Coprate Advertising Impact On Brand

Marketing Letters 10:1 (1999): 63–73 © 1999 Kluwer Academic Publishers, Manufactured in The Netherlands

Corporate Advertising Pass-through onto the Brand: Some Experimental Evidence DANIEL A. SHEININ College of Business and Management, University of Maryland at College Park, R.H. Smith School of Business, 3454 Van Munching Hall, College Park, MD 20742, E-mail: [email protected] GABRIEL J. BIEHAL College of Business and Management, University of Maryland at College Park, R.H. Smith School of Business, 3447 Van Munching Hall, College Park, MD 20742, E-mail: [email protected]

Abstract While researchers and managers have speculated that corporate advertising can have an impact on brand knowledge, to date there is no empirical support for such “pass-through.” We report on a laboratory study manipulating the presence or absence in a brand ad of retrieval cues for subjects’ corporate advertising and their pre-existing brand knowledge. We found evidence of pass-through, but it was moderated by the retrieval of subjects’ pre-existing brand knowledge. Our results have theoretical and managerial implications, and suggest a number of interesting opportunities for future research. Key words: Corporate advertising, brand management

Corporate advertising, which communicates information about a company, represents over $9 billion in annual spending (Belch and Belch 1996). Designed primarily to build corporate image or increase investment (Javagli, Traylor, Gross and Lampman 1994; Schumann, Hathcote and West 1991), corporate advertising may also influence consumers’ knowledge about products marketed by the corporation (Hartigan and Finch 1986; Winkleman 1985). This influence is critical for brand managers to understand because they have the responsibility for building and managing consumers’ brand knowledge. While previous corporate ad research has focussed on classification schemes based on different messages and targets, and the implications for corporate image (Rothschild 1987; Schumann et al. 1991), it has not discussed in-depth or empirically examined possible “pass-through” to brand knowledge. Moreover, prior brand research has emphasized the use of brand advertising tactics without considering possible contributions from corporate advertising. In this paper, we propose pass-through occurs, but is moderated by the retrieval of consumers’ pre-existing brand knowledge. We report on empirical evidence confirming these propositions in a lab setting. In an experiment, to form corporate ad knowledge we first expose all subjects to the same corporate ad. Next, they see a brand ad differing in the presence or absence of retrieval cues for knowledge about both the corporate ad and an already known brand. Product beliefs and evaluations are then examined for pass-through.

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 1 SESSION: 11

SHEININ, BIEHAL

64

In the next section, we discuss the experiment’s theoretical background and present research propositions. After describing our results, we discuss theoretical and managerial implications, and future research directions. 1. Conceptual development 1.1 Theoretical background In general, research shows exposure to an ad leads consumers to form knowledge about it, which is stored in memory (MacKenzie, Lutz and Belch 1986). Thus, consumers form knowledge about corporate ads following exposure to them, which is stored as part of their corporate knowledge. This process is consistent with corporate advertising’s traditional purpose to build and maintain a positive corporate image (Rothschild 1987; Schumann et al. 1991). In addition to corporate ad knowledge, consumers have brand knowledge stored in memory. Both corporate ad and brand knowledge contain beliefs, affect, and attitudes (Garbett 1983; MacKenzie et al. 1986; Keller 1993), all of which may have an impact on product evaluations. Typically, consumers’ product evaluations are based on a combination of stimulus information and knowledge retrieved from memory (Lynch and Srull 1982; Alba and Hutchinson 1987). Knowledge retrieval, in turn, is influenced by retrieval cues, like names or identifying labels. Such cues activate knowledge in memory, leading to its retrieval (Keller 1987). Corporate ad and brand knowledge retrieval cues may thus influence product evaluations by causing consumers to retrieve and use certain parts of their knowledge during product processing. Retrieved corporate ad knowledge may pass through and influence product evaluations in a number of ways. For example, consumers may use their retrieved corporate ad knowledge to infer new product beliefs. New beliefs are inferred using retrieved knowledge (Fiske and Pavelchak 1986). For example, if consumers see a product ad from a company which they recall advertises its environmental concerns, they may infer the company’s products are environmentally safe. Such inferred beliefs therefore originate in corporate ad knowledge but become associated with advertised products. Consumers may also change existing product beliefs using retrieved corporate ad knowledge. Existing beliefs can change through processing new, relevant information (Crocker, Fiske and Taylor 1984). For example, if consumers are exposed to a product ad from a company which they recall does corporate advertising on quality issues, they may strengthen existing product quality beliefs. Finally, if consumers either infer new product beliefs or change existing ones, they may change their existing product attitudes. 1.2 Research propositions Within this general framework, a lack of empirical work leaves unanswered two important questions: (1) can corporate ad knowledge pass-through onto consumers’ knowledge of products marketed by the corporation? and (2) what factors moderate pass-through?

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 2 SESSION: 11

CORPORATE ADVERTISING PASS-THROUGH ONTO THE BRAND

65

To address the first question we start with a product ad without brand retrieval cues, and compare product evaluations with and without corporate ad retrieval cues. This provides a straightforward test of pass-through. In this situation, we expect corporate ad retrieval cues, like the corporate name, will cause consumers to have different product knowledge. As previously indicated, consumers make product evaluations in the context of their retrieved knowledge (Lynch and Srull 1982). Thus, using the processes previously described, corporate ad knowledge retrieval cues should lead to greater pass-through. The second question concerns moderators of pass-through. Since much research shows information processing is influenced by the retrieval of consumers’ pre-existing knowledge (Alba, Hutchinson and Lynch 1991), it seems likely that pass-through will be similarly affected. Therefore, we now consider a product ad with brand retrieval cues, like the brand’s name. If the comparative impact of the presence and absence of corporate ad retrieval cues differs when brand retrieval cues are present versus absent, we can conclude pass-through is moderated by the retrieval of consumers’ brand knowledge. In this situation, we expect no change in product evaluations regardless of corporate ad retrieval cues. Although such cues may still help consumers retrieve their corporate ad knowledge, for two reasons they are less likely to use it to make product evaluations. First, consumers are less likely to use a particular knowledge source if several sources have been retrieved from memory (Alba et al. 1991). Second, since short-term memory has limited capacity, consumers retrieving multiple knowledge sources tend to use the most diagnostic one and place less emphasis on or completely disregard others (Lynch, Marmorstein and Weigold 1988). As corporate ad knowledge should be less diagnostic for product evaluation than brand knowledge and ad information, product evaluation should be unaffected by corporate ad retrieval cues in the presence of brand retrieval cues. Therefore, the retrieval of pre-existing brand knowledge should moderate corporate ad knowledge passthrough.

2. Method 2.1 Pretests To guide stimulus design, we conducted two pretests using a number of open-ended and scaled questions. The first (n ⫽ 82) identified the attributes subjects associated with “good” and “bad” corporations, and involving product categories. The following corporate attributes were mentioned by at least 33% of subjects: financial success, environmental friendliness, extent of recycling, management quality, brand quality, diversity and undergraduate hiring. These attributes were used to design the corporate ad and as dependent measures. Finally, this pretest identified four, highly-involving product categories: hiking shoes, athletic shoes, blue jeans, and backpacks. The second pretest (n ⫽ 73) established perceptions of and attitudes toward the four product categories, selected brands within them, and possible corporate names. Based on the results we selected the hiking shoe category. Subjects had the following strong and consistent category beliefs about hiking shoes: rugged, available in a variety of colors,

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 3 SESSION: 11

SHEININ, BIEHAL

66

durable, waterproof, has extra support, fashionable, and stylish. These product category attributes were used to design a brand ad which would contain the retrieval cue manipulations. Our brand name selection balanced two considerations. On the one hand, a brand for which subjects had strong beliefs and attitudes, which are difficult to change (Crocker et al. 1984), would be unlikely to show pass-through. On the other hand, the moderating hypothesis assumes subjects have some pre-existing brand knowledge. The “New Balance” brand name met our design needs. Pretests showed that, compared with other tested hiking shoe brands, like Nike and Timberland, subjects were moderately familiar with New Balance, had neutral attitudes toward it, and did not know the corporate manufacturer. Subjects associated the brand with good quality, design and comfort, which were used as dependent measures. Finally, because we wanted subjects’ corporate knowledge to stem exclusively from exposure to the corporate ad, we evaluated six fictitious company names. All had some a priori reasonable link with the tested product category, but no expected recognition as a corporate name. Therefore, we chose “Shenandoah” as the corporate name. Detailed pretest results are available from the authors.

2.2 Ad stimuli, retrieval manipulations and design The corporate ad headline was, “Young Managers Are Our Future.” The picture, taken from a real ad, was a 4x2.5 inch color photograph, centered on an 11x8.5 inch page. It showed three young professionals, an Asian woman, and African-American and Caucasian males. The copy described the corporation as follows: “At the Shenandoah Company our young designers are leading us into the global business future. Each year, we hire hundreds of top college graduates in business and engineering. They are responsible for developing 17 products which received quality awards from leading consumer magazines in 1994. In addition to hiring many graduates, we provide internships for college students in environmental studies. Our interns receive college credit while they help us develop products which can be completely recycled. We are proud to have received a Presidential Achievement Award for our efforts in environmental improvement. Our young talent is the main reason we have grown 25% a year to reach $1 billion in global sales of outdoor products.” Finally, the Shenandoah company name was placed at the bottom of the ad. Because they are central in knowledge representations (Keller 1993; Mitchell and Dacin 1996), we used the company and brand names as retrieval cues. Thus, the Shenandoah company name was placed at the bottom of the brand ad in the corporation ad retrieval cue present conditions, and the New Balance name was inserted in the brand ad copy in the brand retrieval cue present conditions. The brand ad headline was, “The Look of Right Now.” Two pictures showed individuals engaged in rugged outdoor activities. The third showed the hiking shoes, which had no brand identification on them. Finally, the copy was: “Our [New Balance] shoes are the hip-hop way to climb a mountain, a trail and a campus. [New Balance]. Go anywhere. Do anything.” Note the corporate ad information was not directly related to product quality.

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 4 SESSION: 11

CORPORATE ADVERTISING PASS-THROUGH ONTO THE BRAND

67

Thus, the study was a 2 (corporate ad retrieval cue present/absent) x 2 (brand retrieval cue present/absent) between-subjects design. The absent/absent condition represented the baseline to test for pass-through. 2.3 Subjects Subjects were undergraduate juniors and seniors at a large Atlantic Region state university who received extra class credit as a participation incentive. The original sample size was 182, but 20 subjects were eliminated because they were unfamiliar with the New Balance brand name. Since these subjects did not have pre-existing brand knowledge, retrieval cues would be ineffective. Thus, the final sample size was 162. 2.4 Experimental procedure Subjects were tested in groups of about fifteen in a 30-minute session. They received one questionnaire booklet which they filled out individually. Subjects first read general instructions on using the booklet, then received task instructions. They first removed the corporate ad from an envelope, and were instructed to “...look at the ad carefully and thoroughly, much as you would if you saw an ad that interested you. As you examine the ad you should try to relate the ad to your own personal attitudes, lifestyle and concerns.” Subjects could take as much time as they liked on the task. Upon completion, they returned the ad to the envelope, and then wrote down their corporate ad thoughts. Next, they filled out nine company belief items. Finally, they recorded their attitudes toward the company and the corporate ad. To clear short term memory, a simple distractor task followed. Subjects then read instructions about the brand ad task, which were identical to those used previously. Next, they removed the brand ad from an envelope, examined it, and, after returning it to the envelope, wrote down their brand ad thoughts. They then rated seventeen beliefs which could originate in four sources, namely the corporate ad, the brand ad, and their pre-existing brand and product category knowledge. Subjects then indicated their attitudes toward the brand and brand ad. Next, they completed questions on brand and category familiarity, product category usage and involvement level, demographic information, and the importance of the corporate attributes described in the corporate ad. They were also asked to guess the research purpose (none did). Finally, subjects were fully debriefed, either in-class or upon completing the questionnaire. 2.5 Measures Attitude toward the brand, AB, was the sum of three seven-point semantic differential scales anchored by dislike/like, bad/good, and negative/positive. Attitudes toward the brand ad, corporate ad, and the corporation (ABad, ACad, and AC, respectively) used the same scales.

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 5 SESSION: 11

SHEININ, BIEHAL

68

We measured beliefs in two ways. First, nine corporate and seventeen brand beliefs were measured using seven-point semantic differential scales, anchored by unlikely/likely. Second, corporate ad and brand ad beliefs, used for structural modelling, were derived from subjects’ thought protocols. Two independent coders trained by the authors coded the protocols for belief valence (positive, negative and neutral). A log of coding discrepancies indicated a 98% agreement between the two coders. To control for individual differences in the number of thoughts, two indices were formed by dividing a subjects’s total number of positive corporate (brand) ad thoughts by total corporate (brand) ad thoughts.

3. Results 3.1 Preliminary factor analyses We performed three factor analyses with varimax rotation to refine and check measures. The first used the nine corporate beliefs, and the resulting Scree Test revealed a singlefactor solution accounting for 40% of variance. The four beliefs underlying the factor were summed to develop the corporate beliefs measure, BC (Cronbach alpha ⫽ 0.84). The second factor analysis of the seventeen brand beliefs revealed a three-factor solution accounting for 60% of variance. The factors were interpreted as representing brand quality (Cronbach alpha ⫽ 0.84), brand image (Cronbach alpha ⫽ 0.85) and corporate-ad derived beliefs, that is, beliefs originating in the corporate ad subjects associated with the brand (Cronbach alpha ⫽ 0.71). Finally, a confirmatory factor analysis to check the validity of the four attitude constructs showed the constructs were valid and independent (AGFI ⫽ .869, RMR ⫽ .068, p ⬍ .0001).

3.2 Data Analysis First, task involvement, category familiarity, category involvement, brand familiarity, brand interest, and brand knowledge were examined as covariates in 2 (corporate ad retrieval cue) x 2 (brand retrieval cue) ANCOVAs. None were significant. Then, to test the research propositions, 2 (corporate ad retrieval) x 2 (brand retrieval cue) ANOVAs were run with different dependent variables. Finally, to explore the processes underlying AB formation, we estimated structural equations models. With brand quality as the dependent variable there was a significant interaction effect (F3,158 ⫽ 9.81; p ⬍ 0.01; ␻2 ⫽ .04; see Figure 1, Graph A). With New Balance absent, brand quality was stronger when Shenandoah was present (M ⫽ 5.40) versus absent (M ⫽ 4.61; t39 ⫽ 2.72; p ⬍ 0.01). However, with New Balance present, brand quality was the same when Shenandoah was present (M ⫽ 4.92) and absent (M ⫽ 5.29; p ⬎ 0.10). This general pattern of results was found with the other measures.

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 6 SESSION: 11

CORPORATE ADVERTISING PASS-THROUGH ONTO THE BRAND

69

Figure 1. The three significant interactions.

With brand image as the dependent variable the interaction effect was weaker (F3,158 ⫽ 3.39; p ⬍ 0.07; ␻2 ⫽ .02). With New Balance absent, brand image was stronger with Shenandoah present (M ⫽ 5.04) versus absent (M ⫽ 4.16; t39⫽ 2.49; p ⬍ 0.05. However,

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 7 SESSION: 11

SHEININ, BIEHAL

70

with New Balance present, brand image was the same when Shenandoah was present (M ⫽ 4.72) and absent (M ⫽ 4.71; p ⬎ 0.90). When corporate-ad derived brand beliefs were the dependent variable there was also a significant interaction effect (F3,158 ⫽ 5.12; p ⬍ 0.05; ␻2 ⫽ .02; see Figure 1, Graph B). Without the New Balance cue, corporate-ad derived beliefs were stronger with the Shenandoah cue present (M ⫽ 4.51) versus absent (M ⫽ 3.77; t39 ⫽ 2.10; p ⬍ 0.05). However, with the New Balance cue these beliefs were the same when Shenandoah was present (M ⫽ 3.42) and absent (M ⫽ 3.70; p ⬎ 0.30). Finally, with AB as the dependent variable there was a significant interaction effect (F3,158 ⫽ 7.01; p ⬍ 0.01; ␻2 ⫽ .03; see Figure 1, Graph C). With New Balance absent, AB was more positive with Shenandoah present (M ⫽ 4.71) versus absent (M ⫽ 3.53; t39 ⫽ 3.83; p ⬍ 0.001). However, with New Balance present, AB was the same when Shenandoah was present (M ⫽ 4.24) and absent (M ⫽ 4.27; p ⬎ 0.90). All these results are consistent with the proposition that pass-through occurs, but is moderated by the retrieval of subjects’ pre-existing brand knowledge. One reviewer suggested examining the potential moderating effect of brand familiarity on pass-through in the corporate ad and brand retrieval cues present condition. Familiarity was a significant moderator on brand attitude (p ⬍ .05), but not on brand quality, brand image, or corporate ad-derived beliefs (each p ⬎ .20). Thus, at least on brand attitude, corporate ad pass-through was significantly less when subjects were more familiar with the brand. Next, to examine the processes underlying pass-through, structural equations were estimated for each experimental condition. Guided by previous research (MacKenzie et al. 1986), the models included relationships among corporate ad beliefs (BCad) and attitude (ACad), corporate beliefs (BC) and attitude (AC), brand ad beliefs (BBad) and attitude (ABad), and brand quality (BB) and attitude (AB). BCad and BBad were derived from subjects’ thought protocols. We also ran structural equations using brand image instead of brand quality. Because the results are similar, we report only the relationships specified above. The model for subjects exposed to brand ads with Shenandoah but without New Balance retrieval cues had good fit (AGFI ⫽ 0.99, RMR ⫽ 0.0008, p ⬍ 0.21). Overall, there are two interesting findings (see Figure 2, Model A). First, ACad showed a dual-mediation effect of BCad on both BC and AC. This finding is consistent with the literature on brand advertising impacts (MacKenzie et al. 1986). Second, pass-through was found at both the belief and the attitudinal levels (BC was significantly related to BB, and ACad and AC were both significantly related to AB, respectively). The model for subjects exposed to brand ads containing both Shenandoah and New Balance cues had moderate fit (AGFI ⫽ 0.81, RMR ⫽ 0.0060, p ⬍ 0.69). The results (Figure 2, Model B) also show a dual-mediation effect of ACad on BC and AC. However, there was no evidence of corporate advertising pass-through onto either brand beliefs or attitudes. Instead, only BB significantly explained AB. The results for the remaining two conditions are essentially the same, that is, there was no evidence of pass-through, and we do not describe them further.

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 8 SESSION: 11

CORPORATE ADVERTISING PASS-THROUGH ONTO THE BRAND

71

Figure 2. Structural equations analysis.

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 9 SESSION: 11

SHEININ, BIEHAL

72 4. Discussion

The experiment demonstrates that corporate ad pass-through to brand knowledge can occur, but is moderated by the retrieval of consumers’ pre-existing brand knowledge. Further, even though corporate ad information was not directly related to brand quality, when pass-through occurred both brand beliefs and attitudes were affected. By providing empirical validation of pass-through, our research confirms managers’ and researchers’ intuitions (Hartigan and Finch 1986; Winkleman 1985) and extends the corporate advertising literature. It also extends the brand literature, by broadening the scope of ad processing models and giving additional insights into the importance of advertising memory on consumers’ brand information processing. Our results have important implications for brand managers. If brand managers want to take advantage of positive corporate ad knowledge, they should consider using certain tactics, like adding corporate ad symbols or slogans to brand ads, or including the corporate name prominently in their product names. Such tactics may help consumers transfer their corporate ad knowledge onto the brand. Also, they should measure corporate ad knowledge in Tracking Studies to understand its role in brand ad effectiveness and brand knowledge change. The significance of the results reported here depends on a number of limitations common to much academic advertising research designed to understand fundamental processing phenomena, like student samples and a single, forced ad exposure. Other study limitations suggest interesting avenues for future research. We did not measure evaluations of corporate and brand names, but different relative evaluations may moderate pass-through. For example, extremely negative evaluations of the corporate name may pass through regardless of brand knowledge retrieval. Future research should examine how differences in evaluations of corporation and brand names moderate pass-through. In addition, we manipulate brand knowledge retrieval, but not brand knowledge itself. We used New Balance, a moderately familiar brand, but future research might show pass-through differs greatly for familiar versus unknown or newly-introduced brands. This would be consistent with our finding that familiarity moderated pass-through onto brand attitude. Understanding such moderating effects would help guide managers’ use of corporate advertising, for example, to prepare markets for new product launches (Garbett 1983). Finally, we exposed all subjects to the same corporate ad. In reality, corporate advertising uses a variety of messages and creative strategies (Garbett 1983). Future research may show pass-through is influenced by the tactics employed in corporate advertising. For example, 3M’s corporate advertising illustrates its innovative positioning using a picture of its Post It Notes product, a tactic which may encourage pass-through. In sum, we believe research into corporate advertising pass-through has much to offer both managers and academic researchers alike. Acknowledgments The authors, who contributed equally to this research, thank Professor Richard M. Durand and two anonymous reviewers for their comments on an earlier draft.

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 10

SESSION: 11

CORPORATE ADVERTISING PASS-THROUGH ONTO THE BRAND

73

References Alba, Joseph W. and J. Wesley Hutchinson. (1987). “Dimensions of Consumer Expertise,” Journal of Consumer Research 13 (March), 411–54. Lynch, Jr., ______, ______, and John G. (1991). “Memory and Decision Making,” in The Handbook of Consumer Behavior, Thomas S. Robertson and Harold H. Kassarjian editors, Prentice Hall, Englewood Cliffs, NJ, 1–49. Crocker, Jennifer, Susan T. Fiske and Shelley Taylor. (1984). “Schematic Bases of Belief Change,” in Attitudinal Judgment, ed. J.R. Eiser, New York: Springer-Verlag, 197–226. Fiske, Susan T. and Mark A. Pavelchak. (1986). “Category-Based Versus Piecemeal-Based Affective Responses: Developments in Schema-triggered Affect,” in The Handbook of Motivation and Cognition: Foundations of Social Behavior, eds. Richard M. Sorrentino and E. Tory Higgins, New York: Guilford, 167–203. Garbett, Thomas F. (1983). “Researching Corporate Advertising,” Journal of Advertising Research 23, 1, February/March, 33–37. Hartigan, Maureen F. and Peter Finch. (1986). “The New Emphasis on Strategy in Corporate Advertising,” Business Marketing (February), 42–49. Heath, Robert L. and Richard Alan Nelson. (1985). “Image and Issue Advertising: A Corporate and Public Policy Perspective,” Journal of Marketing 49, 58–68. Javagli, Rajshekhar, Mark B. Taylor, Andrew C. Gross and Edward Lampman. (1994). “Awareness of Sponsorship and Corporate Image: An Empirical Investigation,” Journal of Advertising 23, (December), 47–58. Keller, Kevin Lane. (1993). “Conceptualizing, Measuring, and Managing Customer-Based Brand Equity,” Journal of Marketing 57 (January), 1–22. ______. (1987). “Memory Factors in Advertising: The Effect of Advertising Retrieval Cues on Brand Evaluations,” Journal of Consumer Research 14 (December), 316–333. Lynch, John G. and Thomas K. Srull. (1982). “Memory and Attentional Factors in Consumer Choice: Concepts and Research Methods,” Journal of Consumer Research 9, 18–37. ______, Howard Marmorstein and Michael F. Weigold. (1988). “Choices from Sets Including Remembered Brands: Use of Recalled Attributes and Prior Overall Evaluations,” Journal of Consumer Research 15 (September), 169–84. MacKenzie, Scott B., Richard Lutz, and George E. Belch. (1986). “The Role of Attitude Toward the Ad as a Mediator of Advertising Effectiveness: A Test of Competing Hypotheses,” Journal of Marketing Research 23, (May), 130–143. Mitchell, Andrew A. and Peter A. Dacin. (1996). “The Assessment of Alternative Measures of Consumer Expertise,” Journal of Consumer Research Volume 23, 3 (December), 219–39. Rothschild, Michael L. (1987). Advertising, Massachusetts: Heath and Company. Schumann, David M., Jan M. Hathcote, and Susan West. (1991). “Corporate Advertising in America: A Review of Published Studies on Use, Measurement, and Effectiveness,” Journal of Advertising 20 (September), 35–56. Winkleman, Michael. (1985). “Corporate Advertising,” Public Relations Journal 41 (December), 38–39.

Kluwer Journal @ats-ss11/data11/kluwer/journals/mark/v10n1art5

COMPOSED: 11/20/98 8:43 am.

PG.POS. 11 SESSION: 11