Is beauty best? Highs versus normally attractive models in advertising Essay

Abstract (Summary) Several studies investigating the positive effects of including highly attractive models (HAMs) in advertising have failed to unilaterally support their use. A study explores the differential effects of pairing highly versus normally attractive models with different types of attractiveness-relevant products. Contrary to past research, the results suggest that HAMs are not the most effective choice for all categories of attractiveness-relevant products. The research also explores the method by which the match between model attractiveness and product type influences advertising effectiveness. Results suggest that a match between a model and a product improves ad effectiveness not necessarily through the elicitation of product arguments from model appearance, but instead by heightening perceptions of the model’s expertise about the product.  » Jump to indexing (document details) Full Text (7266 words) Copyright American Academy of Advertising Spring 2001 [Headnote] Several studies investigating the positive effects of including highly attractive models (HAMs) in advertising have failed to unilaterally support their use. This paper explores the differential effects of pairing highly versus normally attractive models with different types of attractiveness-relevant products. Contrary to past research (Kahle and Homer 1985; Kamins 1990), the results suggest that HAMS are not the most effective choice for all categories of attractiveness-relevant products. This research also explores the method by which the match between model attractiveness and product type influences advertising effectiveness. Results suggest that a match between a model and a product improves ad effectiveness not necessarily through the elicitation of product arguments from model appearance, but instead by heightening perceptions of the model’s expertise about the product. The inconsistent support for the use of highly attractive models (HAMs) in advertising has led one group of researchers to state that â€Å"physical attractiveness seems to have been granted greater influence than can be supported empirically† (Caballero, Lumpkin, and Madden 1989, p. 21). To better understand these inconsistencies, researchers have explored the importance of a convergence between the product and the message communicated by a model’s image, that is, a model-product type match-up (e. g. , Baker and Churchill 1977; Caballero and Solomon 1984; Joseph 1982; Kahle and Homer 1985; Kamins 1990; Park and Young 1986; Peterson and Kerin 1977). Although prior research has argued that HAMs are most effectively matched with attractiveness-relevant products (e. g., Kahle and Homer 1985; Kamins 1990; Peterson and Kerin 1977), the mixed results from HAM-attractiveness-relevant product match-ups may be because there are different types of attractiveness-relevant products (Bloch and Richins 1992) that may not all be appropriate for use with highly attractive people. Furthermore, much of the past model-product type match-up research has tended to compare HAMs with unattractive models (e. g. , Caballero and Solomon 1984; Kahle and Homer 1985; Kamins 1990) instead of considering more realistic, normally attractive models (NAMs) as counterpoints. The sparse use of unattractive people in advertising (Caballero and Solomon 1984) and the greater usage of NAMs suggest comparing HAMs with NAMs is more ecologically valid. The purpose of the present research is two-fold. First, the different attractiveness-relevant product types (Bloch and Richins 1992) are incorporated into the model-product type match-up literature, and the types of attractiveness-relevant products that are most effectively paired with HAMs or NAMs are delineated. Because beautiful people may be perceived as having better lives that are free of the problems of normal people (Dion, Berscheid, and Walster 1972; Kanner 1994), advertisements for problem-solving attractiveness-relevant products may be more effective if the models are normal looking. Second, this research attempts to determine whether model-product type match-ups influence ad effectiveness either directly, through the elicitation of product arguments from the picture, or indirectly, through his or her perceived credibility. Previous research has explored the effect of model-product type match-up on perceptions of model credibility (Kamins 1990; Maddux and Rogers 1980), product evaluations, purchase intentions, or other measures of ad effectiveness (Caballero, Lumpkin, and Madden 1989; Kahle and Homer 1985; Kamins 1990). However, the relationship between spokesperson credibility and ad effectiveness (i. e. , product evaluations, purchase intentions) has not been explored under match-up conditions. Literature Review Highly Versus Normally Attractive Models The appearances of HAMs are both idealized and unrealistic and have been called â€Å"haunting images of perfection† (Richins 1991, p. 71). The elements of HAM beauty include a beautiful facial appearance (Richins 1991), as well as thinness (Striegel-Moore, Silberstein, and Rodin 1986). Contrary to these icons of flawlessness, â€Å"normal† attractiveness is defined here as a more average or moderate weight, height, and facial beauty, that is, more representative of a â€Å"real† woman. Normally attractive models are considered attractive but not beautiful in the idealized manner of HAMs. An important difference between HAMs and NAMs is in the attributions made about each. Highly attractive models tend to be associated with the â€Å"what is beautiful is good† stereotype, in that beautiful people are believed to have more positive life outcomes (e. g. , more successful careers, better marriages) and not suffer from the problems of â€Å"normal† people (Dion, Berscheid, and Walster 1972; Kanner 1994; Walster et al. 1966). Whereas NAMs are considered to be somewhat attractive, they are perceived to be normal people to whom these more positive life outcomes are not attributed. Problem-Solving Versus Enhancing Products Attractiveness-relevant product types are introduced into the match-up literature to delineate not only those that are best paired with HAMs, but also those that may be best paired with NAMs. A major distinction that can be drawn between types of attractiveness-relevant products is whether the product is associated with a potentially problematic area of life and appearance (Bloch and Richins 1992). One group of attractiveness-relevant products (called problem-solving products) serves to fix or hide beauty liabilities or flaws such as acne or dandruff. Other products (called enhancing products) serve more aesthetic purposes by enhancing beauty (e. g. , jewelry, lipstick, perfume) instead of masking defects. Enhancing products either may be inherently beautiful (e. g. , jewelry) or may enhance beauty through their application to the user (e. g. , lipstick). Regardless of their method of enhancing beauty though, the use of enhancing products is not catalyzed by the existence of a beauty problem. A product’s classification depends on whether the product-related body attribute is perceived positively or negatively (e. g. , a consumer may use mascara either to thicken puny eyelashes or to enhance beautiful eyes). Advantages of Normally Attractive Models? Prior research suggests that the match-up between product type and model beauty may be more important than the models attractiveness alone and that a NAM may sometimes be more effective. Kahle and Homer (1985) demonstrate that a â€Å"well-matched† celebrity endorser for an attractiveness-relevant product should be physically attractive (versus unattractive) to convey information about the quality and benefits of that product. Kamins (1990) findings also indicate that physically attractive (versus unattractive) celebrities are best matched with attractiveness-relevant (versus irrelevant) products. Although these findings do not take into account the advantages of NAMs or recognize the important distinctions between different kinds of attractiveness-relevant product types, both Kahle and Homer (1985) and Kamins (1990) support the importance of matching spokesperson image with product characteristics. Similarly, Baker and Churchill (1977) find that when the product is related to romance (e. g., perfume), men had higher purchase intentions when the female model was attractive. However, when the product was unrelated to romance (e. g. , coffee), male subjects indicated greater purchase intent if the model was less attractive. In a study by Caballero and Solomon (1984), more tissues were purchased when they were paired with an unattractive (versus attractive) model. Although the authors proposed that this response was due to the unusual (for advertising) and perhaps more noticeable nature of the model, an alternative explanation may be that an unattractive appearance may be more convergent with sickness. Consumers may not perceive the HAM as suffering from such mundane problems as an illness that requires tissues, whereas the less attractive model’s more realistic appearance may have made her a more credible spokesperson. HAM Versus NAM Match- Ups Some of the research investigating model-product type match-ups has assessed match-up effectiveness by measuring spokesperson credibility (Kamins 1990; Maddux and Rogers 1980). Two components of spokesperson credibility typically recognized are source expertise and trustworthiness. Expertise refers to the perceived ability of a source to make valid claims, whereas trustworthiness pertains to the believed willingness of the spokesperson to make those claims (Hovland, Janis, and Kelley 1953; Ohanian 1990). Model Attractiveness and Trustworthiness. There is evidence to suggest that NAMs may be perceived as more trustworthy than HAMs because of consumers’ perceived similarity between themselves and the NAMs. Deshpande and Stayman (1994) demonstrate that enhanced identification and similarity with a source can increase perceptions of source trustworthiness. If a woman perceives herself to be physically more similar to a NAM, a greater perceived attitudinal similarity between the receiver and the source may drive the receiver to like the source more (O’Keefe 1990; see also Berscheid 1985; Byrne 1969). A greater liking for the source may influence source trustworthiness positively (e. g. , O’Keefe 1990; Simons, Berkowitz, and Moyer 1970), which may in turn influence the consumer to like and trust the source more. Therefore, H1: NAMs are perceived to be more trustworthy than HAMs. There is no hypothesized relationship or interaction between product type and model condition on trustworthiness. Trustworthiness pertains to a personality trait of the endorser, regardless of the endorsed product. Model Attractiveness-Product Type Match-Up and Model Expertise. Models may be more credible if they either have some physical characteristic that is associated with a product schema or demonstrate some characteristic that indicates that the product has accomplished what it claims, which thus suggests expertise based on experience (Lynch and Schuler 1994). Stemming from their attributions about beautiful people, consumers may believe that HAMs know more about the presentation and elements of a beautiful image, including the kinds of products used to enhance beauty. In addition, HAMs may be perceived as having more experience accentuating beautiful features or surrounding themselves with aesthetic things. Thus, H2: HAMs are perceived to have greater source expertise for enhancing products than are NAMs. Highly attractive models may not be perceived as having the problems that purportedly are solved by problem-solving products (e.g. , Dion, Berscheid, and Walster 1972; Kanner 1994) and therefore are believed to have little expertise using such products. Alternatively, consumers may be more likely to believe that a NAM has struggled with and, as evidenced in the ad, â€Å"conquered† the problem using the advertised product. Therefore, NAMs may be more congruous with the schema associated with enhancing products (Lynch and Schuler 1994). H3: NAMs are perceived to have greater source expertise for problem-solving products than are HAMs. Model Attractiveness-Product Type Match-Up and Product Evaluations. If a viewer can convert the visual imagery of an endorser into product information, the model’s image may serve as an argument for product efficacy (Bloch and Richins 1992; Downs and Harrison 1985; Kahle and Homer 1985; Lynch and Schuler 1994; see also Rossiter and Percy 1980). There is general evidence that product arguments may be elicited from pictures (e. g. , Mitchell and Olson 1981) and that, in high involvement situations, pictures containing product-relevant information can strengthen product beliefs and result in more positive product attitudes than can pictures that only influence affect (Miniard et al.1991). If a viewer believes that a model possesses some physical characteristic that indicates the model has improved his or her appearance with the product, then the viewer may believe that the product was responsible for that improvement (Lynch and Schuler 1994). The pairing of an enhancing product and a HAM may serve to reinforce the argument for product quality and efficacy for enhancing beauty by demonstrating the enhancer’s ability to highlight beautiful features (Petty and Cacioppo 1980). This stronger argument for product quality may enhance ad efficacy by improving evaluations of the product and heightening product purchase intentions (Caballero, Lumpkin, and Madden 1989; Kahle and Homer 1985; Kamins 1990). Therefore, 114: Ads for enhancing products that include HAMs will be more effective than those that include NAMs. Specifically, we expect that ads for enhancing products that include HAMs will result in greater product evaluations and higher purchase intentions than will those ads containing NAMs. With regard to problem-solving products, it might initially appear that a HAM could serve as a more positive argument for a product’s effectiveness. The more beautiful appearance of the HAM (versus NAM) might make the product appear as if it is more effective in heightening the user’s beauty. However, for the efficacy of problem-solving products to be assessed from the model’s photo, the viewer must first believe that the model had a beauty problem prior to product use. Because HAMs are not perceived as having the problems of normal people (e. g., Dion, Berscheid, and Walster 1972; Kanner 1994), consumers may not believe that the HAM suffered from a beauty problem that required the use of the advertised product. In contrast, consumers may believe that NAMs are more likely to have experienced beauty problems (e. g. , Dion, Berscheid, and Walster 1972; Kanner 1994). According to attribution theory, if an outcome occurs despite the presence of something that might have prevented that outcome (an inhibitory cause), a consumer may give more weight to the cause that is perceived to be responsible for that outcome (the facilitative cause; Kelley 1972). A HAM’s overall appearance might suggest multiple facilitative causes for the model’s superior appearance in the productrelated attribute, and the greater the number of other potential causes for the overall superior appearance, the more the problem-solving product may be discounted as responsible for solving the beauty problem (Kelley 1972). However, a NAMs’ overall typical appearance might suggest that, prior to product use, the appearance of the NAM’s product-related attribute was normal. Because the product (facilitative cause) may be responsible for a beautiful product-related physical attribute in the face of the inhibitory cause of the model’s â€Å"normalcy,† the perceived efficacy of the problem-solving product is augmented. With this improvement in product evaluations and likely improvement in purchase intentions (Caballero, Lumpkin, and Madden 1989; Kahle and Homer 1985; Kamins 1990), the following hypothesis is offered: H5: Ads for problem-solving products that include NAMs will be more effective than those that include HAMs. Specifically, we expect that ads for problem-solving products that include NAMs will result in greater product evaluations and higher purchase intentions than will those ads containing HAMs. The Mediating Role of Source Credibility Although model-product type match-up effectiveness has measured both spokesperson credibility and other measures of ad effectiveness, such as product evaluations or purchase intentions, no relationship between these two constructs has been examined. Past credibility research has suggested that greater spokesperson credibility (i. e., expertise and trustworthiness) tends to generate greater attitude chane (e. g. , Dholakia and Sternthal 1977; Harman and Coney 1982; Hovland, Janis, and Kelley 1953; Sternthal, Dholakia, and Leavitt 1978). Therefore, any influence that a model-product type match-up may have on spokesperson credibility subsequently might offect product evaluations or other measures of ad effectiveness. Therefore, H6A: Spokesperson expertise will mediate the effects of model-product match-up on ad effectiveness. H6B: Spokesperson trustworthiness will mediate the effects of model beauty on ad effectiveness. Study I Pretests. Two pretests were conducted to select product and model stimuli. The purpose of the first pretest (n=25) was to select two products to represent enhancing products and two to represent problem-solving products. Subjects were asked to categorize a list of products according to the provided definitions of each product type. The two products selected to represent the problem-solving category were acne concealer and acne medicine because of their categorization as problem-solving products (100% and 92%, respectively). With regard to enhancing products, 88% of people categorized lipstick and jewelry as enhancers. Because acne medicine, acne concealer, and lipstick all pertain to the face, the jewelry selected for use as stimuli in this experiment was earrings (i. e. , the only widely accepted jewelry associated with the face). For the second pretest, two judges selected full-color model photographs from popular women’s magazines on the basis of subjective criteria of which photos might represent each model condition. Photos intended to represent NAMs were selected from â€Å"Reader Makeover† issues of the magazines so that, while the physical features of the NAMs were still moderately attractive, the hair and makeup were professionally styled. Two undergraduate classes viewed five photos of NAMs (n=72) while two other classes evaluated five photos of HAMs (n=65). Respondents were asked to complete five questions regarding the beauty and four items assessing the normalcy of the model (all seven-point scales), and each set of items subsequently was summed. The HAM photo was selected on the basis of beauty extremity (M=29. 37) and subjects’ beliefs that she was leading a less-than-normal life (M=13. 31). The NAM photo was selected on a rating of moderate beauty (M=22. 06; p