," Journal of Advertising 11, (July), 15-24.
Kalwani, M. U. and Silk A. (1982), "On TheReliability And Predictive Validity OfPurchase Intention Measures, Marketing Science, I (Summer), 243-286.
Keller, K. L. (1998), Strategic Brand Management. Upper Saddle River, NJ: Prentice Hall.
Kelman, H. and Hovland, C. (1953), "Reinstatement Of TheCommunicator In Delayed Measurement Of Opinion Change," Journal of Abnormal and Social Psychology, 48, 327-335.
References
Lafferty, B. A. and Goldsmith, R. E. (1999), "Corporate Credibility's Role In Consumers' Attitudes And Purchase Intentions When A High Versus A Low CredibilityEndorser Is Used In TheAd," Journal of Business Research, 44, (February), 109-116.
Lutz, R. J. (1985), "Affective And Cognitive Antecedents Of Attitude Toward TheAd: A Conceptual Framework." In L. F. Alwitt and A. A. Mitchell (eds) Psychological Processes and Advertising Effects.- Theory, Research and Application. Hillsdale, NJ: Lawrence Erlbaurn Associates, 45-63.
Lutz, R. J., MacKenzie, S. B., and Belch, G. E. (1983), "Attitude Toward TheAd As A Mediator Of Advertising Effectiveness: Determinants And Consequences," Advances in Consumer Research, R. P. Bagozzi and A. M. Tybout (eds) Ann Arbor, MI: Association for Consumer Research, 532-539.
MacKenzie, S. B. and Lutz, R. J. (1989), "An Empirical Examination of theStructural Antecedents of Attitude-Toward-The-Ad in an Advertising Pretesting Context," Journal of Marketing, 53 (April), 48-65.
References
MacKenzie, S. B., Lutz, R. J. and Belch, G. E. (1986), "TheRole of Attitude Toward TheAd as a Mediator of Advertising Effectiveness: A Test of Competing Explanations," Journal of Marketing Research, 23 (May), 130-143.
Miller, G. and Baseheart, J. (1969), "Source Trustworthiness, Opinionated Statements, and Response to Persuasive Communication," Speech Monographs, 36, 1-7.
Mitchell, Andrew A. (1986), "TheEffect of Verbal and Visual Components of Advertisements on Brand Attitudes and Attitude Toward theAdvertisement," Journal of Consumer Research, 13 (June), 12-25.
Moore, D. L., Hausknecht, D. and Thamodaran, K. (1988), "Time Compression, Response Opportunity, and Persuasion," Journal of Consumer Research, 13 (June),12-24.
References
Newell, S. J. (1993), Developing A Measurement Scale And A Theoretical ModelDefining Corporate CredibilityAnd Determining Its Role As An Antecedent Of Consumers' Attitude Toward TheAdvertisement. Doctoral Dissertation, Florida State University.
Newell, S. J. and R. E. Goldsmith (2001), "TheDevelopment Of A Scale To Measure Perceived Corporate Credibility," Journal of Business Research, 52 (3), 235-247.
Ohanian, R. (1990), "Construction And Validation OfA Scale To Measure Celebrity Endorsers' Perceived Expertise, Trustworthiness, and Attractiveness," Journal of Advertising, 19 (3), 39-52.
Petty, R. E. and Cacioppo, J. T. (1983), "Central And Peripheral Routes To Persuasion: Application To Advertising." In L. Percy and A. G. Woodside, (eds) Advertising and Consumer Psychology. Toronto: Lexington Books, 3-23.
Petty, R. E., Cacioppo, J. T. and Goldman, R. (1981), "Personal Involvement As A Determinant Of Argument-Based Persuasion," Journal ofPersonality and Social Psychology, 41 (November), 847-855.
Pope, N. K. L. and Voges, K. E. (1999), "Sponsorship And Image: A Replication And Extension," Journal of Marketing Communications, 5(l),17-28.
Schulman, G. and Worrall, C. (1970), "Salience Patterns, Source Credibility, And TheSleeper Effect," Public Opinion Quarterly, 34, 371-382.
References
Simonin, B. L., and Ruth, J. A. (1998), "Is A Company Known By TheCompany It Keeps? Assessing TheSpillover Effects Of Brand Alliances On Consumer Brand Attitudes?" Journal of Marketing Research, 35 (February), 30-42.
Sobol, Marion G., Gail E. Farrelly, and Jessica S. Taper (1992), Shaping theCorporate Image, New York: Quorum Books.
Stayman, Douglas M. and Rajeev Batra (1991), "Encoding and Retrieval of Ad Affect in Memory," Journal of Marketing Research, 28 (May), 232-240.
Stem, B. (1994), "A Revised Communication ModelFor Advertising: Multiple Dimensions Of TheSource, TheMessage, And TheRecipient," Journal of Advertising, 23 (June), 5-15.
Sternthal, B., Phillips, L. W. and Dholakia, R. (1978), "ThePersuasive Effect Of Source Credibility: A Situational Analysis," Public Opinion Quarterly, 42 (Fall), 285-314.
References
Sternthal, B., Dholakia, R. and Leavitt, C. (1978), "ThePersuasive Effects Of Source Credibility: Tests Of Cognitive Response," Journal of Consumer Research, 4 (March), 252-260.
Stuart, H. and Kerr, G. (1999), "Marketing Communication And Corporate Identity: Are They Integrated?" Journal of Marketing Communications, 5(4), 169179.
Tripp, C., Jensen, T. D., and Carlson, L. (1994), "TheEffects Of Multiple Product Endorsements By Celebrities On Consumers' Attitudes And Intentions," Journal of Consumer Research, 20 (March), 535-547.
Warren, 1. (1969), "TheEffect Of CredibilityIn Sources Of Testimony On Audience Attitudes Toward Speaker And Message," Speech Monographs, (36) 456-458.
Wu, C. and Shaffer, D. (1987), "Susceptibility To Persuasive Appeals As A Function Of Source CredibilityAnd Prior Experience With TheAttitude Object," Journal of Personality and Social Psychology, 52 (April), 677-688.
Yi, Y. (1990), "Cognitive And Affective Priming Effects In TheContext For Print Advertisements," Journal of Advertising, 19 (2), 40-48.
References
Yoon, Kak, Russell N. Laczniak, Darrell D. Muehling, and Bonnie B. Reece (1995), "A Revised Modelof Advertising Processing: Extending the DualMediation Hypothesis," Journal of Current Issues and Research in Advertising, 17 (Fall), 53-67.
AuthorAffiliation
Barbara A. Lafferty TheUniversity of South Florida
AuthorAffiliation
Ronald E. Goldsmith Florida State University
AuthorAffiliation
Stephen J. Newell Western Michigan University
AuthorAffiliation
AUTHOR BIOGRAPHY
AuthorAffiliation
Barbara A. Lafferty (Ph.D., Florida State University) is an assistant professor of marketing in theCollege of Business Administration at the University of South Florida. Her research interests are in theareas of source credibility, corporate reputation, and cause-brand alliances. She has published articles in such journals as the Journal of Advertising, Journal of Business Research, European Journal of Marketing, Corporate Reputation Review, and the Journal of Global Marketing. Her work also appears in numerous conference proceedings.
AuthorAffiliation
AUTHOR BIOGRAPHY
AuthorAffiliation
Ronald E. Goldsmith (Ph.D., University of Alabama in Tuscaloosa) is currently theRichard Baker professor of marketing at the Florida State University. His articles have appeared in theJournal of the Academy of Marketing Science, the Journal of Advertising, the Journal of Business Research, the Service Industries Journal, the Journal of Retailing, Marketing Intelligence and Planning, the European Journal of Marketing, The Journal of Marketing Theory and Practice and others. His research interests include developing measures of consumer behavior constructs, diffusion of innovations, and personality in consumer behavior.
AuthorAffiliation
AUTHOR BIOGRAPHY
AuthorAffiliation
Stephen J. Newell (Ph.D., Florida State University) is an associate professor of marketing in theHaworth College of Business at Western Michigan University. His research interests include corporate credibility, advertisement recall and buyers' perceptions of sales people. His articles have been published a number of journals including theJournal of Advertising, Journal of Business Research, Journal of Marketing Theory and Practice, Journal of Marketing Education, and theJournal of Consumer Affairs. He has also published in various conference proceedings.
Copyright Association of Marketing Theory and Practice Summer 2002

study expand

2012-01-05T14:18:00.000+07:00

A study expands and integrates prior price-perceived value models within the context of price comparison advertising. More specifically, the conceptual model explicates the effects of advertised selling and reference prices on buyer's internal reference prices, perceptions of quality, acquisition value, transaction value, and purchase and search intentions. Two experiments are used to test the hypothesis. Results indicate that the experiments support the hypothesis that buyer's internal reference prices are influenced by both advertised selling and reference prices as well as the buyer's perception of the product's quality. In addition, the effect of advertised selling price on buyers' acquisition value was mediated by their perceptions on transaction value, and the effects of perceived transaction value on buyers' behavioral interns were mediated by their acquisition value perceptions.

Headnote

Theauthors expand and integrate prior price-perceived value models within thecontext of price comparison advertising. More specifically, theconceptual model explicates the effects ofadvertised selling and reference prices on buyers' internal reference prices, perceptions ofquality, acquisition value, transaction value, and purchase and search intentions. Two experimental studies test theconceptual model. The resultsacross these two studies, both individually and combined, support thehypothesis that buyers' internal reference pricesare influenced by both advertised selling and reference pricesas well as the buyers' perception of theproduct's quality. Theauthors also find that the effect ofadvertised selling price on buyers' acquisition value was mediated by their perceptions oftransaction value. In addition, the effects ofperceived transaction value on buyers' behavioral intentions were mediated by their acquisition value perceptions. Theauthors suggest directions for further research and implications for managers.

To compete successfully in a value-conscious environment, sellers must stress thevalue oftheir offerings. One value-based strategy involves emphasizing thevalue ofacquiring theproduct (i.e., acquisition value) (Monroeand Chapman 1987). Sellers can increase acquisition value perceptions by enhancing buyers' perceptions of theproduct's quality or benefits relative to theselling price(Bolton and Drew 1991; Dodds, Monroe, and Grewal1991; Monroeand Krishnan1985; Zeithaml 1988). Thus, firms might opt for one ofthree value-based positioning strategies-high quality, low price, or some balance ofquality to price.

Sellers also can compare a lower selling priceto a higher advertised reference price(e.g., was $200, now $150) to enhance buyers' value perceptions. This value-oriented strategy is aimed at enhancing buyers' deal perceptions (or transaction value). Stressing the pricebargain the buyerwould be getting by undertaking thetransaction can effectively promote theoffering (i.e., increasing thesalience of thereduction in theselling price). Unfortunately, previous research has not examined the effects of price-comparisonor reference-price advertising on buyers' perceptions ofacquisition value, transaction value, or behavioral intentions. There is need for researchers to understand why this pricetactic (i.e., price-comparison advertising) seemingly works. From such information, effective value-oriented promotions could be developed. Furthermore, pricepromotions have become so widespread that several State's Attorneys' General offices (e.g., Maryland, New York, Colorado) have become concerned that some sellers use "value pricing" to deceive buyers(Grewaland Compeau 1992; Kaufmann, Smith, and Ortemeyer 1994).

In this article, we provide an understanding ofhow price-comparison advertisingcould influence buyers' perceptions ofvalue and establish a framework for addressing thedeception issue. Theconceptual argument suggests that advertised reference pricesin these deal-oriented advertisements can enhance buyers' internal reference prices(Lichtenstein and Bearden 1989; Urbany, Bearden, and Weilbaker 1988). These enhanced internal reference prices, when compared with thelower selling price, resultin higher transaction value perceptions. Theincrease in perceived transaction value enhances purchases and reduces search behavior for lower prices. If sellers intentionally increase theadvertised reference pricesabove normal retail prices, that is, inflate advertised reference prices, theresulting inflated perceptions oftransaction value would be deceptive. Harm to both buyersand competitors could resultfrom the effect of theinflated transaction value on buyers' search and purchase behaviors.

In addition, past research has not addressed theconceptual distinction between acquisition and transaction value, nor whether it is empirically feasible to make this distinction. Theresearch reported in this article addresses this important limitation. We develop a conceptual model that outlines how advertised pricing tactics (i.e., comparison price-advertising) affect buyers' purchase decisions. Then we present two tests of theconceptual model and discuss the resultsand their implications.

TheConceptual Model

In price-comparison advertising, a higher advertised comparison price(commonly termed advertised reference price) is compared with a lower advertised selling price. Buyers' judgments ofthese advertised pricesdepend not only on the pricesper se, but also on thecontextual cues presented within theadvertisement, situational influences surrounding buyers, and buyers' internal reference prices(Rajenderan and Tellis 1994). Theproposed model has two exogenous constructs (advertised reference priceand advertised selling price) and six endogenous constructs (buyers' perceptions ofproduct quality, their internal reference price, perceived transaction value, perceived acquisition value, willingness to buy, and search intentions) (see Figure 1). Each ofthese constructs and therelationships between them are explained subsequently.

Perceived Quality

Several past studies have examined the effects ofinformation cues, such as price, on buyers' perceptions ofquality (see reviews by Monroeand Krishnan1985; Rao and Monroe1989; Zeithaml 1988). Perceived quality is defined as a buyer's estimate ofa product's cumulative excellence (Zeithaml 1988). Thegeneral consensus ofthese studies is that priceis less likely to have a significant effect on buyers' perceptions ofquality in thepresence ofother attributes and when buyersare familiar with theproduct or product category (Rao and Monroe1988, 1989). In addition, there is evidence in thedomain ofcomparative price advertisingthat these advertised prices(both thereference priceand theselling price) do not have an effect on buyers' perceptions ofquality (see Grewal1989; Urbany and Bearden 1990).

Consequently, subjects exposed to a comparative priceoffer for a well-known brand (Dodds, Monroe, and Grewal1991), in thepresence ofseveral cues (including a picture of theproduct) (Grewal1989; Rao and Monroe1989), and having familiarity and knowledge of theproduct category (Rao and Monroe1988) are not likely to use these advertised pricesto shift their perceptions ofquality. Therefore, in theproposed model, we do not expect theadvertised selling priceand theadvertised reference priceto affect buyers' perceptions ofquality. However, we test for these paths.

Internal Reference Price

Theconcept ofinternal reference price, while operationally elusive, is an important cornerstone for behavioral priceresearch. In this research, an internal reference priceis defined as a price(or pricescale) in buyers' memories that serves as a basis for judging or comparingactual prices(Monroe1973; Monroe, Grewal, and Compeau 1991).

Della Bitta, Monroe, and McGinnis (1981) use adaptation-level theory to argue that buyers' internal reference pricesare influenced by thekey focal cues in an advertisement: theadvertised selling priceand theadvertised reference price. Furthermore, adaptation-level theory suggests that these internal reference pricesare influenced by residual cues (e.g., previously acquired information that has been assimilated to form perceptions and/or expectations of thequality ofproducts in a product category or a specific brand). Buyersforming an initial level ofperceived quality for theproduct and/or brand depend oninformation in theadvertisement and onpreviously acquired information (Herr 1989). Using this level ofperceived quality for theproduct and/or brand and theadvertised prices(sales and reference) as a basis, buyersdevelop internal reference prices(or pricescales) to be used during subsequent judgments ofvalue. Thepossibility ofa perceived quality to the pricemapping phenomenon has been illustrated by Monroe(1973). Consequently, advertised reference price, advertised selling price, and perceived product quality positively influence buyers' internal reference prices(Lichtenstein and Bearden 1989; Urbany, Bearden, and Weilbaker 1988). This conceptualization leads to thefollowing three paths in theconceptual model:

HI: There is a positive relationship between buyers' perceptions ofquality and their internal reference price.

H2: There is a positive relationship between advertised selling priceand buyers' internal reference price.

H3: There is a positive relationship between advertised reference priceand buyers' internal reference price.

Buyers' internal reference pricesadapt to thestimuli pricespresented in theadvertisement. That is, buyerseither adjust their internal reference priceor accept theadvertised reference priceto make judgments about theproduct's value and thevalue of thedeal. Our conceptual model onhow comparison price advertisinginfluences buyers' perceptions ofvalue explicitly recognizes this adaptive nature of buyers' internal reference prices.

Perceived Acquisition Value

Past acquisition value-based models (e.g., Dodds, Monroe, and Grewal1991; Zeithaml 1988) have defined this concept as theperceived net gains associated with theproducts or services acquired. That is, theperceived acquisition value of theproduct will be positively influenced by thebenefits buyersbelieve they are getting by acquiring and using theproduct and negatively influenced by themoney given up to acquire theproduct (i.e., theselling price). One important element ofthis "get" component is product quality or buyers' perceptions ofproduct quality. 1

Several researchers have conceptualized acquisition value in this manner, though they used different termssuch as "bargain value" (Keon 1980), "perceived value" (Dodds, Monroe, and Grewal1991; Lichtenstein and Bearden 1989; Monroeand Krishnan1985; Urbany, Bearden, and Weilbaker 1988), "perceived worth" (Szybillo and Jacoby 1974), "acquisition utility" (Thaler 1985), and "value consciousness" (Lichtenstein, Netemeyer, and Burton 1990; Lichtenstein, Ridgeway, and Netemeyer 1993). Therefore, defining perceived acquisition value as the buyers' net gain (or tradeoff) from acquiring theproduct or service represents "a more global and enduring kind ofvalue which takes into account both priceand quality" (Urbany and Bearden 1990, p. 4). This conceptualization leads to thefollowing paths in theconceptual model:

H4: There is a positive relationship between buyers' perceptions ofquality and their perceived acquisition value.

H5: There is a negative relationship between theadvertised selling priceand buyers' perceptions ofacquisition value.

Perceived Transaction Value

Buyersexposed to price-comparisonadvertisements and similar pricepromotions are presented with an expressed deal or bargain in terms ofa selling pricethat is explicitly reduced in magnitude. They are likely to assess themerits or value ofsuch a deal by comparing theselling priceto their internal reference prices(Monroeand Chapman 1987; Thaler 1985). For example, in a recent interpretive study of buyers' reactions to price-deals, one shopper indicated thefollowing: "Sometimes if I get a good deal at thediscount rack, I feel good about that and I'll stroll through theother parts [of themall or store] and not feel guilty if I buy more expensive, originally priced items" (Grewaland Compeau 993, p. 11).

Therefore, a buyer, onexamining thefinancial terms of the priceoffer, might perceive additional value beyond that provided by acquisition value. Thus, perceived transaction value is theperception ofpsychological satisfaction or pleasure obtained from taking advantage of thefinancial terms of the pricedeal (Lichtenstein, Netemeyer, and Burton 1990; Monroeand Chapman 1987; Thaler 1985; Urbany and Bearden 1989). This conceptualization leads to thefollowing paths in theconceptual model:

H6: There is a positive relationship between buyers' internal reference priceand their perceived transaction value.

H7: There is a negative relationship between theadvertised selling priceand buyers' perceptions oftransaction value.

Effects ofAcquisition and Transaction Value onWillingness to Buy and Search Intentions

Willingness to buy is defined as thelikelihood that the buyerintends to purchase theproduct (Dodds, Monroe, and Grewal1991). All things being equal, willingness to buy is positively related to overall perceptions ofacquisition and transaction value (Della Bitta, Monroe, and McGinnis 1981; Monroeand Chapman 1987; Urbany and Dickson 1990; Zeithaml 1988). Buyers' willingness to buy is positively linked to their perceptions ofacquisition and transaction value:

H8: There is a positive relationship between buyers' perceptions ofacquisition value and their willingness to buy. H9: There is a positive relationship between buyers' perceptions oftransaction value and their willingness to buy. Search intention is defined as a buyer's willingness to search for additional priceinformation. Stigler (1961) suggests that because ofvariations in pricein themarketplace buyersgenerally are uncertain what thelowest available priceis. To reduce this uncertainty, buyersmust seek information from sellers. Willingness to search for priceinformation is contingent on buyers' trading off theperceived benefits (e.g., money saved) relative to thecosts of thesearch (e.g., time, money, effort spent in conducting thesearch) (Marmorstein, Grewal, and Fishe 1992). Previous research shows that when buyersare exposed to an advertised regular pricecoupled with a lower sale price, their willingness to conduct additional search declines because ofan increase in their perceptions ofvalue (Della Bitta, Monroe, and McGinnis 1981; Urbany, Bearden, and Weilbaker 1988). Therefore, buyers' intentions to search is linked negatively to their perceptions ofacquisition and transaction value:

H10: There is a negative relationship between buyers' perceptions ofacquisition value and their intentions to search.

H11: There is a negative relationship between buyers' perceptions oftransaction value and their intentions to search.

An Alternative Model

A key issuethat must be addressed is whether perceived transaction value and perceived acquisition value are interrelated. Similar to Thaler's (1985) conceptualization and Monroeand Chapman's (1987) model, theproposed model (Figure 1) assumes that buyers' perceptions ofacquisition and transaction value are independent ofeach other. However, there are several potential reasons that suggest these two value dimensions are not independent ofeach other. Conceptually and operationally, theinterrelationship between these two value dimensions has not been addressed by previous research (cf. Urbany and Bearden 1990).

It is reasonable to propose that a pricepromotion that leads to positive perceived transaction value (i.e., greater psychological pleasure associated with obtaining favorable financial terms) would in turn influence buyers' perceptions of thevalue ofacquiring theproduct or receiving theservice (i.e., greater net gain by reducing thefinancial outlay). It is proposed that positive perceived transaction value enhances buyers' evaluations of thevalue ofacquiring theproduct. Buyers' perceptions oftransaction value are situation specific, and though their assessments ofacquisition value are more holistic evaluations of theproduct's value, it is likely that their transaction value influences their perceptions ofacquisition value and not vice versa.

Thetwo constructs have an overlapping antecedent construct, theadvertised selling price. Drawing onequity notions from satisfaction research (see Bolton and Drew 1991; Oliver and Swan 1989), transaction value could be considered akin to thefairness construct (i.e., theequity/pleasure associated with getting a fair price) (see also Huppertz, Arenson, and Evans 1978), and acquisition value similar to theoverall satisfaction construct (Grewal1995). Research in this post-purchase domain suggests that buyers' assessment ofequity affects their overall evaluations. Theparallel in thepre-purchase domain is that perceived transaction value affects perceived acquisition value.2 This conceptualization leads to thefollowing additional path: H12: There is a positive relationship between buyers' perceptions oftransaction value and their perceived acquisition value.

Hl2 suggests that sellers also might influence buyers' evaluations of thevalue of theproduct (i.e., acquisition value) indirectly through the effects ofcomparative advertised reference prices on buyers' perceived transaction value. This suggestion also leads to thepossibility that the effect ofadvertised selling price on buyers' acquisition value might be mediated by their perceptions oftransaction value. It also leads to thepossibility that the effect ofperceived transaction value onbehavioral intentions could be mediated by perceived acquisition value.

On thebasis ofthese proposed revisions, an alternative model (Figure 2) also is tested. Thealternative model hypothesizes a link between perceived transaction value and perceived acquisition value. The effects ofadvertised selling price onacquisition value are expected to be mediated by transaction value (i.e., no significant direct effect ofadvertised selling price onacquisition value). Therevised model also hypothesizes that the effects ofperceived transaction value onbehavioral intentions are mediated by perceived acquisition value (i.e., no significant direct effects ofperceived transaction value onpurchase and search intentions). In addition, theno-effectpaths pertaining to advertised selling priceand advertised reference price on buyers' perceptions ofquality have been dropped.

Research Method Research Plan

These hypotheses (and models) were tested using causal modeling. Two studies were conducted. Both studies used a 2 x 2 between-subjects experimental design, that is, two selling pricelevels ($249.95 and $349.95) and two advertised reference pricelevels ($400 and $500).3 In both studies, thesubjects were shown a booklet containing an advertisement for a bicycle and a questionnaire. Theadvertisement used a known brand name (Raleigh USA). Thesubjects then responded to a set ofquestions that assessed thevarious latent constructs.

In Study 1, a laboratory experiment, thesubjects were 361 undergraduate students at a western state university. Themean age ofrespondents in Study I was 23 years (range 18-49), 52.9% of therespondents were male, and 76.1% owned a bicycle. In Study 2, an experimental survey assessing thereplicability and theboundary conditions ofStudy l's results, thesubjects were staff employees at thesame university. Of the600 employees surveyed, 328 responded, a 54.6% response rate. To motivate a response, $125 in prizes were awarded by lottery. Themean age ofrespondents in Study 2 was 41 years (range 24-62), median family income was $40,000-$50,000, 37.6% of therespondents were male, and 71.6% owned a bicycle.

Pretests

A series ofpretests were conducted. Thefirst provided information about thesubjects' knowledge, involvement, and acceptable pricerange for bicycles. The resultsindicate that a bicycle was a personally relevant product and thus was selected as thetest product. Because Rao and Monroe(1988) found that product knowledge affected buyers' subjective product evaluations, another objective ofthis pretest was to determine a product about which subjects were knowledgeable. The resultsindicate that subjects exhibited high levels ofproduct familiarity with this specific product (i.e., a Raleigh USA bicycle). The resultsand a market survey of pricessuggest that $400 was around theaverage market priceand $500 was an above-average market price(i.e., an inflated advertised reference price) for this bicycle. Three additional pretests were used to develop thescales for thevarious latent constructs, particularly acquisition value and transaction value.

Measures

Thescales used to measure thelatent constructs are provided in Table 1. Theconstructs of buyers' perceptions ofquality, acquisition value, transaction value, willingness to buy, and search intention were assessed using seven-point category rating scales. Buyers' internal reference priceswere dollar estimates provided by thesubjects.

Perceived quality. Buyers' quality perceptions were measured using three Likert statements (Dodds, Monroe, and Grewal1991; Rao and Monroe1988) that assessed theproduct's quality, durability, and reliability.

Internal reference price. Buyers' internal reference pricewas assessed using two common measures: average market priceestimate and fair priceestimate. These two items were based onscales developed by Lichtenstein and Bearden ( 1989) and Urbany, Bearden, and Weilbaker (1988).

Perceived acquisition value. Buyers' acquisition value was measured using nine Likert statements that built onpast scales ofperceived value (e.g., Chapman and Monroe1990; Dodds, Monroe, and Grewal1991). However, theproposed measure ofperceived acquisition value is more comprehensive than thethree- or four- item scales previously used that focused on"good value for themoney." We explicitly attempted to capture thetrade-off between a product's benefits and thecost ofits acquisition. For example, sample items included thefollowing: "I feel that acquiring this bicycle meets both my high-quality and low-pricerequirements;" "I would value this bicycle as it would meet my needs for a reasonable price;" and "This bicycle would be a worthwhile acquisition because it would help me exercise at a reasonable price."

Perceived transaction value. Past research has had considerable problems measuring buyers' perceptions oftransaction value and developing a scale that discriminates adequately from perceived acquisition value. On thebasis ofour pretests and research by Lichtenstein, Netemeyer, and Burton ( 1990), we measured perceived transaction value using three Likert statements. These statements seem to capture theessence oftransaction value-thepleasure buyersget from finding and taking advantage ofa pricedeal (e.g., "taking advantage ofa pricedeal like this makes me feel good"). Principal component analysis of theacquisition and transaction value scales demonstrates that thetwo scales discriminate in both studies (see Table 2).

Willingness to buy. A three-item scale, based onDodds, Monroe, and Grewal's (1991) study, measured buyers' willingness to buy. Thespecific items were anchored from "very low" to "very high."

Search intentions. Buyers' intentions to search for additional information (e.g., visit other stores to check their prices) were measured using three Likert statements that were based onprior research by Della Bitta, Monroe, and McGinnis (1981).

Analysis and Results

Plan for Data Analysis

Thedata from thetwo studies were analyzed in two stages. Themeasurement model was assessed to confirm that thescales were unidimensional and reliable. When thereliability of themeasures had been established, thestructural model was tested using LISREL-VII causal modeling procedures (Joreskog and Sorbom 1989). This testing determined thestrength ofindividual relationships, themodel's goodness offit, and thevarious hypothesized paths. Thetwo-step procedure followed here reduces thenumber ofinterpretational confounds. PRELIS was used to generate theinput matrix.

Measurement Properties of theScales

Thescales used to measure thelatent constructs in themodel are provided in Table 1. Item reliability, variance extracted, and construct reliability also are shown. Theassessment of themeasurement properties ofall six scales indicated that thefactor loadings (lambdas) were high and significant (p < .001), which satisfies thecriteria for convergent validity. Anderson (1987) suggests thefollowing criterion for assessing discriminant validity between scales: Thecorrelation between two latent constructs plus or minus two standard errors does not include one. All six scales met this criterion in both studies. Furthermore, Fornell and Larcker (1981) suggest that discriminant validity can be assessed by determining whether thevariance extracted estimates for two constructs are greater than thesquare of theparameter estimate between them . Thesix measured constructs met this criterion in both studies. We also assessed thediscriminant validity of thescales using confirmatory factor analysis procedures (Anderson and Gerbing 1988). The results ofeach pairwise construct comparisonsuggest that thetwo factor solution was better than thesingle factor solution (see theAppendix).

Fornell and Larcker (1981) also stress theimportance ofexamining composite reliability and variance extracted. Bagozzi and Yi (1988) suggest two criteria: Composite reliability should be greater than or equal to .60, and variance extracted should be greater than or equal to .50. For both studies, all six composite reliabilities were greater than .75, and all six variances extracted were greater than .55 (see Table l). Finally, we analyzed thestructural model using summated scales and obtained similar results, which suggests that themeasurement-structure interaction was minimal.

Model Fit

Thecausal models in Figures I and 2 were assessed using a full-information method. Thecausal models were specified as shown in thefigures. ThePHI, PSI, TD, and TE matrices were diagonal and free. Thetheta-deltas associated with advertised selling priceand advertised reference pricewere fixed. Errors were treated independently to avoid interpretational confounds. Thelambda matrices (both X and Y) were full and fixed. Then theindividual items associated with theexogeneous and endogeneous constructs were freed. However, one of thelambdas for each construct was set to 1.0 to properly define themeasurement (see Joreskog and Sorbom 1989).

Theoverall fit of thestructural model was determined initially by examining theZ2 statistics for each study, which were significant. A significant 2 statistic could indicate an inadequate fit, but this statistic is sensitive to sample size and model complexity; therefore, rejection ofa model on thebasis ofthis evidence alone is inappropriate (Bagozzi and Yi 1988; Bearden, Sharma, and Teel 1982; Marsh, Balla, and McDonald 1988). Accordingly, other measures offit compensating for sample size also were applied: Bentler and Bonett's (1980) normed fit index (A), Tucker and Lewis's (1973) non-normed fit index (p), and Bentler's (1990) comparative fit index (CFI). Each ofthese indices showed an adequate fit: A was .88 (Study 1) and .89 (Study 2); p was .91 (Study 1) and .90 (Study 2); CFI was .93 and .91 for Study I and 2, respectively (see Table 3).

Hypotheses Tests

Thestandardized estimates for thevarious model paths and theassociated t-values for thetwo studies are provided in Table 3. Thestructural path estimates for themodel furnished in Table 3 should be read with thecaveat that our data are causal only for the effects ofadvertised reference priceand advertised sale price. For all other relationships, our data are correlational, and thecausal direction is based onprior theory.

Advertised pricesand perceived quality. As expected, advertised selling priceand advertised reference pricedid not affect buyers' perceptions ofquality significantly.

Influences oninternal reference price. These resultssupport H1, H2, and H3, which indicates that buyers' internal reference pricesare functions ofperceived quality, advertised selling price, and advertised reference price. Previously, Urbany, Bearden, and Weilbaker ( 1988) demonstrated thecapability ofadvertised reference pricesto serve as anchors and to shift internal reference pricesin their direction. Finally, Hyun (1993), using Korean subjects, demonstrated a positive relationship between advertised selling pricesand subjects' internal reference prices. Thus, our resultscomplement past research and are consistent with predictions based onadaptation-level theory.

Influences onperceived acquisition value. In both studies, perceived acquisition value is a positive function ofsubjects' perceptions ofquality (H4 is supported). This relationship has previous empirical support (Dodds, Monroe, and Grewal1991; Hyun 1993). Therefore, we provide empirical support in a price comparisoncontext for theproposition that perceived acquisition value is influenced, in part, by buyers' perceptions ofquality. (Substantively, as mentioned previously, theliterature stresses that perceived quality is an important part of the"value equation.")

Thehypothesized influence of theadvertised selling price onperceptions ofacquisition value was supported in both studies (H5 is supported). Theoverall available evidence presented here supports thetheoretical proposition and thelay belief that selling priceis a negative element of buyers' perceptions ofacquisition value.

Influences onperceived transaction value. Themodel in Figure 1 suggests that perceived transaction value is a function of buyers' internal reference pricesand theactual selling price. As is shown in Table 3, these relationships were supported by both studies (H6 and H7 are supported). There is a significant negative relationship between theactual selling priceand subjects' perceptions oftransaction value. Moreover, there is a positive relationship between subjects' internal reference pricesand their perceptions oftransaction value. Thus, thetheoretical arguments for theinfluence ofselling priceand buyers' internal reference prices on buyers' perceptions oftransaction value (i.e., theperceived merits of theoffer) have empirical support.

Value perceptions and behavioral intentions. Della Bitta, Monroe, and McGinnis ( 1981 ) observe that buyers' positive perceptions ofvalue were a necessary but insufficient condition to induce willingness to buy. Other research has found a positive relationship between perceptions ofvalue and willingness to buy (Dodds, Monroe, and Grewal1991; Hyun 1993). However, previous research efforts measured themore global construct-perceived value. In thecurrent empirical effort, we decompose perceived value into two theoretical components: perceived acquisition value and perceived transaction value.

The resultsshown in Table 3 indicate a significant positive relationship between perceived acquisition value and willingness to buy (HK is supported). Thedirect relationship between perceived transaction value and willingness to buy, though positive, is weak overall and not statistically significant in thesecond study (partial support for Hg). Similarly, thenegative relationship between perceived acquisition value and search intentions is significant in both studies (Hio is supported). Therelationship between perceived transaction value and search intentions was not supported (no support for HI,).

Test of theAlternative Model, H12, and Mediating

Hypotheses

Alternative model. A key theoretical argument of thealternative model presented in Figure 2 is that perceived transaction value has a positive influence onperceived acquisition value. Theempirical relationships provided in Table 4 strongly support this relationship in both studies (H12 is supported). Previous efforts to decompose perceived value into its theoretical acquisition value and transaction value components have had measurement flaws. As is demonstrated in Tables 1 and 2, theresearch reported here has overcome theinherent measurement difficulties presented by these two concepts.

The results of thealternative model also are presented in Table 4. Themodel hypotheses are supported in both studies. Furthermore, therevised model (with four fewer paths) fits thedata as well as thecomplete model (e.g., theCFI statistic was thesame for both models). These resultsfurther validate our propositions that the effect ofselling price onperceived acquisition value is mediated by perceived transaction value and that the effects ofperceived transaction value onpurchase and search intentions are mediated by

Discussion

Conceptual Developments

As has been noted, price comparison advertisingis a widely used pricepromotion tactic. Although research investigating issues on therelative effectiveness ofthis tactic spans nearly 20 years, we are still trying to understand how and why it works. Drawing onprior research (e.g., Monroeand Chapman 1987; Thaler 1985), we provide a theoretical argument for why such pricepromotional tactics (and other similar forms) influence buyerbehavior.

Previous empirical efforts to decompose theconcept ofoverall perceived value into thetwo independent constructs ofperceived acquisition value and perceived transaction value encountered measurement problems (e.g., Chapman and Monroe1990; Grewal1989). Examining prior research efforts (e.g., Chapman and Monroe1990; Grewal1989) and comments onThaler's original conceptualization (Bearden et al. 1992) led to therevised model in Figure 2 and to thestronger measurement model presented here.

Prior research (i.e., Monroeand Chapman 1987; Thaler 1985) typically modeled these two value dimensions as independent ofeach other. There is thepossibility that one of thetwo value components is actually an antecedent of theother. Because, in theabsence ofa pricepromotion, thebasic perceived value model (see Dodds, Monroe, and Grewal1991; Zeithaml 1988) postulates that buyers' perceptions ofvalue are formed from a mental trade-off between perceived quality (or benefits) and price, it seems logical that buyerswould perceive a pricepromotion as enhancing theoverall value ofan acquisition. Thus, themodel in Figure 2 shows theconceptual adjustments, which suggests that perceived acquisition value is a function ofperceived quality and perceived transaction value (i.e., assessment of the priceoffer). Theadvertised selling priceaffects perceived acquisition value by its effect onperceived transaction value. As simple as this conceptualization seems, it represents an important addition to theresearch literature on therelationship between priceand buyers' perceptions ofquality and value.

Although this research emphasizes exploration of therelative effects of price comparison advertising on buyers' perceptions ofvalue, important insights also have been found about theinternal reference priceconcept. It has been conceptually argued and empirically confirmed that buyers' internal reference priceis influenced by both theseller's advertised (higher) reference priceand theadvertised (lower) selling price. Previous research demonstrates theinfluence of theadvertised reference price on buyers' internal reference prices(Urbany, Bearden, and Weilbaker 1988). As conceptualized by adaptation-level theory, this research also demonstrates that both pricespresented in a price comparisonadvertisement influence buyers' internal reference prices. Moreover, it has been shown that buyers' internal reference pricesare influenced by their relative assessments ofproduct quality. This finding is consistent with thetenets ofcognitive reference points and categorization theory (Herr 1989; Monroe, Grewal, and Compeau 1991).

Theconceptual argument by Thaler (1985) and Monroeand Chapman (1987) that perceived transaction value is a function of theselling priceand buyers' internal reference pricehas been confirmed empirically. Although Urbany and Bearden (1989) show thepositive relationship between buyers' internal reference priceand their perceptions oftransaction value, this is thefirst published research effort to demonstrate theapparent implied mental comparisonbetween theadvertised selling priceand buyers' internal reference prices.

Also ofconcern is whether perceived transaction value has a direct or indirect influence onwillingness to buy or intentions to search. Previous research offers convincing evidence on thedirect relationship between perceived (acquisition) value and measures ofbehavioral intentions. Our empirical evidence supports theidea that perceived transaction value influences willingness to buy and intentions to search through its effect onperceived acquisition value. This particular finding strengthens theargument that, in comparative price advertisingpromotions, perceived transaction value enhances buyers' perceptions ofacquisition value, and that these two components ofperceived value are not independent constructs.

Measure Development

Operationally, this research develops and validates separate measures ofperceived acquisition value and perceived transaction value (Table 1). One aspect ofthis measure development was to understand thepsychological pleasure that buyersmight experience when buying a product ondeal (i.e., obtaining a bargain). For further discussion onthis point, see Schindler (1989). Another contribution ofthis research is thedevelopment ofa unidimensional multi-item scale to measure buyers' internal reference price. Past research predominantly used single items that tapped into various aspects of theinternal reference pricescale (or continuum) (see Monroe, Grewal, and Compeau 1991). In addition, measures for perceived product quality and purchase and search intentions were further refined and validated (see Table 1).

Managerial Implications

A key managerial implication ofthis research is thedemonstration ofhow advertised prices(both reference and sale) provided in price comparisonadvertisements affect buyers' internal reference prices(i.e., higher advertised priceslead to higher internal reference prices). In turn, these internal reference pricesare linked to buyers' perceptions ofvalue and behavioral intentions. Focusing ononly thefinal price(or sale price) to theexclusion of thecontextual advertised (or display) reference priceby advertisers and retailers might be a strategic mistake. This issuecould be one of thereasons why the"everyday low price" strategy used by several retailers has not been successful.

The results ofour studies, in conjunction with past research (e.g., Lichtenstein and Bearden 1989; Urbany, Bearden, and Weilbaker 1988), suggest that inflated advertised reference priceshave thepotential to be deceptive (in our studies, the$500 advertised reference pricewas considerably higher than theaverage market price). These higher advertised reference pricesenhance buyers' perceptions oftransaction value, acquisition value, and purchase intentions and reduce buyerslikelihood ofsearching for a lower price. Furthermore, thepotential for deception is likely to be more pronounced for buyerswho have less priceand/or product knowledge. Theconcerns voiced by several State's Attorneys' General offices that value pricing could be used to deceive consumers seem to have merit. Consequently, State's Attorneys' General offices and theFederal Trade Commission must monitor such practices. In cases in which retailers and manufacturers use fictitious or inflated advertised reference pricesin their advertisements, appropriate action must be taken. Appropriate action by such agencies might include cease and desist orders, fines, and posted notices ofhow theretailer established theadvertised reference price.

Thecurrent research findings support thenotion that product quality perceptions enhance acquisition value and willingness to buy. Furthermore, past research shows that a high-quality position is important in developing brand equity and leads to higher market share and profitability in thelong run (Curry 1985; Jacobson and Aaker 1987; Phillips, Chang, and Buzzell 1983). Our results(crosssectional study) in conjunction with those based onPIMS databases (i.e., longitudinal) suggest that developing and maintaining a high-quality position is important for shortterm adoption and long-termdevelopment ofmarket share.

Thelack ofassociation between priceand quality perceptions in our two studies supports past research findings that a high-quality position is not necessarily incompatible with a low cost (or price) position (Phillips, Chang, and Buzzell 1983). Many manufacturers try to maximize value to buyersby offering above-average quality at reasonable prices(Curry 1985). This positioning can be achieved through well-designed pricepromotions that emphasize thefairness or reasonableness oftheir selling pricesand thereby enhance buyers' perceptions oftransaction value. Our findings suggest that buyers' perceptions oftransaction value enhances willingness to buy through their perceptions ofacquisition value.

unprecedented ease and accuracy. Managers must understand thevariables affecting theacquisition value of theproduct. They also must understand where their product fits ona continuum ranging from satisfying unique needs (e.g., CAT scanner) to satisfying undifferentiated needs (e.g., corn syrup) (Dolan 1995; Nagle and Holden 1995). Thus, manufacturers can position products that are unique using an acquisition value-enhancing strategy and those that are relatively undifferentiated from competitors using a transaction value-enhancing strategy.

Thestudy resultsalso suggest that thevarious value strategies (i.e., deal value versus product value) are important predictors ofbehaviors. In addition, past research suggests that thecost ofserving buyersand theeffectiveness ofvalue strategies might vary across segments (Lichtenstein, Netemeyer, and Burton 1990; Shapiro et al. 1987). That is, some segments are sensitive toward price, whereas others are more benefit oriented (in our studies, therelative effect ofacquisition value versus transaction value onwillingness to buy was greater for thenonstudent sample). Therefore, value perceived by buyerswill vary across segments. For some buyers, acquisition value might be more important than transaction value or vice versa. Managers should determine which value strategy is appropriate for their target segments and develop their positioning strategies appropriately.

Limitations and Avenues for Further Research

Further research should explore this study's limitations. For example, subjects were exposed to one type ofsemantic cue, theoriginal priceand selling pricecombination. Research using other product and semantic cues, such as "compare at, selling price," "MSLP, selling price," and "total value, selling price" would be worthwhile (see Grewal, Marmorstein, and Sharma 1996).

Another limitation ofthis research was Thaler's ( 1985) suggestion that buyers' acquisition utility (or value) should be equivalent to the comparison oftheir reservation pricewith theactual selling price. This research, following Monroeand Chapman's (1987) conceptualization, considered perceived acquisition value as a comparisonbetween buyers' perceptions ofquality and selling price. Although theapproach used here is consistent with theextant marketing practitioner beliefs, therelative role ofreservation price(or maximum acceptable price) should be explored as an alternative way to measure theget component of buyers' perceptions ofacquisition value. In addition, research must explore whether other factors such as usage flexibility, usage convenience, and need congruence affect buyers' perceptions ofacquisition value.

A related issuepertains to thefunctional form ofperceived acquisition value. Monroe( 1990) suggests a ratio (or proportional) model, but points out that this is only one means ofillustrating the comparison. Associated research on price-quality and price-warranty trade-offs supports a subtractive model (e.g., Levin and Johnson 1984; White and Truly 1989). Our resultsalso support a subtractive model.6 However, White and Truly (1989), using some important methodological variations, also show that some oftheir subjects apparently followed a proportional model ofinformation integration. Therefore, an important research issueis whether theinformation integration implied by theformation ofperceived acquisition value is represented best as a subtractive, ratio, or averaging model or as some other functional form. Finally, there is a need for research that involves examining whether thedegree ofbelievability of theadvertised reference price, believability of theselling price, believeabilty of theoverall priceoffer, and priceconsciousness influence buyers' perceptions oftransaction value.

Another issuethat warrants additional research addresses thedistinction between acquisition value and overall value. Overall perceived value can be conceptualized with many distinct components (Forbes and Mehta 1978), which could include thevalue of theacquisition, thevalue associated with thestart-up (e.g., a cellular phone with a recharger similar to an existing cellular phone will have greater value), and thevalue associated with reselling theproduct (e.g., some cars have greater resale value). Thus, research in business-to-business settings might need to distinguish, conceptually and operationally, acquisition value from overall value.

Research also must address how buyersform their internal reference prices. In this research, respondents' internal reference priceswere operationalized through point estimates oftheir expected normal, average, and fair prices. However, Klein and Oglethorpe (1987) suggest that internal reference pricescould be operationalized in a variety ofother manners, including by expected prices, pricelast paid, or aspiration price. Further-more, internal reference pricesalso may be operationalized as a pricerange (Monroe, Grewal, and Compeau 1991; Urbany and Dickson 1990). Therefore, further research must address whether these different bases for, and ways ofmeasuring, internal reference pricesyield similar results.

Although this study examines theeffectiveness of theframework in thecontext of price-comparisonadvertisements, further research is needed to test its effectiveness in explaining buyers' behavioral reactions to other pricepromotions (e.g., coupons, rebates) as well as pricechanges. Thegeneralizability of themodel should be examined by assessing thefit of themodel for different samples of buyers. Value-conscious segments, deal-prone segments, and segments that do or do not believe thedifference between theadvertised selling priceand theadvertised reference pricemight vary. Furthermore, thepredispositions of thesubjects could affect their perceptions. It would be useful to look at theindividual characteristics ofsubjects more closely and assess such variables as involvement, priceconsciousness, knowledge, and inclination to take risks.

Footnote

1 Thaler (1985) operationally defines this get component (or value equivalent) as theamount ofmoney that would leave theperson indifferent about receiving themoney or theproduct as a gift. In economic theory, thevalue equivalent is similar to thereservation price(themaximum price the buyeris willing to pay), and therefore acquisition value could be considered comparable to consumer surplus (Monroeand Chapman 1987; Thaler 1985). However, operationally defining thevalue equivalent (get component) simply as a reservation priceis a limited view, because it does not include the buyers' quality evaluation, except by indirect inference. Moreover, empirical evidence verifying that buyersuse such reservation priceswhen assessing thevalue ofa product is not available. For example, Bearden and collegues (1992) find no significant relationship between three price-estimate measures of thesubjects' reservation priceand their perceptions ofacquisition value or their willingness to buy.

Footnote

We thank an anonymous reviewer for this suggestion.

Footnote

3. We do not predict an interaction termbetween advertised selling priceand advertised reference price, in line with previous models, such as Monroeand Chapman's (1987). However, we did test for theinteraction term. MANOVA analyses onboth data sets indicated that theinteractions were not significant in either data set (p > .05). Furthermore, using LISREL procedures and modeling an interaction term, we did not find a significant effect of theinteraction oninternal reference price, acquisition value, and transaction value. Furthermore, the resultswith an interaction termsuggest a worse fit.

Footnote

4. 0ne of thepretests' (n = 400) resultsindicated that scales used by past research (e.g., Chapman and Monroe1990) to assess overall perceived value did not adequately discriminate acquisition value from transaction value. The resultssuggest a two-factor solution. That is, acquisition value and transaction value load onseparate factors, but overall value loads onboth factors. Thus, this research explicitly focuses onacquisition and transaction value and not onoverall value.

Footnote

5. Three conditions must be met to establish mediation : ( I ) theindependent variable affects themediator : (2) theindependent variable affects thedependent variable : and (3) when both theindependent variable and themediator are regressed on thedependent variable, themediator is significant, whereas the effect of theindependent variable is reduced. ANOVA and ANCOVA procedures suggested by Hastak and Olson (1989) (similar to Baron and Kenny [19861 procedures) were followed and supported theproposition that perceived transaction value mediates the effect ofselling price onperceived acquisition value. ANOVA resultsindicated a significant effect ofselling price onperceived transaction value (Study I: F([.344)= 12.IO, p < .OOI: Study 2: F(1,36)= 48.64, p < .()QI ) and perceived acquisition value (Study 1: F(I 344) = 7.57, p < OI; Study 2: F(1.316) = 25.87, lp < .OOI ). Furthermore, the effect ofselling price onperceived acquisition value was nonsignificant when perceived transaction value was treated as a covariate (Study

Footnote

1: Fti 3)= 1.26, p > .05; Study ?: Fti,315)=.57, p > .OS). In addition, thecovariate was significant (Study 1: Ft l 343) = 130.16, p < OOI: Study 2: FkI.315= 273.91, p < .OOI ). Procedures suggested by Baron and Kenny (1986) were followed to assess whether perceived acquisition value mediates the effect ofperceived transaction value on buyers' willingness to buy. We find that perceived transaction value significantly enhanced perceived acquisition value (Study 1: tt3,rt;46) = 11.43, p < .OOI; Study 2: t(318)= 16.59, p < .OOI ) and willingness to buy (Study I: t(353= 7.25,jp < .001; Study 2; t(3i = 9.99, p < .001). Furthermore, when both transaction value and acquisition value were regressed on buyers' willingness to buy, acquisition value significantly affected willingness to buy (Study 1: t(343= 10.28, p < .OOI: Study 2: t(317)= 12.29, p < .OOI ), while the effect oftransaction value was reduced and nonsignificant (Study 1: t(34= 1.41, p > .05; Study 2: t(317= 1.12; p > .OS). Thus, theregression resultssupport theproposition that the effect ofperceived transaction value onwillingness to buy is mediated by perceived acquisition value. Similarly, we find that perceived

Footnote

transaction value significantly reduced search intentions only in Study 2 (Study I: th356) = -.66, p > .05; Study 2: t(322, = -3.34, p < .OOI ). Thus, mediation could only be tested in Study 2. When both transaction value and acquisition value were regressed on buyers` search intentions, acquisition value significantly affected search intentions (Study 2: t(316)= -3.74, p < .001), while the effect oftransaction value was reduced and non-significant (Study 2: t(316) = .09, p > .OS). As was found for willingness to buy, acquisition value serves to mediate therelationship between perceived transaction value and search intentions. We also tested themediation through nested models. We ran theproposed model (Figure 1 ) with an additional path (transaction value to acquisition value). The resultswere X2 of630 (Study 1) and 841 (Study 2), both with df = 263. A nested model with thethree fewer paths (no linkage between selling priceand acquisition value, no linkages from acquisition value to willingness to buy and search intentions) had Z2 of644.24 (Study I ) and 846.63 (Study 2), both with df = 266. Similar to theregression results, thenested model approach supports themediation hypotheses for Study 2.

Footnote

6. Acquisition value is conceptualized as a function ofperceived quality of theproduct and theselling price. Monroe( 1990) uses a ratio model but points out that this is only a way ofillustrating the comparison. Associated literature on price-quality trade-offs have found greater support for a subtractive model (e.g., Levin and Johnson 1984; White and Truly 1989). We conducted two regression analyses:

Model 1: Perceived acquisition value as a function ofperceived quality and advertised selling price.

Model 2: Perceived acquisition value as a function ofperceived quality, advertised selling price, and an interaction term.

Footnote

The results of theregression analysis indicate thefollowing: Model 1: Study I-F(2,301) = 78.67, adjusted R2 = .34; Study

2-F(2, 321) = 63.14, adjusted R2 = .28. Model 2: Study I-F(3,3oo) = 53. [, adjusted R2 = .347; Study 2-F(3.32o) = 42.66, adjusted R2 = .28.

Footnote

Interaction was not significant in either study. The resultssuggest that acquisition value might be represented best by a subtractive model. Resultsalso suggest that transaction value is represented by a subtractive model. Model 1: Study 1-F(2.3o) = 36.77, adjusted R2 = .19; Study 2-F(233 1)= 12.73, adjusted R2 = .07. Model 2: Study II-F(3,303) = 24.451, adjusted R2 = .195; Study 2-F(3,330) = 8.47, adjusted R2 = .07. Interaction was not significant in either study.

References

REFERENCES

References

Anderson, James C. (1987), "An Approach for Confirmatory Measurement and Structural Equation Modeling ofOrganizational Properties," Management Science, 33 (4), 525-41. and David W. Gerbing (l988), "Structural Equation Modeling in Practice: A Review and Recommended Two-Step Approach," Psychological Bulletin, 103 (3), 411-23. Bagozzi, Richard P. and Youjae Yi (1988), "On theEvaluation ofStructural Equation Models," Journal of theAcademy ofMarketing Science, 16 (Spring), 74-94.

Baron, Reuben M. and David A. Kenny (1986), "TheModeratorMediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations," Journal ofPersonality and Social Psychology, 51 (6), 1173-82.

References

Bearden, William O., Ajit Kaicker, Melinda Smith de Borrero, and Joel E. Urbany (1992), "Examining Alternative Operational Measures ofInternal Reference Prices," in Advances in ConsumCer Research. Vol. 19, John F. Sherry, Jr. and Brian Sternthal, eds. Provo, UT: Association for Consumer Research, 629-35.

, Subhash Sharma, and Jesse E. Teel (1982), "Sample Size Effects onChi-Square and Other Statistics Used in Evaluating Causal Models," Journal ofMarketing Research, 19 (November) 425 30.

Bentler, Peter M. (1990), "Comparative Fit Indexes in Structural Models," Psychological Bulletin, 107 (2), 238-46.

and Douglas G. Bonett (1980), "Significance Tests and Goodness ofFit in theAnalysis ofCovariance Structures," Psychological Bulletin, 88 (3), 588-606.

References

Bolton, Ruth N. and James H. Drew ( 1991), "A Multistage Model ofCustomers' Assessments ofService Quality and Value," Joun-al ofConsumer Research, 17 (March), 375-84. Chapman, Joseph D. and Kent B. Monroe(1990), "TheFraming Effects ofAlternative PricePromotions on Buyers' Subjective

References

Product Evaluations," working paper, Department ofMarketing, Ball State University.

Curry, David J. (1985), "Measuring Priceand Quality Competition," Journal ofMarketing, 49 (Spring), 106-17. Della Bitta, Albert J., Kent B. Monroe, and John M. McGinnis ( 1981), "Consumer Perceptions ofComparative PriceAdvertisements," Journal ofMarketing Research, 18 (November), 416-27.

Dodds, William B., Kent B. Monroe, and Dhruv Grewal(1991), "The Effects of Price, Brand, and Store Information on Buyers' Product Evaluations," Journal ofMarketing Research, 28 (August), 307-19.

References

Dolan, Robert J. (1995), "How Do You Know When the Priceis Right'?" Harvard Business Review, 73 (September/October), 174-83.

Forbes, John L. and Nitin T. Mehta ( 1978), "Value-Based Strategies for Industrial Producs," Busines Horizon.sz 21 (October), 25-31. Fornell, Claes and David F. Larcker (1981), "Evaluating Structural Equation Models with Unobservable Variables and Measurement Error," Journal ofMarketing Researrh, 18 (February), 39-50.

References

Grewal, Dhruv(1989), "The Effect ofIntrinsic, Extrinsic Cues and Reference Prices on Buyers' Perceptions ofQuality and Value," unpublished dissertation, Virginia Polytechnic Institute and State University.

(1995), "Product Quality Expectations: Towards an Understanding ofTheir Antecedents and Consequences," Journal ofBusiness and Psychology, 9 (Spring), 22540. and Larry D. Compeau (1992), "Comparative Price Advertising: Informative or Deceptive?" Journal ofPublic Policy & Marketing, II (Spring), 5242.

and ( 1993), "Consumers' Interpretations ofSemantic Phrases in Comparative PriceAdvertisements: Some Prelimi

References

nary Evidence ona Public Policy Issue," in Advances in Consumer Research, Vol. 20, Leigh McAlister and Michael Rothschild, eds. Provo, UT: Association for Consumer Research, 479-80.

, Howard Marmorstein, and Arun Sharma (1996), "Communicating PriceInformation Through Semantic Cues: The Effects ofSituation and Discount Size," Journal ofConsumer Research, 23 (September), 148-55.

Hastak, Manoj and Jerry C. Olson (1989), "Assessing theRole ofBrand-Related Cognitive Responses as Mediators ofCommunication Effects onCognitive Structure," Journal ofConsumer Research, 15 (March), 444-56.

Herr, Paul M. (1989), "Priming Price: Prior Knowledge and Context Effects," Journal ofConsumer Research, 16 (June), 67-75. Huppertz, John W., Sidney J. Arenson, and Richard H. Evans (1978), "An Application ofEquity Theory to Buyer-Seller Exchange Situations," Journal ofMarketing Research, 15 (May) 250(60. Hyun, Soeun (1993), "The Effects ofContextual Cues onConsumers' Perceptions ofComparative PriceAdvertisements," doctoral dissertation, Department ofMarketing, Virginia Polytechnic Institute and State University.

References

Jacobson, Robert and David A. Aaker (1987), "TheStrategic Role ofProduct Quality," Journal ofMarketing, 51 (October), 3 -44. Joreskog, Karl G. and Dag Sorbom (1989), LISREL VII: Analysis ofLinear Structural Relationships by theMethods ofMaximum Likelihood. Mooresville, IN: Scientific Software, Inc. Kaufmann, Patrick J., N. Craig Smith, and Gwen Ortmeyer (1994), "Deception in Retailer High-Low Pricing: A Rule ofReason Approach," Journal ofRetailing, 70 (Summer), I IS-38. Keon, John W. (1980), "TheBargain Value Model and a Comparison ofManagerial Implications with theLinear Learning Model," Management Science, 26 (November), 1117-30. Klein, Noreen M. and Janet E. Oglethorpe (1987), "Reference Points in Consumer Decision Making,'` in Advances in Consumer Research, Vol. 14, Melanie Wallendorf and Paul F. Anderson, eds. Provo, UT: Association for Consumer Research, 183-87. Levin, Irvin P. and Richard D. Johnson (1984), "Estimating PriceQuality Tradeoffs Using Comparative Judgments," Journal ofConsumer Research, II (June), 593-60.

References

Lichtenstein, Donald R. and William O. Bearden (1989), "Contextual Influences onPerceptions ofMerchant-Supplied Reference Prices," Journal ofConsumer Research, 15 (September), 55466. , Richard G. Netemeyer, and Scot Burton (1990), "Distinguishing Coupon Proneness from Value Consciousness: An Acquisition-Transaction Utility Theory Perspective," Journal ofMarketing, 54 (July), 54-67.

, Nancy M. Ridgeway, and Richard G. Netemeyer (1993), "PricePerceptions and Consumer Shopping Behavior: A Field Study," Journal ofMarketing Research, 30 (May), 234-45. Marmorstein, Howard, Dhruv Grewal, and Raymond P. H. Fishe (1992), "TheValue ofTime Spent in Price ComparisonShopping: Survey and Experimental Evidence," Journal ofConsumer Research, 19 (June), 5241.

References

Marsh, Herbert W., John R. Balla, and Roderick P. McDonald (1988), "Goodness-of-Fit Indexes in Confirmatory Factor Analysis: The Effects ofSample Size," Psychological Bulletin, 103 (3), 39110.

Monroe, Kent B. (1973), "Buyers' Subjective Perceptions of Price," Journal ofMarketing Research, 10 (February), 70-80. (1990), Pricing: Making Profitable Decisions, 2d ed. New York: McGraw-Hill.

and Joseph D. Chapman (1987), "Framing Effects on BuyersSubjective Product Evaluations," in Advances in Consumer Research, Vol. 14, Melanie Wallendorf and Paul Anderson, eds. Provo, UT: Association for Consumer Research, 193-97. , Dhruv Grewal, and Larry D. Compeau (1991), "TheConcept ofReference Prices: Theoretical Justifications and Research Issues," presented at theAssociation for Consumer Research Conference, Chicago (October).

References

and R. Krishnan(1985), "The Effect of Price onSubjective Product Evaluations," in Perceived Quality: How Consumers View Stores and Merchandise, Jacob Jacoby and Jerry C. Olson, eds. Lexington, MA: Lexington Books, 209-32. Nagle, Thomas T. and Reed K. Holden (1995), TheStrategy and

Tactics ofPricing, 2d ed. Englewoods Cliffs, NJ: Prentice Hall. Oliver, R. L. and J. E. Swan ( 1989), "Consumer Perceptions ofInterpersonal Equity and Satisfaction in Transactions: A Field Survey Approach," Journal ofMarketing, 53 (April), 21-35. Phillips, Lynn W., Dae R. Chang, and Robert D. Buzzell (1983), "Product Quality, Cost Position and Business Performance: A Test ofSome Key Hypotheses," Journal ofMarketing, 47 (Spring), 2643.

Rajendran, K. N. and Gerard J. Tellis ( 1994), "Contextual and Temporal Components ofReference Price," Journal ofMarketing, 58 (January), 22-34.

References

Rao. Akshay R. and Kent B. Monroe(1988), "TheModerating Effect ofPrior Knowledge onCue Utilization in Product Evaluations," Journal ofConsumer Research, IS (September), 253-64.

and (1989), "The Effect of Price, Brand Name, and Store Name on Buyers' Subjective Product Assessments: An Integrative Review," Journal ofMarketing Research, 26 (August), 351-57.

Schindler, Robert M. ( 1989), "TheExcitement ofGetting a Bargain: Some Hypotheses Concerning theOrigins and Effects ofSmartShopper Feelings," in Advances in Consumer Research, Vol. 16, Thomas K. Srull, ed. Provo, UT: Association for Consumer Research, 447-53.

Shapiro, Benson P. and Barbara B. Jackson ( 1978), "Industrial Pricing to Meet Customer Needs," Harvard Business Review, (November-December), 119-127.

, V. Kasturi Rangan, Rowland T. Moriarty, and Elliot B. Ross (1987), "Manage Customers for Profits (Not just Sales)," Harvard Business Review, 65 (September/October), 101-108.

References

Stigler, George J. (1961), "TheEconomics ofInformation," Journal ofPolitical Economy, 69 (June), 213-25. Szybillo, George J. and Jacob Jacoby (1974), "Intrinsic Versus Extrinsic Cues as Determinants ofPerceived Quality," Journal ofApplied Psychology, 59 (February), 74-78. Thaler, Richard (1985), "Mental Accounting and Consumer

Choice," Marketing Science, 4 (Summer) 199-214. Tucker, Ledyard R. and Charles Lewis (1973), "A Reliability Coefficient for Maximum Likelihood Factor Analysis," Psychometrika, 38 (1), 1-10.

Urbany, Joel E. and William O. Bearden (1989), "Reference Price Effects onPerceptions ofPerceived Offer Value, Normal Prices, and Transaction Utility," in Enhancing Knowledge Development in Marketing, Paul Bloom et al., eds. Chicago: American Marketing Association, 45-49.

and (1990), "The Effects ofAdvertised PriceInformation onValue Perceptions," unpublished manuscript, Department ofMarketing, University ofSouth Carolina.

, and Dan C. Weilbaker (1988), "The Effect ofPlausible and Exaggerated Reference Prices onConsumer Perceptions and PriceSearch," Journal ofConsumer Research, 15 (June), 95-110.

References

and Peter R. Dickson (1990), "Consumer Knowledge ofNormal Prices: An Exploratory Study and Framework," Working Paper Series, Report No. 90-112, Cambridge, MA: Marketing Science Institute.

White, J. Dennis and Elise L. Truly (1989), "Price-Quality Integration in Warranty Evaluation," Journal ofBusiness Research, 19 (September), 109-25.

Zeithaml, Valarie A. (1988), "Consumer Perceptions of Price, Quality, and Value: A Means-End Model and Synthesis ofEvidence," Journal ofMarketing, 52 (July), 2-22.

AuthorAffiliation

Dhruv Grewalis Associate Professor ofMarketing, Department ofMarketing, University ofMiami. Kent B. Monroeis J. M. Jones Professor ofMarketing, University ofIllinois. R. Krishnanis a professor, Department ofMarketing, California Polytechnic Institute and State University. This article has benefited from comments and suggestions ofJames Littlefield, Julie Ozanne, Edward Fern, Dennis Hinkle, Michael Levy, Arun Sharma, Howard Marmorstein, Diana Grewal, Carolyn Costley, Banwari Mittal, Tamara Mangleburg, and A. Parasuraman. Theauthors also acknowledge thevaluable feedback provided by thethree anonymous JM reviewers. Financial support from a Cunningham Fellowship, Marketing Science Institute, and University ofMiami Summer Research Grant is gratefully acknowledged.

a study expand

2012-01-05T14:07:00.001+07:00

Headnote

TheConceptual Model

Perceived Quality

Internal Reference Price

HI: There is a positive relationship between buyers' perceptions ofquality and their internal reference price.

H2: There is a positive relationship between advertised selling priceand buyers' internal reference price.

H3: There is a positive relationship between advertised reference priceand buyers' internal reference price. Buyers' internal reference pricesadapt to thestimuli pricespresented in theadvertisement. That is, buyerseither adjust their internal reference priceor accept theadvertised reference priceto make judgments about theproduct's value and thevalue of thedeal. Our conceptual model onhow comparison price advertisinginfluences buyers' perceptions ofvalue explicitly recognizes this adaptive nature of buyers' internal reference prices.

Perceived Acquisition Value

H4: There is a positive relationship between buyers' perceptions ofquality and their perceived acquisition value. H5: There is a negative relationship between theadvertised selling priceand buyers' perceptions ofacquisition value.

Perceived Transaction Value

H6: There is a positive relationship between buyers' internal reference priceand their perceived transaction value.

H7: There is a negative relationship between theadvertised selling priceand buyers' perceptions oftransaction value.

Effects ofAcquisition and Transaction Value onWillingness to Buy and Search Intentions

H10: There is a negative relationship between buyers' perceptions ofacquisition value and their intentions to search.

H11: There is a negative relationship between buyers' perceptions oftransaction value and their intentions to search.

An Alternative Model

Research Method Research Plan

Pretests

Measures

Willingness to buy. A three-item scale, based onDodds, Monroe, and Grewal's (1991) study, measured buyers' willingness to buy. Thespecific items were anchored from "very low" to "very high."

Analysis and Results

Plan for Data Analysis

Measurement Properties of theScales

Model Fit

Hypotheses Tests

Advertised pricesand perceived quality. As expected, advertised selling priceand advertised reference pricedid not affect buyers' perceptions ofquality significantly.

Test of theAlternative Model, H12, and Mediating

Hypotheses

Discussion

Conceptual Developments

Measure Development

Managerial Implications

Our study resultssuggest that acquisition value has considerable influence on buyers' willingness to buy. It must be noted that customers balance thebenefits of thepurchase against thecosts. Benefits can be functional, operational (e.g., durability, reliability), or personal (Shapiro and Jackson 1978). Costs include both financial (sale price) and nonfinancial aspects, such as time and effort (Zeithaml 1988). Today's information technology (e.g., through theInternet, consumer reports) enables buyersto compare benefits and priceswith unprecedented ease and accuracy. Managers must understand thevariables affecting theacquisition value of theproduct. They also must understand where their product fits ona continuum ranging from satisfying unique needs (e.g., CAT scanner) to satisfying undifferentiated needs (e.g., corn syrup) (Dolan 1995; Nagle and Holden 1995). Thus, manufacturers can position products that are unique using an acquisition value-enhancing strategy and those that are relatively undifferentiated from competitors using a transaction value-enhancing strategy.

Limitations and Avenues for Further Research

Footnote

2We thank an anonymous reviewer for this suggestion.

Footnote

3We do not predict an interaction termbetween advertised selling priceand advertised reference price, in line with previous models, such as Monroeand Chapman's (1987). However, we did test for theinteraction term. MANOVA analyses onboth data sets indicated that theinteractions were not significant in either data set (p > .05). Furthermore, using LISREL procedures and modeling an interaction term, we did not find a significant effect of theinteraction oninternal reference price, acquisition value, and transaction value. Furthermore, the resultswith an interaction termsuggest a worse fit.

Footnote

40ne of thepretests' (n = 400) resultsindicated that scales used by past research (e.g., Chapman and Monroe1990) to assess overall perceived value did not adequately discriminate acquisition value from transaction value. The resultssuggest a two-factor solution. That is, acquisition value and transaction value load onseparate factors, but overall value loads onboth factors. Thus, this research explicitly focuses onacquisition and transaction value and not onoverall value.

Footnote

5Three conditions must be met to establish mediation: ( I ) theindependent variable affects themediator: (2) theindependent variable affects thedependent variable: and (3) when both theindependent variable and themediator are regressed on thedependent variable, themediator is significant, whereas the effect of theindependent variable is reduced. ANOVA and ANCOVA procedures suggested by Hastak and Olson (1989) (similar to Baron and Kenny [19861 procedures) were followed and supported theproposition that perceived transaction value mediates the effect ofselling price onperceived acquisition value. ANOVA resultsindicated a significant effect ofselling price onperceived transaction value (Study I: F([.344)= 12.IO, p < .OOI: Study 2: F(1,36)= 48.64, p < .()QI ) and perceived acquisition value (Study 1: F(I 344) = 7.57, p < OI; Study 2: F(1.316) = 25.87, lp < .OOI ). Furthermore, the effect ofselling price onperceived acquisition value was nonsignificant when perceived transaction value was treated as a covariate (Study

Footnote

6Acquisition value is conceptualized as a function ofperceived quality of theproduct and theselling price. Monroe( 1990) uses a ratio model but points out that this is only a way ofillustrating the comparison. Associated literature on price-quality trade-offs have found greater support for a subtractive model (e.g., Levin and Johnson 1984; White and Truly 1989). We conducted two regression analyses:

Model 1: Perceived acquisition value as a function ofperceived quality and advertised selling price.

Model 2: Perceived acquisition value as a function ofperceived quality, advertised selling price, and an interaction term.

Footnote

The results of theregression analysis indicate thefollowing: Model 1: Study I-F(2,301) = 78.67, adjusted R2 = .34; Study

2-F(2, 321) = 63.14, adjusted R2 = .28. Model 2: Study I-F(3,3oo) = 53. [, adjusted R2 = .347; Study 2-F(3.32o) = 42.66, adjusted R2 = .28.

Footnote

References

REFERENCES

References

, Subhash Sharma, and Jesse E. Teel (1982), "Sample Size Effects onChi-Square and Other Statistics Used in Evaluating Causal Models," Journal ofMarketing Research, 19 (November) 425 30.

Bentler, Peter M. (1990), "Comparative Fit Indexes in Structural Models," Psychological Bulletin, 107 (2), 238-46.

and Douglas G. Bonett (1980), "Significance Tests and Goodness ofFit in theAnalysis ofCovariance Structures," Psychological Bulletin, 88 (3), 588-606.

References

Product Evaluations," working paper, Department ofMarketing, Ball State University.

Dodds, William B., Kent B. Monroe, and Dhruv Grewal(1991), "The Effects of Price, Brand, and Store Information on Buyers' Product Evaluations," Journal ofMarketing Research, 28 (August), 307-19.

References

Dolan, Robert J. (1995), "How Do You Know When the Priceis Right'?" Harvard Business Review, 73 (September/October), 174-83.

References

and ( 1993), "Consumers' Interpretations ofSemantic Phrases in Comparative PriceAdvertisements: Some Prelimi

References

nary Evidence ona Public Policy Issue," in Advances in Consumer Research, Vol. 20, Leigh McAlister and Michael Rothschild, eds. Provo, UT: Association for Consumer Research, 479-80.

, Howard Marmorstein, and Arun Sharma (1996), "Communicating PriceInformation Through Semantic Cues: The Effects ofSituation and Discount Size," Journal ofConsumer Research, 23 (September), 148-55.

References

Marsh, Herbert W., John R. Balla, and Roderick P. McDonald (1988), "Goodness-of-Fit Indexes in Confirmatory Factor Analysis: The Effects ofSample Size," Psychological Bulletin, 103 (3), 39110.

Monroe, Kent B. (1973), "Buyers' Subjective Perceptions of Price," Journal ofMarketing Research, 10 (February), 70-80. (1990), Pricing: Making Profitable Decisions, 2d ed. New York: McGraw-Hill.

References

Rajendran, K. N. and Gerard J. Tellis ( 1994), "Contextual and Temporal Components ofReference Price," Journal ofMarketing, 58 (January), 22-34.

References

Rao. Akshay R. and Kent B. Monroe(1988), "TheModerating Effect ofPrior Knowledge onCue Utilization in Product Evaluations," Journal ofConsumer Research, IS (September), 253-64.

and (1989), "The Effect of Price, Brand Name, and Store Name on Buyers' Subjective Product Assessments: An Integrative Review," Journal ofMarketing Research, 26 (August), 351-57.

Shapiro, Benson P. and Barbara B. Jackson ( 1978), "Industrial Pricing to Meet Customer Needs," Harvard Business Review, (November-December), 119-127.

, V. Kasturi Rangan, Rowland T. Moriarty, and Elliot B. Ross (1987), "Manage Customers for Profits (Not just Sales)," Harvard Business Review, 65 (September/October), 101-108.

References

Choice," Marketing Science, 4 (Summer) 199-214. Tucker, Ledyard R. and Charles Lewis (1973), "A Reliability Coefficient for Maximum Likelihood Factor Analysis," Psychometrika, 38 (1), 1-10.

and (1990), "The Effects ofAdvertised PriceInformation onValue Perceptions," unpublished manuscript, Department ofMarketing, University ofSouth Carolina.

, and Dan C. Weilbaker (1988), "The Effect ofPlausible and Exaggerated Reference Prices onConsumer Perceptions and PriceSearch," Journal ofConsumer Research, 15 (June), 95-110.

References

and Peter R. Dickson (1990), "Consumer Knowledge ofNormal Prices: An Exploratory Study and Framework," Working Paper Series, Report No. 90-112, Cambridge, MA: Marketing Science Institute.

White, J. Dennis and Elise L. Truly (1989), "Price-Quality Integration in Warranty Evaluation," Journal ofBusiness Research, 19 (September), 109-25.

Zeithaml, Valarie A. (1988), "Consumer Perceptions of Price, Quality, and Value: A Means-End Model and Synthesis ofEvidence," Journal ofMarketing, 52 (July), 2-22.

AuthorAffiliation

the dual

2012-01-05T14:06:00.003+07:00

The dual credibility model: The influence of corporate and endorser credibility on attitudes and purchase intentions

A study proposes a theory of the combined influence of corporate and endorser credibility. Participants assessed the credibility of the companies and spokespersons as well as their attitudes toward the ads and brands, and their intent to purchase the advertised product. The covariance matrix was subjected to a path analysis. The model fit the data, and the findings corroborated prior research indicating that both types of source credibility have an impact on attitudes and purchase intentions albeit a differential one. The results suggest that the Dual Credibility Model partially predicts and explains advertising effectiveness for these dual sources of credibility.

Headnote

This study proposes a theory of thecombined influence of corporateand endorser credibility. Participants assessed the credibility of thecompanies and spokespersons as well as their attitudestoward theads and brands, and their intent to purchase theadvertised product. Thecovariance matrix was subjected to a path analysis. The modelfit thedata, and thefindings corroborated prior research indicating that both types ofsource credibilityhave an impact on attitudesand purchase intentions albeit a differential one. Theresults suggest that the Dual Credibility Modelpartially predicts and explains advertising effectiveness for these dualsources of credibility.

INTRODUCTION

In thequest to maximize theeffectiveness ofadvertising, marketers and researchers have endeavored to understand therelationship between consumers' attitudestoward an advertisement (Aad), their attitudestoward theadvertised brand (AB), and how these attitudes influencetheir purchase intent (PI). The DualMediation Hypothesis (DMH) (Lutz, MacKenzie, and Belch 1983; MacKenzie, Lutz, and Belch 1986; MacKenzie and Lutz 1989; Homer 1990; Brown and Stayman 1992) has been shown to be robust in depicting thecausal sequence ofAad, AB, and PI in an advertising context, thus providing marketers with a better understanding of theinterrelationships among these variables. Since thedevelopment ofthis model, researchers and marketers have explored thedeterminants ofconsumers' attitudestoward advertisements and brands in greater detail in order to develop more persuasive advertising and marketing campaigns.

Source credibilityhas been proposed as an important antecedent to attitudestoward an advertisement and ultimately, advertising effectiveness (e.g., Lutz et al. 1983). Originally, theconcept of credibilitymeant the credibility of the endorseror thespokesperson in thead (e.g., Bergin 1962; Aronson, Turner, and Carlsmith 1963; Bochner and Insko 1966; Sternthal, Phillips, and Dholakia 1978; Ohanian 1990). Many studies support thegeneralization that perceived source (i.e., spokesperson) credibility influences attitudesand behavioral intentions (e.g., Sternthal, Dholakia, and Leavitt 1978; Harmon and Coney 1982; Wu and Shaffer 1987; Moore, Hausknecht, and Thamodaran 1988). Spokesperson credibilityhas been operationalized generally to include three dimensions: expertise, trustworthiness, and attractiveness (DeSarbo and Harshman 1985; Ohanian 1990). Theinclusion ofattractiveness was prompted by research suggesting that a physically attractive communicator is liked more and has a positive impact onopinion change and product evaluations (e.g., Joseph 1982).

More recently, corporate credibilityalso has been shown to play an important role in influencing attitudesand purchase intentions (Lafferty and Goldsmith 1999; Goldsmith et al. 2000a, 2000b). While this form ofsource credibilitydoes appear to have an effect on attitudestoward thead, studies also suggest that it has a stronger effect on attitudestoward thebrand (Lafferty and Goldsmith 1999; Goldsmith et al. 2000a, 2000b). In earlier studies, the influence of corporate credibility onconsumers' attitudesand purchase intentions had only been alluded to in thepersuasion literature (e.g., Kalwani and Silk 1982; MacKenzie and Lutz 1989; Goldberg and Hartwick 1990) or in managerially focused discussions (Gregory 1991; Sobol et al. 1992). Previous studies have referred to corporate credibilityas advertiser credibility(Lutz et al. 1983; Lutz 1985; Mackenzie and Lutz 1989), advertiser reputation (Goldberg and Hartwick 1990) and only more recently, as a determinant of corporatereputation (Fombrun 1996; Keller 1998). Although it is acknowledged to be an important part of corporateimage and reputation, corporate credibilityhas not been systematically studied in thecontext ofits effect on attitudesand purchase intentions.

Corporate credibilityforms part ofa positive corporateimage or reputation (Gregory 1991; Sobol et al. 1992; Fombrun 1996; Keller 1998). A corporateimage is thetotality ofimpressions that a firm makes on theminds ofconsumers and is evoked by the corporatename or logo (Pope and Voges 1999). Corporate credibility, or theextent to which consumers, investors, and other constituents believe in a company's trustworthiness and expertise, makes up a portion ofa corporation's image (Fombrun 1996). A corporation's credibilitycan be harmed when it is suspected ofproducing dangerous or inferior products, when it lies to consumers or others, or when it is reported to be in violation oflegal or ethical norms (Frombrun 1996). Theethics literature indicates that when ethical perceptions are exceeded by company actions, thecompany is rewarded in terms ofmore positive attitudesand behavioral intentions (e.g., Creyer and Ross 1997). Corporate credibilityplays a critical role in thefirm's ability to secure loans, create partnerships, and market products (Gregory 1991; Haley 1996). Corporate credibilityis also important in shaping corporateidentity (Stuart and Kerr 1999). Consumers who perceive a company as credible are more likely to evaluate thefirm's advertisements favorably and to buy thecompany's products (Keller 1998).

Goldsmith et al. (2000a) describe thecausal relationships that exist for corporate credibilityand endorser credibilityin thecontext of thelatent variables in theDMH model. Thedevelopment oftheir proposed modelwas based onan earlier study suggesting theeffect ofboth antecedents onAad, AB, and PI (Lafferty and Goldsmith 1999). Thepurpose ofthis study is to empirically test the Dual Credibility Model(DCM) proposed by Goldsmith et al. (2000a) in an advertising context and suggest theoretical, managerial, and methodological implications from thefindings.

THE DUAL CREDIBILITY MODEL

The Dual Credibility Model(see Figure 1) builds upon thelatent variables in theDMH model, which specifies a direct causal sequence from Aad to AB to PI (see MacKenzie et al. 1986). TheDMH specifies that ad cognitions represent an antecedent to Aad and have a direct effect onthis variable. Within thecontext ofad cognitions, endorser credibilitycan be one of thevariables consumers use to judge theadvertisement and thus, its persuasiveness (e.g., Sternthal, Phillips, and Dholakia 1978; Harmon and Coney 1982; Moore et al. 1988). This, in turn, can lead to attitudeformations toward theadvertisement. Seven hypothesized paths among thefive variables comprise the model.

Theeffects of endorseror source credibility onadvertising effectiveness have been well documented in themarketing and social psychology literature (e.g., Aronson, Turner, and Carlsmith 1963; Sternthal, Phillips, and Dholakia 1978; Harmon and Coney 1982;). According to Fishbein and Ajzen (1975), source credibilityaffects theprobability that a message claim will be accepted. In general, a message delivered by a high credibilitysource will be accepted more readily and is more likely to lead to greater attitudechange (Kelman and Hovland 1953; Johnson, Torcivia, and Poprick 1968; Miller and Baseheart 1969; Warren 1969; Schulman and Worrall 1970). In thecontext ofAad -> AB -> PI, theperceived credibility of the endorserhas been shown to have an influential effect onAad in thestudy by Lafferty and Goldsmith (1999) and is one of thetwo exogenous variables in theproposed DCM (Goldsmith et al. 2000a). Thus, theDCM hypothesizes that

H1: Endorser credibilityis positively and directly related to attitude-toward-the-ad.

Thefindings by Lafferty and Goldsmith (1999) also showed that corporate credibility influencesAad directly. While theeffects onAad were weaker than theeffects produced by thecredible endorser, theresults suggested that a direct path between thetwo variables did exist. MacKenzie and Lutz (1989) also indicated thelikelihood ofthis relationship. Their modelshowed that advertiser credibility(synonymous with corporate credibility) had a strong positive effect on attitudetoward theadvertiser, which in turn, had a strong positive effect onAad. Thecausal path from corporate credibilityto Aad was confirmed in thestudy by Goldsmith et al. (2000a), and thus corporate credibilityrepresents thesecond exogenous variable in theDCM. Therefore, it is hypothesized that

H2: Corporate credibilityis positively and directly related to Aad.

Corporate credibilityalso has been shown to have a direct effect onAB (Lafferty and Goldsmith 1999). MacKenzie and Lutz (1989) indicated that advertiser credibilityis more a central processing cue. This would suggest that consumers' evaluations ofthe company's credibilitywould be incorporated into their assessment of thebrand, prompting more cognitive processing. According to theElaboration Likelihood Model(ELM), when consumers are more motivated to centrally process an ad, thebrand-relevant aspects increase and theperipheral cues decrease (Petty and Cacioppo 1983). Thedirect effect of corporate credibility onAB was confirmed in thestudy by Goldsmith et al. (2000a). Thus, theDCM posits that

H3: Corporate credibilityis positively and directly related to AB.

Lafferty and Goldsmith (1999) also showed that corporate credibilityhad a direct effect onPI. This empirical evidence is consistent with studies by Newell (1993) and Davis (1994) who found a positive effect of corporate credibility onall three outcome variables, Aad, AB, and PI. Perceptions of thecompany's trustworthiness and expertise are part of theinformation consumers use to judge thecompany's products and determine whether they are inclined to buy them (Fombrun 1996). Firestone Tire Company exemplifies this phenomenon. Thecompany suffered serious problems in 2000 from all thenegative publicity it received regarding its Wilderness tires and their propensity to explode causing SUV's, particularly Ford Explorers, to flip and roll. Firestone stock fell 60 percent from August 2000 to January 2001 (Dawson 2001) and sales fell 40% in September and October 2000 (Davis 2000). Thestudy by Goldsmith et al. (2000a) also supported thedirect relationship between corporate credibilityand PI. Therefore, it is hypothesized for theDCM that

H4: Corporate credibilityis positively and directly related to Pl.

Substantial research has been done on thecausal sequence ofAad, AB, and PI to explain advertising effectiveness (Heath and Gaeth 1994). Themajority offindings suggest that theorder ofeffects appears to be robust (MacKenzie et al. 1986; MacKenzie and Lutz, 1989; Homer 1990; Brown and Stayman 1992). This causal sequence is reflected as thelatent variables in theDMH modeland is also theoutcome variables in theDCM. Thus,

115: Aad is positively and directly related to AB. H6: AB is positively and directly related to PI.

Finally, there is evidence for a direct relationship between Aad and PI in certain circumstances (MacKenzie et al. 1986; Cox and Locander 1987; Biehal et al. 1992). This appears to be true when affective responses are evoked ( Petty, Cacioppo, and Goldman 1981; Gorn 1982; Batra and Ray 1985; Cox and Locander 1987;). Consumer may decide what they would buy without completely processing all brand information (see Biehal et al. 1992). Theads in this study contained no strong arguments and used minimal copy thus, providing limited information about theproduct and relied primarily on the influence of the endorserand on theinformation provided about thecompany to test the model. It is likely that thephotograph of the endorserand thecompany description generated some feelings or affect either positive or negative owing to thelack ofproduct information (Mitchell 1986; Stayman and Batra 1991). Thus, theevaluations stimulated by theads would more than likely be chiefly affective or emotional in nature and not largely analytical. This affective response to theads could influence thedeclarations ofpurchase intent directly without all theeffects being mediated by thebrand evaluations. Any cognitive processing that does occur to influencepurchase intent would primarily be an outcome of attitudetoward thebrand. Thespecific selection ofproduct and type ofad used for this study is similar to thestimuli used by Goldsmith et al. (2000a) and therefore is hypothesized to produce similar results. Thus:

H7: When an ad contains few and weak arguments, Aad is positively and directly related to PI.

In sum, theDCM posits both direct and indirect differential relationships between endorserand corporate credibilitywith important reactions to advertisements.

METHOD

Subjects

Thedata came from 315 undergraduate students who were taking an introductory marketing course and volunteered to participate in an advertising experiment. Thedata were collected in a manner similar to that used by Simonin and Ruth (1998), who randomly exposed small groups ofsubjects to different combinations ofbrands ofmicroprocessors and automobiles and then tested a modelusing theaggregated data. Data were combined from two groups ofsubjects who were exposed to different combinations of endorsersand companies. Thefirst group consisted of161 women marketing students at a university in theU.S. Midwest. They were randomly assigned to one offour experimental treatments in a 2 (high/low corporate credibility) x 2 (high/low endorser credibility) between-subjects factorial design with female endorsers. Thesecond group ofsubjects was 154 male students from thesame Midwest University. They were randomly assigned to one offour experimental treatments in thesame 2 x 2 design, but male endorserswere used in their ads. Men and women saw endorsers of thesame sex because a pretest showed same sex endorsersto be viewed by theparticipants as more realistic. Women, especially, were not effective endorsers ofathletic shoes for men. Thus, theonly difference in thetreatment between thetwo samples was the endorserused in thead (see Stimuli section for details). In both cases, pretests were conducted to select suitable high and low credibilitysources for themale and female endorsers. Both high credibility endorserswere athletes, and both low credibility endorserswere actors ofsimilar weight and proportions. Because identical measures ofall constructs were taken in both groups, cross validation procedures were employed to test therobustness of thefindings.

Stimuli

Subjects saw one oftwo ads for Pride, a purportedly new brand ofathletic shoes. This is a relevant product category that thestudent subjects were highly familiar with and with which most had purchasing experience. Thestudy used a fictitious brand ofathletic shoes to prevent prior knowledge ofa brand from biasing theresults of thestudy. Thewomen subjects in thehigh endorser credibilitycondition viewed an ad featuring Florence Griffith-Joyner, theOlympic gold medallist, as its spokesperson. (Thestudy was conducted prior to theuntimely death ofMs. Griffith-Joyner.) Women in thelow endorser credibilitycondition saw thesame ad with Roseann Barr, theactress, as thespokesperson. These endorserswere used after a pretest suggested they represented opposite poles of endorser credibilityfor athletic shoes. For themen, discussions with students suggested that thegolfer, Tiger Woods, would be a positive spokesperson for running shoes; and Wayne Knight, theactor who played Newman onSeinfeld, was suggested as an appropriate substitute for Roseanne Barr. Thequestionnaire booklet was otherwise identical, and thesame procedures were followed.

Providing descriptions of thefictitious company that manufactured theshoes influenced perceptions ofhigh and low corporate credibility. Once again, a fictitious company was selected to prevent prior knowledge ofa company unduly influencing therespondents. One corporatehistory presented a positive image while theother presented thecompany in a negative manner. This manipulation was adapted from a procedure used by Goldberg and Hartwick (1990) who successfully used it to manipulate thereputation ofa company. Company descriptions were approximately equal in length and, except for thepositive or negative tone, presented virtually thesame information. Perceptions of thecompany's length oftime in business, their degree ofnational and international success, thesize of thecompany, and recognition oftheir social consciousness were manipulated to reflect a positive versus a negative (less positive) image. Theonly difference in content was thepositive manipulation contained information on thecompany's community and environmental contributions while thenegative manipulation contained information onpoor quality control and alleged SEC violations. These elements are typical cues used by consumers to assess how credible a corporation might be.

Procedure

Four booklets containing different endorser/credibilitycombinations were prepared. Thefront page ofeach booklet contained general instructions for thestudy and thecover story. Subjects were told that thestudy assessed student reactions to magazine advertisements. They were instructed to read each page carefully and to look at thead as if they were seeing it in a magazine (Yi 1990). Thesubjects were informed that when they finished, they should keep thebooklet face down until it was collected.

Thesecond page of thebooklet contained thedescription of thecorporation either negative or positive depending on thetreatment (Goldberg and Hartwick 1990). Thestudents were instructed to read theexcerpt, which was said to be from a Wall Street Journal article. Following thedescription, they were instructed to proceed without turning back.

Thenext page showed thead for Pride athletic shoes that contained a picture depicting one of thefour endorsers. Thelayout for each ad as well as theheadline and copy were identical in each of thefour conditions to eliminate thepossibility ofany potential confounds. Thecopy consisted ofa short quotation that was attributed to the endorserstressing thenatural fit of theshoe. No other product attributes or information about benefits was provided. Thus, theonly differences among thefour treatments were the endorserand the credibility of thecorporation.

Measures

Six 7-point bi-polar adjective scales assessed endorser credibility. These scales were derived from those developed by Ohanian (1990) for measuring celebrity endorsers' perceived attractiveness, trustworthiness, and expertise. Two items, "attractive/unattractive" and "classy/not classy," measured attractiveness; two items with endpoints of"sincere/insincere" and "trustworthy/untrustworthy" measured trustworthiness; and two items with anchors of"expert/not an expert" and "experienced/inexperienced" measured expertise. Corporate credibilitywas measured via the Corporate CredibilityScale developed by Newell and Goldsmith (2001). This is an eight-item, 7-point Likert scale with a five-point, agree/disagree response format. Four items measure corporatehonesty or trustworthiness (e.g., I trust thePride Corporation."), and four items measure corporateexpertise (e.g., "ThePride Corporation has great expertise."). These are two of themajor dimensions of corporate credibility(Fombrun 1996; Haley 1996).

To measure attitudetoward thead, attitudetoward thebrand, and purchase intention (cf., Tripp et al. 1994) participants were first asked to use three 7-point, bi-polar adjective scales to rate their overall impression of theadvertised product (AB). Thescales were anchored with "good/bad," "favorable/ unfavorable," and "satisfactory/unsatisfactory" (Bruner and Hensel 1992). Following this measure of attitude-toward-the-- brand, thesubjects were asked to rate their overall impression of thead (Aad) onthree 7-point bi-polar scales anchored by "good/bad," "favorable/unfavorable," and "pleasant/ unpleasant" (MacKenzie and Lutz, 1989). Next, thesubjects were asked how likely it would be that they would consider buying that brand (PI) thenext time they purchased athletic shoes. Three 7-point bi-polar scales were used, anchored by "very likely/very unlikely," "probable/improbable," and "possible/impossible" (Yi 1990).

Thefinal questions in thebooklet asked for demographic data, and one question asked subjects if they could identify thepurpose ofparticipating in thestudy. This was done to assure that no one guessed thehypotheses, which could bias their responses and invalidate their answers. Theresponses showed that no subject appeared to be aware of theintent of thestudy, so all questionnaires were used.

RESULTS

Preliminary Analyses

Theitems used to measure themanipulations and dependent variables were factor analyzed using theSPSS principal axis factor method followed by an oblique rotation to determine scale dimensionality and item convergent and discriminant validity. Thesix items that measured endorser credibilitywere factor analyzed. Only one factor with an eigenvalue greater than 1.0 was extracted. Thesix items were summed to form a measure of endorser credibility. We assessed internal consistency by computing Cronbach's coefficient alpha (alpha =.93).

Table 1 presents thefactor analysis results for the corporate credibilityscale and theAad, AB, and PI scales. For the corporate credibility, thefactor analysis extracted two factors with eigenvalues greater than 1.0. Four items formed an "honesty" factor with theother four items making up an "expertise" factor (see Newell and Goldsmith 2001). We summed theeight items to form an overall measure of corporate credibility. We assessed internal consistency by computing coefficient alpha (alpha = .92).

Thenine items measuring Aad, AB, and PI were also factor analyzed (see Table 1). Theresults showed a three-factor structure with theitems loading on the"correct" factors. Thethree items measuring Aad were then factor analyzed and found to be unidimensional, with one factor accounting for 81 percent of thevariance. Factor analysis was also performed on thethree items assessing AB and on thethree used to assess PI. Both sets also demonstrated unidimensionality, with one factor accounting for 90 percent of thevariance in AB and one factor accounting for 89 percent of thevariance in PI. Since each set ofitems was unidimensional, theitems comprising each scale were summed to assess overall Aad, AB, and PI for Pride athletic shoes. These three scales had high internal consistency with coefficient alphas of.93, .96, and .96 for Aad, AB, and PI, respectively.

Thus, there is strong evidence that theitems used to measure thevariables in this study possessed convergent and discriminant validity as well as high internal consistency. Thescales were all scored so that higher scores represented higher levels of theconstruct.

Pooling theData

DISCUSSION

Thepurpose ofthis study was to empirically test the Dual Credibility Modelto determine if it could serve as a predictor ofadvertising effectiveness when evaluating dualsources of credibility. Thefindings are consistent with Goldsmith et al. (2000a) and argue for theusefulness ofthis modelin both theoretical and managerial contexts.

Theoretical Implications

Thefindings support Stem's (1994) argument that consumers do not perceive only a single communicator but may differentiate between thepresenters and thecompanies they represent. While additional applications ofthis modelneed to be tested to confirm that theproposed paths are robust under varying conditions, this study supports thehypothesized influences of endorserand corporate credibility onAad, AB, and PI. Similar to Goldsmith et al. (2000a), this study confirmed that endorser credibilityhad a greater effect onAad (.24) than corporate credibility(.083), although both paths were significant. Moreover, this study confirms thehypothesis that corporate credibilityhas a direct effect onall three advertising-related variables, unlike endorser credibility, which mainly influences attitudestoward thead. Credible endorsersseem to influence attitudestoward thebrand and purchase intentions indirectly. This study also lends support to thehypothesized direct path from attitudetoward thead to purchase intent (Biehal et al. 1992). These findings represent an advance in our understanding of thecomplex interactions among thevarious features ofadvertising and consumer cognitions that lead ultimately to purchase. They focus attention on theimpact of corporate credibility onconsumer behavior in a domain hitherto preoccupied with spokesperson and endorsers.

Managerial Implications

This study indicates that corporate credibilityplays an important role in consumer evaluation ofadvertisements. In addition, a company's credibilityseems to have a direct effect onperceptions ofbrands and ofpurchase intentions. Because these company perceptions may be central to thesuccess or failure ofpromotions and sales, managers need to become much more aware ofhow consumers view thecorporation (Fombrun 1996). Consequently, companies must constantly monitor perceptions of corporate credibilityby consumers as well as other stakeholder groups. Annual, or even more frequent, surveys ofconsumers and their feelings about a corporation seem to be fundamental for effective decisionmaking. By tracking consumer perceptions of corporate credibility, managers can detect changes in this important marketing metric and take corrective action if needed. Themeasures for corporate credibilitypresented in this modelcan provide managers with a better idea ofhow well thecompany is viewed in theconsumer's eyes.

Companies that face tough times may be particularly well served by theconstant monitoring ofconsumer perceptions oftheir company's credibility. By understanding how perceptions have changed, and in particular how each component of the credibilityconstruct has been altered, companies such as Exxon during theValdez incident, Microsoft during their antitrust suit, and Anderson Consulting during and after its association with theEnron scandal, may have been better able to develop appropriate responses in order gain back lost trust and/or perceptions ofexpertise. For example, if a company found that after a recent crisis perceptions oftrust among their distributors and feelings ofcompetency and expertise among end consumers had significantly decreased, a two-pronged response could be developed. In this situation, a company might use sales representatives to help solidify their relationships with distributors by increasing visits to accounts and providing increased guarantees onproducts and greater reassurances to their customers about their commitment to serving them better. In addition, advertising and other media driven strategies for consumers could emphasize thelong history ofservice and dependability and may use a highly respected spokesperson to present logical and compelling arguments about why thecompany is still thebest choice. Overall, in times ofcrisis, those companies that have kept informed ofconsumers perceptions ofcompany credibilitymay be better equipped to develop strategies to regain feelings oftrust and/or perceptions ofexpertise among their target markets.

Coupling this measure of corporate credibilitywith themeasure for the credibility ofany endorser thecompany uses will allow managers to see therelative value ofeach credible source. This, in turn, should prove to be a useful measure for promotional and marketing budget allocations. This information can provide managers with a better understanding how consumers view their company, endorsers, and their advertisements. With this information managers can then develop more appropriate promotional strategies for both their products and their corporations.

Well thought-out strategies for maintaining a good reputation or improving a poor image are an essential step to developing effective advertisements. Specific marketing communications strategies to enhance a company's credibilitymay include developing appropriate corporateadvertising that focuses onone or both of thedimensions of credibility. Another means to improve corporatereputation may be through public relations and sponsorships. In general, without a credible reputation, successful advertising is much more difficult to develop. Partnering with charitable causes in cause-brand alliances may also be a good way to enhance corporatereputation.

Methodological Implications

Thefindings promote theuse ofmodeling to study these phenomena. Structural analysis provides researchers with theopportunity to test modelsusing a single comprehensive method that examines multiple relationships simultaneously. Testing a modelin this manner allowed us to test all therelationships among corporateand endorser credibilitysimultaneously with measures ofreaction to advertising in order to corroborate experimental studies. Both methods (experimentation and causal modeling) should be used in tandem to provide a comprehensive explanation ofmarketing phenomena. Moreover, theresults confirmed thereliability and validity of themeasures of endorserand corporate credibility, thus promoting their use by other researchers in any studies ofthese constructs.

Limitations to Generalizability

TheDCM was tested for a single product with a fictitious brand in a single ad exposure condition. Clearly, to establish any sort ofgeneralizable conclusions for this model, more tests to extend these critical dimensions must be examined. Differences for the modelmay exist for durable versus nondurable products, for higher versus lower involvement products, and for authentic and more recognizable brands. Multiple exposures to thead also may influence thevarious relationships. The endorsersselected also limited thestudy's generalizability. While pretests ensured that the endorsersselected were appropriate for thetreatment conditions, other endorsersmay change thestrength of therelationships between variables or thevery relationship. Certainly, theselection of thelow credibility endorsersmay be considered extreme, but they did represent realistic examples oflow credibilityspokespersons. In addition, themanipulation for corporate credibilitypresented two opposite corporateimages. While this manipulation has been used in thepast with success (Goldberg and Hartwick 1990; Lafferty and Goldsmith 1999), it may unduly influence therespondents' impressions ofbrand quality, producing a stronger effect on attitudestoward thebrand. This is not necessarily an inappropriate representation ofa company since this type ofdescription, whether positive or negative, can exist in thereal world.

While theDCM appears robust in this study, theweak link in the modelis thepath from attitudetoward thead to purchase intent (M). This path exists under certain circumstances and possibly will not be robust when thecontent of thead changes or with products that engender less affect. Additional study should be devoted to delineating thecircumstances where this direct path pertains and where it does not.

Specifications of thepotential limits ofour findings, however, do not necessarily diminish their value. TheDCM may represent thebasic structure ofresponse processes that exist when consumers are exposed to a new brand. This study also extends thegeneralizability of thefindings of theDCM by Goldsmith et al. (2000a) since they used a real company, an institutional ad, and a single, relatively high credibility endorserto test theproposed model.

Future Research

Thelimitations provide opportunities for future research to test theviability of theDCM. Different products, different types ofads (institutional, comparative, reminder), different endorsers(CEO's versus celebrities), different companies (real versus fictitious), different media (broadcast versus print) all present opportunities to test the modelunder different conditions. Additional moderating variables also should be examined such as knowledge or involvement to see what impact these variables have on theDCM. These factors may influence thestrength of therelationships among thevariables in the modeland are, therefore, candidates for discussion.

References

REFERENCES

References

Anderson, J. C. and Gerbing, D. W. (1988), "Structural Equation Modeling In Practice: A Review And Recommended Two-Step Approach," Psychological Bulletin, 103(3),411-423.

Aronson, E., Turner, J. A. and Carlsmith, J. M. (1963), "Communicator CredibilityAnd Communications Discrepancy As Determinants Of AttitudeChange," Journal ofAbnormal and Social Psychology, 67 (July), 31-36.

Batra, R. and Ray, M. L. (1985), "How Advertising Works At Contact," in L. Alwitt and A. Mitchell (eds) Psychological Processes and Advertising Effects: Theory, Research, and Application, Hillsdale, NJ: Lawrence Erlbaum, 13-43.

Bergin, A. E. (1962), "TheEffect OfDissonant Persuasive Commications Upon Changes In Self-Referring Attitudes," Journal ofPersonality, 30 (MarchDecember), 423-438.

Biehal, Gabriel, Debra Stephens, and Eleonora Curlo (1992), "AttitudeToward theAd and Brand Choice," Journal ofAdvertising, 21 (3), 19-36.

Bochner, S. and Insko, C. A. (1966), "Communicator Discrepancy, Source Credibility, And Opinion Change," Journal ofPersonality and Social Psychology, 4 (December), 614-621.

References

Brown, S. P. and Stayman, D. M. (1992), "Antecedents And Consequences Of AttitudeToward TheAd: A MetaAnalysis," Journal ofConsumer Research, 19 (June), 34-51.

Bruner, G. C. II and Hensel, P. J. (1992) Marketing Scales Handbook: A Compilation ofMulti-Item Measures, Chicago, IL: American Marketing Association.

Cox, D. and Locander, W. (1987), "Product Novelty: Does It Moderate TheRelationship Between Ad AttitudesAnd Brand Attitudes," Journal ofAdvertising, 16 (3), 39-44.

References

Creyer, Elizabeth and William T. Ross (1997), "The Influence ofFirm Behavior onPurchase Intentions: Do Consumers Really Care about Business Ethics?" Journal ofConsumer Marketing, 14 (6), 421-428.

Davis, Bruce (2000), "Firestone Prepares Effort to Rebuild Damaged Brand,"Advertising Age 71 (December), 8. Davis, J. J. (1994), "Consumer Response To Corporate

Environmental Advertising," Journal ofConsumer Marketing, 11 (2), 25-47.

Dawson, Chester (2001), "What Japanese CEO's can Learn from Bridgestone," Business Week (January), 50.

References

DeSarbo, W. S. and Harshman, R. A. (1985), "Celebrity-- Brand Congruence Analysis," in J. H. Leigh and C. R. Martin, Jr. (eds) Current Issues and Research in Advertising, Ann Arbor, MI: Division ofResearch Graduate School ofBusiness Administration, TheUniversity ofMichigan, 17-52.

Fishbein, Martin and Ajzen, Icek (1975), Belief Attitude, Intention, and Behavior: An Introduction to Theory and Research, Reading, MA: Addison-Wesley.

Fombrun, C. J. (1996), Reputation, Boston, MA: Harvard Business School Press.

References

Goldberg, M. E., and Hartwick, J. (1990), "TheEffects OfAdvertiser Reputation And Extremity OfAdvertising Claim OnAdvertising Effectiveness," Journal ofConsumer Research, 17 (September), 172-179.

Goldsmith, Ronald E., Lafferty, Barbara A., and Newell, Stephen J. (2000a), "TheImpact Of Corporate CredibilityAnd EndorserCelebrity OnConsumer Reaction To Advertisements And Brands," Journal ofAdvertising, (Fall), 43-54.

Goldsmith, Ronald E., Lafferty, Barbara A., and Newell, Stephen J. (2000b), "The Influence of Corporate Credibility onConsumer Attitudesand Purchase Intent," CorporateReputation Review, 3(4),304-318.

Gorn, G. J. (1982), "TheEffects OfMusic In Advertising OnChoice Behavior: A Classical Conditioning Approach," Journal ofMarketing, 46 (Winter), 94101.

References

Gregory, J. (1991), Marketing CorporateImage, Lincolnwood, IL: NTC Business Books.

Harmon, R. and Coney, K. (1982), "ThePersuasive Effects OfSource CredibilityIn Buy And Lease Situations," Journal ofMarketing Research, 19 (May), 255-260.

Haley, E. (1996), "Exploring TheConstruct OfOrganization As Source: Consumers' Understandings OfOrganizational Sponsorship OfAdvocacy Advertising," Journal ofAdvertising, 25 (Summer), 19-35.

Heath, T. B. and Gaeth, G. J. (1994), "Theory And Method In TheStudy OfAd And Brand Attitudes: Toward A Systemic Model." In E. M. Clark, T. C. Brock, and D. W. Stewart, (eds). Attention, Attitude, and Affect in Response to Advertising. Hillsdale, NJ: Lawrence Erlbaum, 125-148.

References

Homer, P. M. (1990), "TheMediating Role Of AttitudeToward TheAd: Some Additional Evidence," Journal ofMarketing Research, 27 (February), 7886.

Johnson, H., Torcivia, J., and Poprick, M. (1968), "Effects OfSource Credibility Of TheRelationship Between Authoritarianism And AttitudeChange," Journal ofPersonality and Social Psychology, 9, 179-183.

Joseph, W. B. (1982), "The Credibility OfPhysically Attractive Communicators: A Review," Journal ofAdvertising 11, (July), 15-24.

Kalwani, M. U. and Silk A. (1982), "On TheReliability And Predictive Validity OfPurchase Intention Measures, Marketing Science, I (Summer), 243-286.

Keller, K. L. (1998), Strategic Brand Management. Upper Saddle River, NJ: Prentice Hall.

Kelman, H. and Hovland, C. (1953), "Reinstatement Of TheCommunicator In Delayed Measurement OfOpinion Change," Journal ofAbnormal and Social Psychology, 48, 327-335.

References

Lafferty, B. A. and Goldsmith, R. E. (1999), "Corporate Credibility's Role In Consumers' AttitudesAnd Purchase Intentions When A High Versus A Low Credibility EndorserIs Used In TheAd," Journal ofBusiness Research, 44, (February), 109-116.

Lutz, R. J. (1985), "Affective And Cognitive Antecedents Of AttitudeToward TheAd: A Conceptual Framework." In L. F. Alwitt and A. A. Mitchell (eds) Psychological Processes and Advertising Effects.- Theory, Research and Application. Hillsdale, NJ: Lawrence Erlbaurn Associates, 45-63.

Lutz, R. J., MacKenzie, S. B., and Belch, G. E. (1983), "AttitudeToward TheAd As A Mediator OfAdvertising Effectiveness: Determinants And Consequences," Advances in Consumer Research, R. P. Bagozzi and A. M. Tybout (eds) Ann Arbor, MI: Association for Consumer Research, 532-539.

MacKenzie, S. B. and Lutz, R. J. (1989), "An Empirical Examination of theStructural Antecedents of Attitude-Toward-The-Ad in an Advertising Pretesting Context," Journal ofMarketing, 53 (April), 48-65.

References

MacKenzie, S. B., Lutz, R. J. and Belch, G. E. (1986), "TheRole of AttitudeToward TheAd as a Mediator ofAdvertising Effectiveness: A Test ofCompeting Explanations," Journal ofMarketing Research, 23 (May), 130-143.

Miller, G. and Baseheart, J. (1969), "Source Trustworthiness, Opinionated Statements, and Response to Persuasive Communication," Speech Monographs, 36, 1-7.

Mitchell, Andrew A. (1986), "TheEffect ofVerbal and Visual Components ofAdvertisements onBrand Attitudesand AttitudeToward theAdvertisement," Journal ofConsumer Research, 13 (June), 12-25.

Moore, D. L., Hausknecht, D. and Thamodaran, K. (1988), "Time Compression, Response Opportunity, and Persuasion," Journal ofConsumer Research, 13 (June),12-24.

References

Newell, S. J. (1993), Developing A Measurement Scale And A Theoretical ModelDefining Corporate CredibilityAnd Determining Its Role As An Antecedent OfConsumers' AttitudeToward TheAdvertisement. Doctoral Dissertation, Florida State University.

Newell, S. J. and R. E. Goldsmith (2001), "TheDevelopment OfA Scale To Measure Perceived Corporate Credibility," Journal ofBusiness Research, 52 (3), 235-247.

Ohanian, R. (1990), "Construction And Validation OfA Scale To Measure Celebrity Endorsers' Perceived Expertise, Trustworthiness, and Attractiveness," Journal ofAdvertising, 19 (3), 39-52.

Petty, R. E. and Cacioppo, J. T. (1983), "Central And Peripheral Routes To Persuasion: Application To Advertising." In L. Percy and A. G. Woodside, (eds) Advertising and Consumer Psychology. Toronto: Lexington Books, 3-23.

Petty, R. E., Cacioppo, J. T. and Goldman, R. (1981), "Personal Involvement As A Determinant OfArgument-Based Persuasion," Journal ofPersonality and Social Psychology, 41 (November), 847-855.

Pope, N. K. L. and Voges, K. E. (1999), "Sponsorship And Image: A Replication And Extension," Journal ofMarketing Communications, 5(l),17-28.

Schulman, G. and Worrall, C. (1970), "Salience Patterns, Source Credibility, And TheSleeper Effect," Public Opinion Quarterly, 34, 371-382.

References

Simonin, B. L., and Ruth, J. A. (1998), "Is A Company Known By TheCompany It Keeps? Assessing TheSpillover Effects OfBrand Alliances OnConsumer Brand Attitudes?" Journal ofMarketing Research, 35 (February), 30-42.

Sobol, Marion G., Gail E. Farrelly, and Jessica S. Taper (1992), Shaping the CorporateImage, New York: Quorum Books.

Stayman, Douglas M. and Rajeev Batra (1991), "Encoding and Retrieval ofAd Affect in Memory," Journal ofMarketing Research, 28 (May), 232-240.

Stem, B. (1994), "A Revised Communication ModelFor Advertising: Multiple Dimensions Of TheSource, TheMessage, And TheRecipient," Journal ofAdvertising, 23 (June), 5-15.

Sternthal, B., Phillips, L. W. and Dholakia, R. (1978), "ThePersuasive Effect OfSource Credibility: A Situational Analysis," Public Opinion Quarterly, 42 (Fall), 285-314.

References

Sternthal, B., Dholakia, R. and Leavitt, C. (1978), "ThePersuasive Effects OfSource Credibility: Tests OfCognitive Response," Journal ofConsumer Research, 4 (March), 252-260.

Stuart, H. and Kerr, G. (1999), "Marketing Communication And CorporateIdentity: Are They Integrated?" Journal ofMarketing Communications, 5(4), 169179.

Tripp, C., Jensen, T. D., and Carlson, L. (1994), "TheEffects OfMultiple Product Endorsements By Celebrities OnConsumers' AttitudesAnd Intentions," Journal ofConsumer Research, 20 (March), 535-547.

Warren, 1. (1969), "TheEffect Of CredibilityIn Sources OfTestimony OnAudience AttitudesToward Speaker And Message," Speech Monographs, (36) 456-458.

Wu, C. and Shaffer, D. (1987), "Susceptibility To Persuasive Appeals As A Function OfSource CredibilityAnd Prior Experience With The AttitudeObject," Journal ofPersonality and Social Psychology, 52 (April), 677-688.

Yi, Y. (1990), "Cognitive And Affective Priming Effects In TheContext For Print Advertisements," Journal ofAdvertising, 19 (2), 40-48.

References

Yoon, Kak, Russell N. Laczniak, Darrell D. Muehling, and Bonnie B. Reece (1995), "A Revised Model ofAdvertising Processing: Extending the DualMediation Hypothesis," Journal ofCurrent Issues and Research in Advertising, 17 (Fall), 53-67.

AuthorAffiliation

Barbara A. Lafferty TheUniversity ofSouth Florida

AuthorAffiliation

Ronald E. Goldsmith Florida State University

AuthorAffiliation

Stephen J. Newell Western Michigan University

AuthorAffiliation

AUTHOR BIOGRAPHY

AuthorAffiliation

Barbara A. Lafferty (Ph.D., Florida State University) is an assistant professor ofmarketing in theCollege ofBusiness Administration at theUniversity ofSouth Florida. Her research interests are in theareas ofsource credibility, corporatereputation, and cause-brand alliances. She has published articles in such journals as theJournal ofAdvertising, Journal ofBusiness Research, European Journal ofMarketing, CorporateReputation Review, and theJournal ofGlobal Marketing. Her work also appears in numerous conference proceedings.

AuthorAffiliation

AUTHOR BIOGRAPHY

AuthorAffiliation

Ronald E. Goldsmith (Ph.D., University ofAlabama in Tuscaloosa) is currently theRichard Baker professor ofmarketing at theFlorida State University. His articles have appeared in theJournal of theAcademy ofMarketing Science, theJournal ofAdvertising, theJournal ofBusiness Research, theService Industries Journal, theJournal ofRetailing, Marketing Intelligence and Planning, theEuropean Journal ofMarketing, TheJournal ofMarketing Theory and Practice and others. His research interests include developing measures ofconsumer behavior constructs, diffusion ofinnovations, and personality in consumer behavior.

AuthorAffiliation

AUTHOR BIOGRAPHY

AuthorAffiliation

Stephen J. Newell (Ph.D., Florida State University) is an associate professor ofmarketing in theHaworth College ofBusiness at Western Michigan University. His research interests include corporate credibility, advertisement recall and buyers' perceptions ofsales people. His articles have been published a number ofjournals including theJournal ofAdvertising, Journal ofBusiness Research, Journal ofMarketing Theory and Practice, Journal ofMarketing Education, and theJournal ofConsumer Affairs. He has also published in various conference proceedings.