When Retailers Target Women Based on Body Shape and Size: The Role of Ethical Evaluation on Purchase Intention

Abstract

The purpose of this research was to assess consumers' ethical evaluation of a company's practice of targeting women based on body shape and size in the clothing retail industry and its impact on the consumer's purchase intention. Many companies target those consumers they have the greatest chance of satisfying, leading to greater profit potential for the firm. The sophistication of target marketing helps marketers achieve efficiency and effectiveness. However, targeting may be an ineffective strategy when it results in controversy. The ethical dilemma of targeting lies within the explicit inclusion or exclusion of groups of customers. Under the framework of Theory of Planned Behavior (TPB) and the Multidimensional Ethics Scale (MES), this research incorporated ethical evaluation into TPB. This study found that consumers evaluate retailers who target women based on body shape and size as less ethical than retailers who do not target women based on a body shape and size. However, the consumer's unethical evaluation of the company did not divert the consumer's intention to buy the company's products. The results demonstrated that when retailers target women based on body shape and size, ethical evaluation significantly influences attitude and subjective norm, but does not significantly influence purchase intention.

Headnote

ABSTRACT:

The purpose of this research was to assess consumers' ethical evaluation of a company's practice of targeting women based on body shape and size in the clothing retail industry and its impact on the consumer's purchase intention. Many companies target those consumers they have the greatest chance of satisfying, leading to greater profit potential for the firm. The sophistication of target marketing helps marketers achieve efficiency and effectiveness. However, targeting may be an ineffective strategy when it results in controversy. The ethical dilemma of targeting lies within the explicit inclusion or exclusion of groups of customers. Under the framework of Theory of Planned Behavior (TPB) and the Multidimensional Ethics Scale (MES), this research incorporated ethical evaluation into TPB.

This study found that consumers evaluate retailers who target women based on body shape and size as less ethical than retailers who do not target women based on a body shape and size. However, the consumer's unethical evaluation of the company did not divert the consumer's intention to buy the company's products. The results demonstrated that when retailers target women based on body shape and size, ethical evaluation significantly influences attitude and subjective norm, but does not significantly influence purchase intention.

KEYWORDS:

Marketing, ethics, theory of planned behavior, purchase intention, targeting, body shape and size

When Retailers Target Women Based on Body Shape and Size: The Role of Ethical Evaluation on Purchase Intention

In 2010, the average American woman was five feet four inches tall and weighed 140 pounds, whereas the average American model was seven inches taller and 23 pounds lighter at five feet eleven inches tall and 117 pounds ("Body Image: Eating Disorders," 2010). That same year, the average Miss America winner was five foot seven and 121 pounds (Martin, 2010).

"American women are constantly bombarded by images of the ideal American woman, with only a small percentage of the women physically able to possibly attain this projected ideal" (Martin, 2010, p. 103). Extreme thinness is a social and cultural ideal, and as a result women often feel great anxiety and pressure to measure up ("Body Image: About Body Image," 2009). Girls and women are inundated with "Barbie Doll-like" images in advertising and media ("Body Image: Eating Disorders," 2010).

Apparel companies such as Abercrombie and Fitch and Hollister target exclusively women who fit the American ideal of beauty by not only using thin models in their advertising, but also offering limited pant sizes. The niche market of women who fit the American ideal of beauty is defined by their body shape and size, similar to other niche markets including petite women, plus size, and big and tall men's wear. Companies have found success in serving niche markets very well, rather than trying to serve the mass market only fairly well (Perreault, Cannon, & McCarthy, 2015).

Targeted marketing activities are designed and executed to be more appealing to the target market than to people in other segments (Ringold, 1995). It is essential for all marketers, and specifically marketers targeting the niche market of women who fit the American ideal of beauty, to understand the systematic differences in how target and nontarget consumers create meaning from advertising (Grier & Brumbaugh, 1999). Marketing efforts can have different meanings for consumers in their target market and outside their target based on the consumer's own cultural, social, and individual experiences (Grier & Brumbaugh, 1999). Different demographic groups evaluate the ethics of targeting differently, especially women, nonwhites, older, and less-educated consumers (Smith & Cooper-Martin, 1997). Knowing how different groups evaluate the targeting efforts of a company enables marketers to recognize, predict, and manage concerns associated with target marketing (Grier & Brumbaugh, 1999).

More focused target marketing has helped marketers achieve greater efficiency and effectiveness, but has been criticized as unethical (Smith & Cooper-Martin, 1997). For example, targeting of potentially harmful products such as lottery tickets, fast food, weight-loss products, and high-interest credit cards has been evaluated as less ethical (Smith & Cooper-Martin, 1997). Similarly, targeting vulnerable market segments such as minority groups, children, the elderly, alcoholics, or those recently bereaved have been perceived as less ethical (Smith & Cooper-Martin, 1997). Market illiterates, typically characterized by low income, low education, and naive in understanding the ways of the marketplace are another audience that might be considered vulnerable (Laczniak & Murphy, 1993).

Several studies have assessed ethical concerns and the controversy of targeting, specifically involving harmful products and vulnerable consumers. In 1997, Smith and Cooper-Martin ran two studies that differed by product type and target demographics which included less/more harmful products and low/high vulnerability. Smith and Cooper-Martin found that public concern exists about certain targeting strategies, particularly for products perceived as more harmful and targets perceived as more vulnerable. The study found that ethical concerns exist about targeting in general (Smith & Cooper-Martin, 1997).

Ten years later, Jones and Middleton (2007) examined the effects of perceptions of product harm and consumer vulnerability on ethical evaluations of target marketing strategies. The results of their study contradicted Smith and Cooper-Martin's (1997) study. Jones and Middleton's findings suggested that "an individual must first be able to recognize the presence of a moral issue before they can make an ethical evaluation" (2007, p. 255).

While harmful products have traditionally been associated with physical harm and vulnerable consumers as those with characteristics that limit their ability to maximize their utility and well-being (Smith & Cooper-Martin, 1997), some clothing retailers target adult female consumers, typically perceived as less vulnerable based on body shape and size with products such as apparel, considered physically harmless by most. A few clothing retailers exclusively target women who fit the American ideal of beauty. While beauty is culturally relative, today's American ideal of beauty emphasizes a physically fit and toned body shape and size (Mendez, Young, Mihalas, Cusumano, & Hoffman, 2006; Solomon, 2015; Vacker & Key, 1993).

Only the thinnest 5% of women fit the ideal female body (Wolszon, 1998). The body mass index (BMI) of many "Miss America" contestant winners fall below the World Health Organization's underweight cutoff, while only 3.26% of women in the U.S. are considered underweight according to their BMI classification (Martin, 2010).

Statement of the Problem

It is important for marketers to know whether a consumer's ethical evaluation of a company based on the company's targeting strategy impacts the consumer's purchase behavior. To compete more effectively, many companies focus on those consumers they have the greatest chance of satisfying (Kotler & Keller, 2006). This practice of target marketing (also known as targeting) is the intentional pursuit of exchange with a specific market segment through advertising or other marketing activities (Ringold, 1995). This practice "provides greater profit potential for the firm" (Perreault et al., 2015, p. 101); however, targeting may be an ineffective strategy when it results in controversy.

The combination of (1) the act or process of targeting, (2) a target market that is perceived to be vulnerable, and (3) a potentially harmful product, makes targeting unethical (Davidson, 2003). The ethical dilemma of targeting lies within the explicit inclusion or exclusion of groups of customers (Grier & Smith, 1997). In the case of targeting exclusively women who fit the American ideal of beauty, the included market segment is only 5% of women while it excludes the other 95% of women (Wolszon, 1998). According to the U.S. Department of Health and Human Services, the average waist circumference for females ages 20-29 is 35.51 inches while only five percent have waist circumference of 27.24 inches or smaller (2012). Trying on jeans can be painful for many women (Solomon, 2015). Clothing retailers targeting young women such as Hollister and Abercrombie and Fitch offer limited pant sizes between a 23" to a 33" waist circumference.

Sheehan (2013, p. 102) believes that, "as social beings, we like to look at beautiful things." Advertising agencies use attractive models to transfer positive affect from the attractive men and women featured in the advertisement to the product (Gulas & McKeage, 2000). Petroshius and Crocker (1989) found that affect transfer impacted sales, reporting that consumers purchase intentions increased after seeing ads for products that featured attractive people. According to Solomon (2015, p. 279), "our desires to match up to these ideals - for better or worse - drive a lot of our purchase decisions."

The desire to look as perfect as the ideal models in advertising can become all-consuming for some people (Swinson, 2011). Swinson, co-founder of the Campaign for Body Confidence, (Swinson, 2011) cites three studies that link negative body image with exposure to idealized images. The first shows that one in four people are depressed about their body (Swinson, 2011). The second study found that one-third of women say they would sacrifice a year of their life to achieve the ideal body weight and shape (Swinson, 2011). And in the third study, nearly fifty percent of girls think the pressure to look good is the worst part of being female (Swinson, 2011).

While ideals of both male and female beauty exist, the focus of this research is on the female ideal of beauty, specifically body shape and size (Englis, Solomon, & Ashmore, 1994). "Critics point to women's fashion and beauty magazines as one of the most influential and potentially damaging media channels because they are directly concerned with the cultural ideal of beauty and provide a vehicle where advertisers can easily link their products to the process of trying to attain beauty" (Englis et al., 1994; Sheehan, 2013, p. 97).

"The sophistication of target marketing and recognition of its importance as a means of achieving efficiency and effectiveness have never been greater" (Smith & Cooper-Martin, 1997, p. 1). At the same time, marketers are expected to make ethical marketing decisions and have respect and concern for the welfare of consumers (Laczniak & Murphy, 1993). This dilemma puts marketers in a tough spot.

This study aimed to assess consumers' ethical evaluation of a company's practice of targeting to women based on body shape and size in the clothing retail industry and its impact on the consumer's planned purchase behavior with said company.

Conceptual Framework and Theoretical Framework

The underpinnings of this study are based on behavioral theory and ethical theory. These theories help to explain how the ethical evaluation of a company's targeting strategy affects a consumer's purchase intention or planned behavior.

The Theory of Planned Behavior (TPB) model, developed by Ajzen (1985, 1991), helps predict and explain intentions and behavior. The TPB model supports the idea that attitude toward the behavior, subjective norm, and perceived behavioral control will impact intentions. The TPB is a well-researched model that has successfully predicted and explained behavior in a variety of settings (Yoon, 2011). "A central factor in TPB is the individual's intention to perform a given behavior. Intentions are assumed to capture the motivational factors that influence a behavior; they are indications of how hard people are willing to try, of how much effort they are planning to exert, in order to perform the behavior" (Ajzen, 1991, p. 181).

The TPB has provided extensive support for the ability to predict a wide range of behaviors for more than twenty years (Smith et al., 2007). The TPB has been used to predict health related behaviors from healthy eating (Astrom & Rise, 2001) to illicit drug use (Conner & McMillan, 1999). It has also been used to support other behaviors from recycling (Terry, Hogg, & White, 1999) to prosocial behavior (Harrison, 1995).

Ethics is the branch of philosophy concerned with morality (Timmons, 2012). Theories of justice, relativism, deontology, and teleology encompass the most notable ideas for social survival (Reidenbach & Robin, 1990). Reidenbach and Robin (1990) developed a multidimensional scale using justice, relativism, deontology, teleology-egoism, and teleology-utilitarianism theories as the basis for ethical evaluation. The three dimensions of Reidenbach and Robin's (1990) multidimensional ethics scale (MES) are moral equity, relativism, and contractualism.

Consistent with consumer behavior theories, Hunt and Vitell (1986) suggest that ethical judgments impact intention. Managers are encouraged to behave in an ethical manner because a firm's ethical behaviors are thought to influence consumers' image of the company, thus product sales (Folkes & Kamins, 1999).

Smith and Cooper-Martin (1997) ran two studies that differed by product types and target demographics using a 2 x 2 full factorial design, which included less/more harmful products and low/high vulnerability. They measured the ethical evaluation using Reidenbach and Robin's (1990) MES and behavioral intentions using Berry's (1977) descriptions of consumer activism. Smith and Cooper-Martin found that public concern exists about certain targeting strategies, particularly for products perceived as more harmful and targets perceived as more vulnerable. The study found that ethical concerns exist about targeting in general (Smith & Cooper-Martin, 1997).

In 1997, Creyer studied the influence of firm behavior on purchase intention. In the study, consumers reported that "the ethicality of a firm's behavior is an important consideration during the purchase decision, ethical corporate behavior is expected, they will reward ethical behavior by a willingness to pay higher prices for that firm's product, and although they may buy from an unethical firm, they want to do so at lower prices which, in effect, punishes the unethical act" (Creyer, 1997, p. 5).

Extant literature on planned behavior, targeting, ethical evaluation, and the impact of ethical evaluations on intentions provides the foundation for the following hypotheses and research model developed for this research study.

Hypothesis 1: Ethical evaluation positively affects attitude toward a company targeting women based on body shape and size.

Hypothesis 2: Ethical evaluation positively affects subjective norm of a company targeting women based on body shape and size.

Hypothesis 3: Ethical evaluation positively affects an individual's intention to buy from a company targeting women based on body shape and size.

Hypothesis 4: Ethical evaluation of a company targeting women based on body shape and size will differ in comparison to the ethical evaluation of a company not targeting women based on body shape and size.

Based on the theoretical framework discussed above, a model integrating ethical evaluation and the Theory of Planned Behavior is shown in Figure 1.

Contribution to the Field

The Theory of Planned Behavior (TPB) has previously been used to explain and predict consumer behavior in a variety of contexts. Prior studies have informed us that targeting is effective, but perceived as less ethical when a company targets a vulnerable target market, or markets a potentially harmful product. Researchers suggested additional research to include different product classes and other characteristics of target vulnerability and to study targeting of harmless products. This study assessed the impact of a consumer's ethical evaluation on attitude, subjective norm, and the consumer's intention to buy from a company that is targeting women based on body shape and size. As a result, this research contributed to extant literature by studying target markets that are not typically perceived as vulnerable and products that are potentially psychologically harmful rather than physically harmful.

In their 1997 study, Smith and Cooper-Martin assessed the correlations between ethical evaluations and behavioral intentions (specifically disapproving behaviors and approving behaviors). This study took the next step by assessing the correlation between ethical evaluation and purchase intention.

Methodology

A web-based survey was conducted via an electronic questionnaire. Female consumers on a web-based research panel were invited to participate. Qualtrics, provider of online survey software and panels, collected 421 responses from females living in the United States, ages 18-29, with a household income level of at least $20,000.

Two scenarios were provided for respondents to evaluate. In both scenarios, the companies are targeting women, ages 18-29. Size charts were provided in each scenario. The scenarios were not primed with possible risks. In both scenarios, respondents were informed of the average waist circumference for U.S. females within the 18-29 age range.

In one scenario, the company offers a wider range of sizes available demonstrating a company not targeting based on body shape and size. Scenario 1 reads:

A clothing retailer, well-known to the public, recently introduced a new product line of pants. This new line of pants is intended to appeal to consumers who are females ages 18-29. The sizing chart for the product line of pants is shown below.

According to the U.S. Department of Health and Human Services, the average waist circumference for females 18-29 is 33" (XL or size 16). In this scenario, the company is offering sizes above and below the average waist circumference of females ages 18-29.

In another scenario, the company limits its size options to demonstrate the use of targeting based on body shape and size. Scenario 2 reads:

A clothing retailer, well known to the public, recently introduced a new product line of pants. This new line of pants is intended to appeal to consumers who are females ages 18-29. The sizing chart for the product line of pants is shown below.

According to the U.S. Department of Health and Human Services, the average waist circumference for females 18-29 is 33" (XL or size 16). In this scenario, the company is only offering sizes below the average waist circumference of females, ages 18-29.

Attitude, subjective norm, perceived behavioral control, and intention (planned behavior) were latent variables in this study. These variables originate from the Theory of Planned Behavior (TPB) and were measured using an adapted version of Liao et al.'s (2010) scale for attitude, subjective norm, perceived behavioral control, and intention.

Ethical evaluation was also a latent variable in this study. Reidenbach and Robin's 1990 Multidimensional Ethics Scale (MES) was used to calculate a measure of ethicality based on moral equity, relativism, and contractualism. For validity purposes, ethical evaluation was also assessed independently.

Findings

For each scenario, each respondent's item responses were averaged to calculate a scale score for each latent variable. Each latent variable scale score was tested for normality using the Shapiro-Wilk test. All measures were non-normal.

For Scenario 1, the calculated scale scores were used to provide descriptive statistics as shown in Table 1. In Scenario 1, the data is negatively skewed with the mean falling below the median for ethical evaluation (E), attitude (A), subjective norm (SN), perceived behavioral control (PBC), and intention (I).

For Scenario 2, the descriptive statistics for the calculated scale scores are shown in Table 2. In Scenario 2, the same negative skewness from Scenario 1 is seen in perceived behavioral control (PBC). Otherwise, the data is positively skewed for ethical evaluation (E), attitude (A), subjective norm (SN), and intention (I).

The means and the medians for each factor were considered to better understand the context of each scenario. Each measured item was measured on a 7-point itemized rating scale with 1 representing strong disagreement and 7 indicating strong agreement. Overall, agreement was stronger for Scenario 1 than Scenario 2 as demonstrated by the higher means and medians for ethical evaluation, attitude, subjective norm, perceived behavioral control, and intention. Respondents were more disagreeable in their ethical evaluation, attitude, subjective norm, and intention in Scenario 2. Albeit to a lesser extent, respondents were more disagreeable in their perceived behavioral control in Scenario 2 in comparison to Scenario 1.

Using the average scale values for the factors, a correlation analysis was conducted using SPSS to get a preliminary measure of the strength and direction of association between factors. A correlation matrix is shown in Tables 3 and 4 based on the scale scores for each latent factor in each scenario. Due to the non-normality of the data, Spearman's Rho was used.

In the correlation analysis, ethical evaluation (E) was found to more strongly correlate to attitude (A), subjective norm (SN), and purchase intention (I) in Scenario 2 than in Scenario 1. The opposite was true for the correlation between ethical evaluation (E) and perceived behavioral control. It is important to note that correlation analysis is a univariate assessment, demonstrating the degree to which one variable changes with another.

Structural Equational Modeling (SEM) was utilized in order to assess the construct validity and theoretical relationship among a set of concepts represented by multiple measured variables (Hair et. al, 2010). First, a single group measurement model was developed and administered for Scenario 1 and then Scenario 2. The measurement model specified the indicators for each construct and enabled an assessment of construct validity (Hair et. al, 2010). Second, a single group structural model was developed and executed for each scenario. The structural model tested how well the measured variables represented the constructs (Hair et. al, 2010). Lastly, the factorial invariance of a measuring instrument was tested using the multi-group functionality of EQS.

Single-Group Measurement Model

The measurement model in SEM defines relations between the scores on the measuring instrument and the underlying constructs they are designed to measure. The measurement model for Scenario 1 and Scenario 2, as shown in Figure 2, shows the relationships between the observed measured items and the unobserved latent factors. The rectangles represent the observed measured items or the observed survey scale items while the circles signify the latent variables that are measured by the observed measured items. The single-headed arrows represent the impact of one variable on another, while the double-headed arrows represent covariances or correlations between pairs of variables (Byrne, 2006). A single-group measurement model for each scenario was run using EQS in robust mode to adjust for the non-normality of the data.

Goodness-of-fit indicates how well the specific model reproduces the observed covariance matrix among the indicator items (Hair, et al.). The goodness-of-fit measures evaluated for the present study include Bentler and Bonett's (1980) Normed Fit Index (NFI) and Bentler's (1990) Comparative Fit Index (CFI). The third fit measure used in this study is the Root Mean Square Error of Approximation (RMSEA), which corrects for both model complexity and sample size by including each in its computation (Hair et al., 2010). These fit measures are summarized in Table 5 for both scenarios.

Bentler and Bonett's (1980) Normed Fit Index (NFI) has been a practical criterion of choice (Byrne, 2006). Since NFI has shown a tendency to underestimate fit in small sample, Bentler revised the NFI to consider sample size in 1990 (Byrne, 2006). The resulting Comparative Fit Index (CFI) and the initial NFI provides a measure of complete covariation in the data (Byrne, 2006). According to Hair et al. (2010), a CFI value above 0.92 for sample sizes greater than 250 with 12-30 observed variables demonstrates goodness-of-fit. Using this guide, both CFI measures represent a good fit. Byrne (2006) cites Hu and Bentler's 1999 research, which is more stringent. Based on the revised cutoff value of 0.95, both NFI and CFI fit measures represent a well-fitting model (Byrne, 2006).

Poor fit measures provide an alternative to the goodness-of-fit measures discussed above. Larger values represent poorer fit. The RMSEA has been cited as one of the most informative criteria in covariance structure modeling (Byrne, 2006). Values less than .05 indicate good fit, .06 to .08 represent reasonable fit, .08 to .10 point to mediocre fit, while values greater than .10 suggest poor fit (Byrne, 2006).

For the final measurement model, the NFI and CFI fit measures for both scenarios represent a well-fitting model (Byrne, 2006). The RMSEA indicates a good fit for Scenario 1 and an extremely good fit for Scenario 2.

Single Group Structural Model

After a good fit was achieved on each measurement model, a structural model for each scenario was developed. As with the final measurement model, EQS was run in robust mode to adjust for the non-normality of the data. The hypothesized relationships are depicted by the single-headed arrows as shown in Figure 3, the structural models for Scenario 1 and 2, respectively.

For the structural model, the NFI and CFI fit measures represent a poor fit for Scenario 1 and a good fit for Scenario 2. Similarly, the RMSEA indicates a poor fit for Scenario 1 and a good fit for Scenario 2. Additionally, the reliability coefficient rho was below .95 in Scenario 1 and greater than .95 in Scenario 2.

While Scenario 2 has a good fit, good model fit alone is insufficient to support a proposed structural theory (Hair, Black, Babin, & Anderson, 2009). Hair et al. posit that a structural model is considered acceptable when it demonstrates acceptable model fit and the path estimates representing the hypotheses are statistically significant (2009). Thus, the estimated model parameters for Scenario 2 are shown in Table 7. Of the six paths modeled, four were statistically significant in Scenario 2. Figure 4 shows the significant paths of the estimated model for Scenario 2.

Estimated model parameters were not calculated nor was a path diagram produced for Scenario 1 for two reasons. First, the structural model fit measures indicated a poor fit for Scenario 1. Second, Hypotheses 1-3 related to the company targeting women based on body shape and size (as in Scenario 2, not Scenario 1).

In the context of this study, the single-group measurement and structural testing suggested that ethical evaluation seemed to differ between the two scenarios. In order to test for the significance of this invariance, measurement equivalence was simultaneously tested across both scenarios using a test of multi-group measurement model invariance.

Multi-Group Measurement Model Invariance Test

To assess whether the items composing a particular measuring instrument operate equivalently across both scenarios, a test for factorial invariance of a measuring instrument was run. In order to seek evidence of multi-group invariance, EQS was run in robust mode to adjust for the non-normality of the data. In EQS, the Lagrange Multiplier (LM) test provides a precise approach to identifying parameters that are not equivalent across groups (Byrne, 2006).

When testing for measurement equivalence, factor loadings for the data in the first group (Scenario 1) are constrained to equal the factor loadings for the data in the second group (Scenario 2). In EQS, equality constraints are specified in the /CONSTRAINTS paragraph of the input file. These equality constraints, as they related to the factor loadings and one error covariance, are shown in Table 9.

The output file and specifically the LM test results were examined to determine which parameters, if any, were not operating equally across the two scenarios. Parameters with probability values for the univariate Chi-Square of less than .05 indicate variance. According to the LM test, two parameters with statistically significant probability values were identified. The two measures, EEETH2 and EEETH3, were found to be operating differentially across Scenarios 1 and 2. Table 10 reviews the specific measures operating differently between the two scenarios as indicated by the LM test. This evidence of multigroup invariance demonstrates that these two measured items on the measurement instrument are not operating equivalently across the two scenarios (Byrne, 2006). Both items found to be non-invariant were designed to measure ethical evaluation (E). These findings are consistent with the model fits determined during the single-group structural model runs, in which Scenario 1 had a poor-fitting model while Scenario 2 had a good-fitting model.

The resulting data analysis from the single-group structural test identified that ethical evaluation contributes to a consumer's attitude and subjective norm when retailers target women based on body shape and size. The multi-group measurement model invariance test indicated that a consumer's ethical evaluation of clothing retailers that target women based on body shape and size (as represented in Scenario 2) was found to be different from a consumer's ethical evaluation of retailers who do not target women based on body shape and size (as represented in Scenario 1). Based on the results from the data analysis, conclusions were reached regarding each hypothesis as shown in Table 11.

Discussion

Based on the findings from the descriptive statistics, exploratory factor analysis, and the structural equation modeling, interpretations can be made. Scenario 1 presented a situation in which a company offers a wider range of sizes available, demonstrating that the company is not targeting women based on body shape and size. Scenario 2, on the other hand, presented a situation in which a company is targeting women based on body shape and size, by limiting size options to consumers. As such, the discussion that follows will center around Scenario 2 and the company targeting women based on body shape and size, unless otherwise noted.

Ethical Evaluation of Targeting Based on Body Shape and Size

The multi-group measurement model invariance test identified two measured items, both designed to measure ethical evaluation (E), that were not operating equivalently across the two scenarios. These findings indicate that ethical evaluation significantly varied between Scenario 1 and 2. This outcome was expected based on the ethical dilemma presented in Scenario 2, but not in Scenario 1. It is evident that respondents were able to recognize the presence of a moral issue in Scenario 2.

Attitude

In the context of this study, ethical evaluation was shown to have a statistically significant influence on attitude. Attitude refers to the degree to which a person has a favorable or unfavorable evaluation of the company (Ajzen, 1991). This finding reinforces Folkes and Kamins' results that firms' ethical actions influence consumers' attitudes toward firms (Folkes & Kamins, 1999). However, the data analysis indicated that attitude did not significantly affect influence, which contradicts Ajzen's TPB in which attitudes are internal dispositions that are expected to induce corresponding behavior (Ajzen, 2012).

The findings of this study both confirmed and contradicted Yoon's (2011) proposed integrated model of TPB and ethics theory. This study supported Yoon's (2011) model in terms of the influence of ethical evaluation on attitude and subjective norm components of the TPB model. The contradiction is related to the influence of ethical evaluation on intention, which is discussed later.

Subjective Norm

When retailers target women based on body shape and size, ethical evaluation has a statistically significant influence on subjective norm. Subjective norm represents the perceived social pressure to consume (or not consume) products from a company (Ajzen, 1991). In Dove's Global Beauty and Confidence Report (Etcoff & Paxton, 2016), researchers report that women are increasingly feeling intense pressure regarding their appearance. American women are inundated with images of the ideal American woman (Martin, 2010). Advertising agencies use attractive models to transfer positive affect from the attractive men and women featured in the advertisement to the product (Gulas & McKeage, 2000). Petroshius and Crocker (1989) found that affect transfer impacted sales, reporting that consumers' purchase intentions increased after seeing ads for products that featured attractive people.

According to Solomon (2015, p. 279), "our desires to match up to these ideals - for better or worse - drive a lot of our purchase decisions." Consistent with Solomon's stance and Petroshius and Crocker's (1989) study, the findings of this study demonstrate that the perceived social pressure plays an impactful role on the likelihood a consumer will purchase a product.

Perceived Behavioral Control

In the context of this study, perceived behavioral control was shown to have a statistically significant influence on intention. Perceived behavioral control is the extent to which people believe that they can perform a given behavior if they are inclined to do so (Ajzen, 2012). This finding is consistent with extant TPB literature. The easier a business makes it for a customer to purchase (such as short lines in physical retail stores, an easy online checkout process, stored credit card information, two-day shipping, etc.) the more likely a consumer will purchase.

Intention

Perhaps the most interesting finding of this study was related to the influence of ethical evaluation on intention. Even though recent studies have demonstrated the predictive ability of Reidenbach and Robin's Multidimensional Ethics Scale on intent to act (Loo, 2004; Schepers, 2003), ethical evaluation of retailers targeting women based on body shape and size was not a statistically significant predictor of purchase intention. The findings of this study primarily negate findings from earlier empirical studies with one exception. First, the findings in this study contradict Creyer's 1997 study in which the ethicality of a firm's behavior was found to be an important consideration during the purchase decision. According to Creyer, consumers reward ethical behavior by paying higher prices for that firm's product, and although consumers may buy from an unethical firm, they want to do so at lower prices which, in effect, punishes the unethical act (1997).

Similarly, the resulting data contradicts Conner and Armitage's (1998) position that moral norms have an important influence on behaviors with a moral or ethical dimension as well as Yoon's (2011) research results that indicated ethical evaluation influences intention. In his meta-analysis, Manstead (2000, p. 29) postulated that there are good conceptual and theoretical grounds for arguing that moral norm represents a potential influence on behavior intention. The results of this study do not support Manstead's postulation.

Despite the difference in overall ethical evaluation of the two scenarios and the lower mean score for ethical evaluation in Scenario 2 (representing disagreement with ethicality of the business), ethical evaluation was not a significant predictor of intention to buy. This means that while the consumer believes the business to be unethical, it does not influence their intention to buy from that business.

The findings were consistent, though, with Carrigan and Attalla's (2001) investigation of good and bad ethical conduct and its effect on consumer purchase behavior. Carrigan and Attalla conducted two focus groups and found that the participants of their focus groups didn't care about ethical consideration in their purchase decision making (2001).

Even though ethical evaluation was hypothesized to play a significant role in predicting intention, several reasons may contribute to its lack of prediction capabilities. Ethical evaluation as an independent factor did significantly influence purchase intention, as demonstrated in the (univariate) correlation analysis. Purchase decisions are rarely that simple as established in the copious stream of TPB research. In SEM, the structural parameter estimates for ethical evaluation were not significant in relation to purchase intention when other TPB factors were simultaneously considered including attitude, subjective norm, and perceived behavioral control.

The findings of this study could possibly be explained by the fact that only 21.3% of the female participants in the study reported wearing a size XL or XXL and above. This could have led to a minority of the participants feeling excluded by the size options offered by the company. Whereas 78.8% of the female participants reported a size L or smaller, which could have led the majority of the female participants to feel included by the size options offered by the company. The company in Scenario 2 targeted consumers based on body shape and size by only offering sizes below the average waist circumference of 33" (XL or size 16) as reported by the U.S. Department of Health and Human Services for females aged 18-29.

This possible explanation was tested in a post hoc analysis. As a result, a significant difference between groups of women based on their pant size (XXS-L in comparison to XL-XXL) was discovered. Specifically, there was a significant difference in the overall ethical evaluation scores for women wearing XXS-L pant sizes (M=3.62, SD=1.53) and women wearing XL-XXL pant sizes (M=2.91, SD=1.43); t(419)=3.93, p=0.000. There was also a significant difference in the purchase intention scores for women wearing XXS-L pant sizes (M=3.68, SD=1.79) and women wearing XL-XXL pant sizes (M=2.54, SD=1.95); t(419)=5.24, p=0.000. This finding provides cause for future research.

It is also possible that the survey respondents did not recognize the extent or severity of the potential psychological harm on women that could come from limiting sizes, or the ethical dilemma of targeting that lies within the explicit exclusion of customers. This corroborates with Jones and Middleton's findings that suggested "an individual must first be able to recognize the presence of a moral issue before they can make an ethical evaluation" (2007, p. 255).

In 2016, Moorman and Day (p. 20) called for research to be done In this area, saying, "Marketing has a high level of perceived misconduct. Therefore, we urge scholars to document whether this reputation is deserved by development methods to document a full range of these behaviors for marketers in different roles." Based on the results of this study, this reputation is undeserved when consumers intentionally buy from a business despite their unethical evaluation of the company's marketing efforts.

Ultimately, this study found that a consumer's judgment of unethical business activity does not have bearing on the likelihood a consumer will purchase a product. The notion that ethical business is good business did not hold true in this scenario.

Implications

There has been an increased attention in both the media and academic literature surrounding ethical consumption (Hassan, Shiu, & Shaw, 2016). As a result, researchers have identified discrepancies between what consumers think, intend to do, and actually do (Hassan et al., 2016).

Marketing managers have been encouraged to behave in an ethical manner because a firm's ethical behaviors have been thought to influence consumers' image of the company, thus product sales. This study reflects that other factors including social pressure and the ease of buying are more important to females ages 18-29 than the ethicality of the company.

Marketing managers should be aware of how the public perceives the ethicality of certain targeting strategies, such as targeting women based on body shape and size. In the context of this study, consumers evaluated the company targeting women based on body shape and size as less ethical in comparison to the company who did not target women based on body shape and size. As a result, retailers should focus on those consumers they have the greatest chance of satisfying, even if that means excluding some consumers. This type of targeting has helped marketers achieve greater efficiency and effectiveness resulting in greater sales and profit potential for the firm.

While the results of this research indicate that ethical evaluation does not impact purchase intention, it does significantly influence a consumer's attitude and subjective norm in the context of a retailer's targeting strategy based on body shape and size. This suggests that companies using potentially controversial targeting strategies should assess consumer attitudes regularly.

In the context of this study, young women (ages 18-29) have grown up with the beauty pressure imposed by women's fashion and beauty magazines. With the long-term exposure to idealized images shown in advertising, it is possible that these young women have accepted this as a social norm and therefore is less of a personal deciding factor in their purchases.

The present study provides evidence that a person may evaluate the actions of the company as unethical, yet express an intention to purchase due to the significant influence of the subjective norm. The dilemma the consumer may be experiencing subsists within the perceived social pressure of subjective norm. The social pressure from their referent groups to buy from the company may be great and influence the consumer's intention to purchase congruent products. Reference group influence can profoundly impact on behavior (Gupta & Ogden, 2009). Some consumers will develop aspirations of belonging to a reference groups such that peer pressure may override their own personal ethical beliefs about a company.

The findings from the present study conflict with the results of Jin Ma et al. (2012) who posit that the influence of subjective norm on purchase decision making may be less prominent among the younger consumer cohort. "This young consumer segment likes to learn about benefits of products and services through their own experiences" (Jin Ma et al., 2012). In the present study though, subjective norm was a significant indicator of purchase intention. A possible reason for this might be that clothing is a publicly worn and visible product to others. The results could have been different if the product was used in private. This is a suggestion for future research. A second reason might be the social dilemma experienced by the consumer. Social dilemma, under the auspices of social exchange theory, whereby the anticipated benefits and costs of engaging in ethical purchase behavior are weighed (Gupta & Ogden, 2009).

Ideally, moral norms should have an important influence on behaviors with moral or ethical dimensions, and work in parallel with attitudes, subjective norms, and PBC (Conner & Armitage, 1998). The results of this study provide evidence that exceptions to Conner and Armitage's ideal may exist. This study suggests that a consumer's ethical evaluation of a company can be overridden by other factors such as the pressure to conform to the ideal norm. In today's society, extreme thinness is a social and cultural ideal, and as a result women often feel great anxiety and pressure to measure up ("Body Image: About Body Image," 2009).

When making decisions in this study, the belief about what social referents (e.g. a spouse, close friends, or colleagues) thought was more important than their own personal ethical evaluation. Marketing managers need to understand how consumers might perceive harmfulness of products, whether it is physical, psychological, or economical. Similarly, marketers need to evaluate the external perception of their targeting strategy. While most marketers focus on those consumers they have the greatest chance of satisfying as this targeting strategy typically provides greater profit potential for the firm, consumers may feel differently regarding the explicit inclusion or exclusion of groups of customers.

The media's coverage and criticism of unethical business practices may generate buzz, sell newspapers, and increase program ratings, but this study showed that consumers did not include the ethicality of the retailer in their purchase decision-making process. However, criticism from media could also influence public perception and ultimately social norm, which did have significant influence on purchase intention.

Criticism could surprise marketing managers if the public perception differs from the marketer's perception of product harm and targeting strategy. It is also worth noting that ethical evaluation differs among customers. While some customers may view a firm as ethical, others could still view it as unethical.

Limitations

Five limitations of this research are noted. First, this study used a survey instrument, which limits respondents to self-report purchase intentions rather than observing actual purchase behavior. As a result, respondents may project their ideal response, rather than their actual response.

Since Qualtrics filtered out surveys with incomplete responses, a second limitation of this study could include a nonresponse bias. It is likely that some consumers on the panel did not complete or submit the survey.

A third limitation of this study includes the use of two generic scenarios in which respondents were asked to evaluate, in lieu of identifying specific brands. In an effort to evaluate the wording of the scenarios, a pre-test was conducted. The purpose of the pretest was to identify whether female students recognized the ethical dilemma demonstrated in the scenarios/size charts, saw the difference between the two scenarios, and understood the questions (wording/clarity). The pre-test also determined whether the repetition was necessary or not. Minor changes were made to the instrument based on the results of the pretest to improve clarity.

Another limitation noted for this research includes a potential testing effect. Since each respondent was presented with two scenarios and similar response statements, a testing effect could have affected respondent's choices. In an attempt to minimize this testing effect, the scenarios were presented on the survey in a random order, so some respondents evaluated Scenario 1 first and Scenario 2 second, and vice versa.

Lastly, a limitation of this research includes the lopsided distribution of pant size reported by participants. Only 21.3% of the female participants in the study reported wearing a size XL or XXL and above, whereas 78.8% of the female participants reported wearing a size XXS, XS, S, M, or L.

Suggestions for Future Research

There are several directions for future research. First, the aforementioned post-hoc analysis pinpointed a significant difference between groups of women based on pant size (XXS-L in comparison to XL-XXL). The current study could be retested using a guaranteed percentage quota in terms of pant sizes reported by women. Structural equation modeling and a multi-group analysis could then be conducted across groups of women based on pant size reported to determine if there is equivalence or variance across the groups.

Since parental status was not collected in the current study, it would be interesting to test whether the ethical evaluation of a company's targeting strategy of body shape and size differs between parents of teenage daughters versus parents of teenage sons, or between parents and nonparents. Since ideals of both male and female beauty exist, future research could be expanded to assess the influence of a company's targeting strategy using the male ideal of beauty, specifically body shape and size.

The conceptual framework could be further examined by testing different scenarios in which ethical dilemmas exist. Using different scenarios would provide an opportunity to refine the ethical dilemma of targeting. Nondemographic characteristics could be used, such as consumers with low self-esteem or body image. Different products might allow for further examination of product harmfulness.

Testing should be conducted to determine if ethical evaluation mediates or moderates the influence of attitude or subjective norm on purchase intention within the Theory of Planned Behavior. Future research could also examine other variables that might influence ethical evaluation. Continued application and further refinement of the MES scale could expand the scale's coverage and improve our understanding of ethical judgments (Jones & Middleton, 2007; Smith & Cooper-Martin, 1997).

References

References

Ajzen, I. (1985). From Intentions to Actions: A Theory of Planned Behavior. In J. Kuhl & J. Beckmann (Eds.), Action Control (pp. 11-39): Springer Berlin Heidelberg.

Ajzen, I. (1991). The theory of planned behavior. Organizational behavior and human decision processes, 50(2), 179-211.

Ajzen, I. (2012). Attitudes and persuasion. The Oxford handbook of personality and social psychology, 367-393.

Astrom, A., & Rise, J. (2001). Young adults' intention to eat healthy food: Extending the theory of planned behaviour. Psychology & Health, 16(2), 223-237.

Berry, J. M. (1977). Lobbying for the people: The political behavior of public interest groups (Vol. 95): Princeton University Press Princeton, NJ.

Body Image: About Body Image. (2009). Retrieved from https://www.womenshealth.gov/body-image/about-body-image/index.html

Body Image: Eating Disorders. (2010). Retrieved from https://www.womenshealth.gov/body-image/eating-disorders/index.html

Byrne, B. M. (2006). Structural equation modeling with EQS: Basic concepts, applications, and programming: Sage.

Carrigan, M., & Attalla, A. (2001). The myth of the ethical consumer-do ethics matter in purchase behaviour? Journal of consumer marketing, 18(7), 560-578.

Conner, M., & Armitage, C. J. (1998). Extending the theory of planned behavior: A review and avenues for further research. Journal of Applied Social Psychology, 28(15), 1429-1464.

Conner, M., & McMillan, B. (1999). Interaction effects in the theory of planned behaviour: Studying cannabis use. British journal of social psychology, 38(2), 195-222.

Creyer, E. H. (1997). The influence of firm behavior on purchase intention: do consumers really care about business ethics? Journal of consumer marketing, 14(6), 421-432.

Davidson, D. K. (2003). Selling sin: The marketing of socially unacceptable products: Greenwood Publishing Group.

Englis, B. G., Solomon, M. R., & Ashmore, R. D. (1994). Beauty before the eyes of beholders: The cultural encoding of beauty types in magazine advertising and music television. Journal of advertising, 23(2), 49-64.

Etcoff, N., & Paxton, S. (2016). The Dove Global Beauty and Confidence Report. Retrieved from

Folkes, V. S., & Kamins, M. A. (1999). Effects of information about firms' ethical and unethical actions on consumers' attitudes. Journal of Consumer Psychology, 8(3), 243-259.

Grier, S. A., & Brumbaugh, M. (1999). Noticing cultural differences: Ad meanings created by target and non-target markets. Journal of advertising, 28(1), 79-93.

Grier, S. A., & Smith, N. C. (1997). The Target and Its "Other": Exploring the Social Context and Interaction Of Consumer Segments. Advances in Consumer Research, 24, 145-147.

Gulas, C. S., & McKeage, K. (2000). Extending social comparison: An examination of the unintended consequences of idealized advertising imagery. Journal of advertising, 29(2), 17-28.

Gupta, S., & Ogden, D. T. (2009). To buy or not to buy? A social dilemma perspective on green buying. Journal of consumer marketing, 26(6), 376-391.

Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2009). Multivariate Data Analysis (7th Edition ed.). Upper Saddle River, NJ: Pearson Prentice-Hall.

Harrison, D. A. (1995). Volunteer motivation and attendance decisions: Competitive theory testing in multiple samples from a homeless shelter. Journal of applied psychology, 80(3), 371.

Hassan, L. M., Shiu, E., & Shaw, D. (2016). Who says there is an intention-behaviour gap? Assessing the empirical evidence of an intention-behaviour gap in ethical consumption. Journal of Business Ethics, 136(2), 219-236.

Hunt, S. D., & Vitell, S. (1986). A general theory of marketing ethics. Journal of macromarketing, 6(1), 5-16.

Jin Ma, Y., Littrell, M. A., & Niehm, L. (2012). Young female consumers' intentions toward fair trade consumption. International Journal of Retail & Distribution Management, 40(1), 41-63.

Jones, J. L., & Middleton, K. L. (2007). Ethical decision-making by consumers: The roles of product harm and consumer vulnerability. Journal of Business Ethics, 70(3), 247-264.

Kotler, P., & Keller, K. L. (2006). Marketing Management, Upper Saddle River. New Yersey.

Laczniak, G. R., & Murphy, P. E. (1993). Ethical marketing decisions: The higher road.

Liao, C., Lin, H.-N., & Liu, Y.-P. (2010). Predicting the use of pirated software: A contingency model integrating perceived risk with the theory of planned behavior. Journal of Business Ethics, 91(2), 237-252.

Loo, R. (2004). Support for Reidenbach and Robin's (1990) eight-item multidimensional ethics scale. The Social Science Journal, 41(2), 289-294.

Manstead, A. S. (2000). The role of moral norm in the attitude-behavior relation.

Martin, J. B. (2010). The development of ideal body image perceptions in the United States. Nutrition Today, 45(3), 98-110.

Mendez, L. M. R., Young, E. L., Mihalas, S. T., Cusumano, D. L., & Hoffman, L. L. (2006). What Teachers Can Do to Reduce Hidden Stressors for Girls in Middle School. Middle School Journal (J3), 38(2), 13-22.

Moorman, C., & Day, G. S. (2016). Organizing for marketing excellence. Journal of Marketing, 80(6), 6-35.

Perreault, W. D., Cannon, J. P., & McCarthy, E. J. (2015). Essentials of marketing: A marketing strategy planning approach (14 ed.). New York, NY: McGraw Hill Education.

Petroshius, S. M., & Crocker, K. E. (1989). An empirical analysis of spokesperson characteristics on advertisement and product evaluations. Journal of the Academy of marketing Science, 17(3), 217-225.

Reidenbach, R. E., & Robin, D. P. (1990). Toward the Development of a Multidimensional Scale for Improving Evaluations of Business Ethics. Journal of Business Ethics, 9(8), 639-653-653.

Ringold, D. J. (1995). Social Criticisms of Target Marketing Process or Product? American Behavioral Scientist, 38(4), 578-592.

Schepers, D. H. (2003). Machiavellianism, profit, and the dimensions of ethical judgment: A study of impact. Journal of Business Ethics, 42(4), 339-352.

Sheehan, K. B. (2013). Controversies in contemporary advertising: Sage Publications.

Smith, C., & Cooper-Martin, E. (1997). Ethics and Target Marketing: The Role of Product Harm and Consumer Vulnerability. The Journal of Marketing, 61(3), 1-20.

Smith, J. R., Terry, D. J., Manstead, A. S. R., Louis, W. R., Kotterman, D., & Wolfs, J. (2007). Interaction Effects in the Theory of Planned Behavior: The Interplay of Self-Identity and Past Behavior. Journal of Applied Social Psychology, 37(11), 2726-2750.

Solomon, M. R. (2015). Consumer Behavior: Buying, Having, and Being (11 ed.). Upper Saddle River, NJ: Pearson Education, Inc.

Swinson, J. (2011). False beauty in advertising and the pressure to look 'good'. CNN. Retrieved from http://www.cnn.com/2011/OPINION/08/08/swinson.airbrushing.ads/

Terry, D. J., Hogg, M. A., & White, K. M. (1999). The theory of planned behaviour: self-identity, social identity and group norms. British journal of social psychology, 38(3), 225-244.

Timmons, M. (2012). Moral theory: an introduction: Rowman & Littlefield Publishers.

U.S. Department of Health and Human Services, C. f. D. C. a. P., National Center for Health Statistics. (2012). Anthropometric reference data for children and adults : United States, 2007-2010. . Retrieved from www.cdc.gov/nchs/data/series/sr_11/sr11_252.pdf

Vacker, B., & Key, W. R. (1993). Beauty and the beholder: The pursuit of beauty through commodities. Psychology & Marketing, 10(6), 471-494.

Wolszon, L. R. (1998). Women's Body Image Theory and Research A Hermeneutic Critique. American Behavioral Scientist, 41(4), 542-557.

Yoon, C. (2011). Theory of planned behavior and ethics theory in digital piracy: An integrated model. Journal of Business Ethics, 100(3), 405-417.