Effects of Gay-Themed Advertising among Young Heterosexual Adults from U . S . and South Korea

Little research has been conducted to explain the effects of gay-themed advertising in a cross-cultural context. Such research has been particularly scarce in South Korea. This study is designed to investigate the effects of cultural orientation, gender, and types of gay-themed advertising in evaluation of gay male and female lesbian print ads. The study results indicate that Korean college students (i.e., collectivists) had lower tolerance of homosexuality than did U.S. college students (i.e., individualists). The study also finds that gender-role beliefs lead males to have lower tolerance of homosexuality. However, gender did not have statistically significant impacts on advertising and brand evaluation. Lastly, the study also found that lesbian imagery print ads could lead to greater tolerance of homosexuality and more favorable evaluations of the advertising and brand than could gay male imagery print ads. The current study sheds some light on the characteristics of U.S. consumers and Korean consumers on tolerance of homosexuality and gay-themed ads. Limitations and areas for further research are discussed.


Introduction
For 2015, the combined buying power of the U.S. lesbian, gay, bisexual, and transgender (LGBT) adult population was estimated at $917 billion (Witeck Communications 2016).In step with the steady growth of LGBT consumer's buying power (Witeck Communications 2016), advertising spending in gay media rose in 2013, reaching an all-time high of $381.4 million, an 18.2% gain over that of 2012(Rivendell Media 2014).Gay media refers to media that predominantly targets a gay or lesbian audience.In general, however, the primary target market for gay media encompasses members of bisexual and transgender audiences.In print ads of gay media, the percentage of gay-specific ads (ads with explicit references to LGBT lifestyles in graphics or messages) was more than 56% (Rivendell Media 2014).
LGBT consumers are deemed to be highly loyal to brands or companies that target them (Angelini and Bradley 2010).In the United States, many companies target LGBT consumers; the industries of these companies range from fashion, airline, hotel, and financial services to car companies and airlines, and include such entities as Accenture, Bank of America, Hyatt Hotel, Levi Strauss, and American Express.What all of these companies hope for is to cultivate brand loyalty among LGBT customers and thereby grow their business (Um 2014).
In contrast, Asian countries have carried out less research on their potential LGBT consumer base.The exact size of the LGBT population has yet to be confirmed.In Korea, it has long been considered a taboo to publicly discuss homosexuality.The traditional Confucian emphasis on familial bonds gave rise to a widespread perception of homosexuality as being detrimental to societal order (Power 2012).Nonetheless, consumers have broadly accepted gay-themed media content as making up a part of an important minority culture (Lyi and Park 1999).In 2005, Korea saw its first gay-themed advertising via a TV commercial.This gay-themed content stirred up not only a great deal of media attention but also of public criticism, with many consumers condemning the TV commercial for corrupting Korean social values and fostering a gay culture.In Korea, it is still uncommon to see gay-themed TV ads and print ads.
The purpose of this study is three-fold.First, the current study investigates the impacts of culture on tolerance toward homosexuality, attitude toward gay-themed ads, and attitude toward brand.Second, the study examines how gender affects tolerance toward homosexuality, attitude toward gay-themed ads, and attitude toward brand.Lastly, this study investigates consumers' responses to types of gay-themed ads (homosexual male imagery print ads vs. lesbian imagery print ads).

Attitude toward Homosexuality in the United States and South Korea
From 1973 to 1991, 70-80% of the public thought homosexuality to be wrong.However, by 2010, that number had fallen to 46% (Keleher and Smith 2012).Since 1991, public acceptance of gays and lesbians has grown dramatically (Keleher and Smith 2012).Acceptance of homosexuality along with the support of gay marriage seems to be on the rise in the United States.In 2001, according to Pew Research polling, Americans opposed same-sex marriage by margins of 57%.Since then, however, support for gay marriage has grown steadily.In 2013 52% of Americans supported gay marriage, compared to 40% who opposed it (Pew Research Center 2015).This cultural shift culminated in the decision by the U.S. Supreme Court that declared that gays have a nation-wide right to marry (Liptak 2015).
In South Korea, even if conservative attitudes continue to prevail, Koreans have by and large become significantly more accepting of homosexuality and LGBT rights.According to a 2013 Gallup poll, 39% of Koreans believed that homosexuality should be accepted by society; just six years earlier, in 2007, only 18% held this view (Gallup Korea 2013).Among the 39 countries surveyed worldwide, the country exhibiting the most significant shift towards greater acceptance of homosexuality was, in fact, Korea (Ipsos 2013).Significantly, there is a very large age gap on this issue: in 2013, 71% of South Koreans aged between 18-29 believed that homosexuality was acceptable, compared to only 16% of South Koreans aged 50 and over (Ipsos 2013).This poll result seems to suggest that South Korea is likely to become more accepting of the LGBT community over time.

Effects of Gay-Themed Ads in United States and South Korea
Studies suggest that gay-themed ads featuring a lesbian or gay male couple or other homosexual and lesbian iconography, such as a rainbow flag or a pink triangle, may lead to negative advertising and brand evaluation (Angelini and Bradley 2010;Tajfel and Turner 1986;Bhat et al. 1998).According to Tsai (2004), gay-themed advertising refers to "the ads targeting gay male and lesbian female consumers by carrying implicit or explicit homosexual references-from vaguely implying same-sex bonding, to explicitly showing self-identified gay characters-and by depicting erotic desire and affection for or between members of the same sex" (p.4).When such gay-themed advertising is placed in mainstream media, marketers tend to deem it as alienating to the general public, causing negative attitudes toward the media itself (Angelini and Bradley 2010).Fleishman-Hillard's HH Out Front Gay PR Group released a report that claimed that 24% of heterosexuals would be less likely to purchase a new or everyday product if a company used gays to market or promote it (Wilke 2007).The study found that, meanwhile, the gay-themed advertising would elicit more positive reactions among gay male and lesbian female consumers (Tuten 2005) and lead to more favorable emotional and attitudinal responses to the ad.In a similar vein, research conducted in Korea indicates that gay-themed advertising lowered heterosexual consumers' advertising and brand evaluation as well as purchase intention (Lyi and Park 1999).When heterosexual consumers were exposed to the gay-themed advertising, they showed higher levels of shock (Lyi and Kim 2000).However, gay male and lesbian female consumers showed more positive attitudes toward the ads and the brand and indicated higher purchase intention (Lyi and Kim 2000).Studies have suggested that differential effects in the evaluation of gay-themed advertising among heterosexual consumers and gay male and lesbian female consumers could be explained by social group membership and social identity theory.First, one study proposed that social groups differ systematically in their affinity, abilities, purposes, and prejudices; consequently, these groups apply a distinct reading strategy when processing an ad, and thus develop its shared reactions to, interpretations of, and meanings of the ad (Bhat et al. 1998).Second, according to social identification theory, an individual's identification with and membership in a social group allows individuals to enhance their group's social standing at the expense of another (Tajfel and Turner 1986).The favoritism shown for a group that one considers oneself a member of is likely to affect one, in a biased manner, when one processes the advertising visuals and messages.
2.3.Cultural Orientation (Individualism vs. Collectivism) and Its Effects in Tolerance toward Homosexuality, Gay-Themed Ads, and the Brand Hofstede (1980) characterized Korea as a collectivistic culture and the United States as an individualistic culture.Individuals in a collectivistic culture are likely to see themselves as interdependent within their groups and tend to behave according to collective social norms (Triandis 1995).However, individuals in an individualistic culture are likely to see themselves as independent of groups and tend to behave according to their personal choices (Triandis 1995).As for people in an individualistic culture, their self-interest carries more weight than the interests of their in-groups (Triandis 1995).It should be noted that both individualistic and collectivistic orientations co-exist within individuals and within cultures.For instance, although Korean culture is more collectivistic, it also embraces some individualistic values.
Similarly, according to prior research, each culture has its own respective and distinctive self-construals.Self-construal is defined as a generalization about the self-derived from past experience (Markus 1977;Markus and Kitayama 1991).Markus (1977) suggested that self-construals are likely to organize and guide the processing of self-related information contained in an individual's social experience.Study findings indicate that people in such Western countries as the United States and Canada hold independent self-construals, whereas people in such Asian countries as Korea and Japan hold interdependent self-construals (Markus 1977;Markus and Kitayama 1991;Triandis 1995;Kim and Markus 1999).Markus and Kitayama (1991) proposed that "the behavior of individuals with independent self-construals tends to be organized and is made meaningful by reference to one's own internal repertoire of thoughts, feelings, and actions" (p.226).However, the behavior of individuals with interdependent self-construals is likely to be guided and further determined by the thoughts, feelings, and actions of others (Markus and Kitayama 1991).As for individuals with interdependent self-construals, "the relationship between the self and others can be characterized as not separate from the social context, but as more connected and less differentiated from others" (Markus and Kitayama 1991, p. 227).
An individuals' respective self-construal and their social values play an important role when it comes to how they perceive homosexuals.People with interdependent self-construals put more emphasis on social harmony and conformity than do people with independent self-construals.Because of this, such people (interdependent self-construals) will deem homosexuality to be harmful or detrimental to society.In terms of social values, embracing diversity is broadly encouraged in the United States, a nation made up of diverse ethnic groups and cultures.Thus, it is likely that putting social values on cultural diversity in the United States may add to the acceptance of homosexuality.South Korea, in contrast, is made up of a largely homogeneous populace.In addition, the influence of Confucianism is still felt in Korean society, though it has weakened over time, suggesting that a majority of Koreans still view homosexuals as problematic and disruptive to family tradition (Power 2012).Most important, in South Korea homosexuality may not be considered to be conforming to a group value.In sum, individuals in a collectivistic culture (i.e., South Korea) are likely to have less tolerance than those in an individualistic culture (i.e., the United States).
In an advertising context, this negative attitude toward homosexuality will likely affect tolerance of homosexuality as well as attitude toward gay-themed ads and its advertised brand.The above discussion leads to the following hypotheses: Hypothesis 1 (H1).People in a collectivistic culture (i.e., South Korea) will have lower tolerance toward homosexuality than those in an individualistic culture (i.e., the United States).

Hypothesis 2 (H2).
People in an individualistic culture will have more favorable (a) attitudes toward the ads and (b) attitudes toward the brand than people in a collectivistic culture.

Effects of Gender in Tolerance toward Homosexuality, Gay-Themed Ads, and the Brand
Gender-role beliefs are linked to sexual prejudice because homosexuality is perceived as a gender-role violation.According to the gender-role belief system, men are more likely to support, to a greater degree, traditional gender-role beliefs than women.In sum, men consider homosexuals to be gender-role deviants whereas women are likely to be more flexible regarding gender-roles and less inclined to reject gender-role deviants (LaMar and Kite 1998).
Prior research indicates that since heterosexual females are less likely to reject lesbians and homosexual men, they feel little social pressure to be hostile to homosexuals (Herek 1984).However, heterosexual males consider heterosexuality to be an important signifier of their masculinity.Thus, they feel obliged to affirm their masculinity by rejecting homosexual men who violate the heterosexual norm (Herek 1984).Men's greater intolerance of homosexuality than that of women may be derived from the gender belief system.According to Kite and Whitley (1996), "gender-associated traits and gender-associated physical characteristics are more narrowly defined for men than for women and violation of the traditional male gender role is seen as more egregious than violation of the traditional female gender role" (p.41).In general, people tend to show more negative reaction to boys who possess female-type traits than to girls who possess male-type traits (Feinman 1981).
In short, gender-role beliefs propose that heterosexual males will condemn homosexual men for violating the male gender code.Heterosexual females are under less social pressure to criticize gender-role violations.Consequently, heterosexual males are more likely to express homophobic attitudes (LaMar and Kite 1998).In a similar vein, one study suggested that women were more tolerant, and by a good measure, of homosexuality (Adolfsen et al. 2006).In an experimental study, Kerns and Fine (1994) found that males showed more negative attitudes toward homosexual men than females; there were no gender differences in attitudes toward lesbians.Thus, the following hypotheses are proposed: Hypothesis 3 (H3).Female consumers will have higher tolerance toward homosexuality than male consumers will.

Hypothesis 4 (H4). Hypothesis 4 (H4)
.Female consumers will have more favorable (a) attitude toward the ads and (b) attitude toward the brand.

Types of Gay-themed Ads: Homosexual Male Imagery Print Ads vs. Lesbian Imagery Print Ads
According to the 2014 Gay Press Report, gay-themed ads continues to rise in the United States (Rivendell Media 2014).Gay-themed ads usually feature a homosexual male or lesbian couple in advertising.Thus, in the current study, types of gay-themed ads could be classified as either homosexual male imagery ads or lesbian imagery ads.Gay-themed ads are more likely to be seen in gay media, in such magazines as Curve, Out, and The Advocate, and on TV cable channels such as Logo TV.Today, however, it is not difficult to come across gay-themed ads in mainstream media.For instance, Amazon.com, in promoting their Kindle, put out a TV commercial (seen on America's broadcast networks) featuring a good-looking homosexual couple at a beach resort (Italie 2013).It seems that marketers are trying to reach the vast majority of homosexual consumers through gay media because of higher reach and readership of mainstream than gay media (Oakenfull and Greenlee 2005).According to one report, gay media reaches less than 50% of the LGBT population, whereas mainstream media reaches more than 90% of it (Taylor 2013).
Individuals' attitudes toward homosexual men and lesbians will affect how they evaluate homosexual male imagery and lesbian imagery print ads.In general, heterosexuals are likely to have more positive attitudes toward lesbians than toward homosexual males (Oakenfull and Greenlee 1999).Oakenfull and Greenlee (2004) found that heterosexual consumers more favorably view ads depicting lesbians than those depicting homosexual males.In addition, heterosexual males could perceive gay-themed ads featuring lesbian couples as "erotic," leading to more favorable ad and brand evaluation than those featuring a homosexual male couple (Louderback and Whitley 1997).Thus, it is plausible to assume that these attitudes will influence heterosexuals evaluating homosexual male imagery and lesbian imagery print ads.
Hypothesis 5 (H5).Types of gay-themed ads will have impact on (a) consumers' attitude toward the ads and (b) attitude toward the brand.

Subjects, Designs, and Procedure
Participating in this cross-cultural study were a total of 457 college students from a major university in America's Southwest and from a major private university in Seoul, Korea.Participants were recruited to take part in an Internet-based experiment.All participated in return for extra course credit.Eleven participants in the United States and seven in Korea identified as either homosexual or lesbian, so after removing them, 439 participants remained for further analyses.Of the 439 participants, 232 were from the United States and 207 were from Korea.Of the 232 U.S. participants, 157 were female (67.7%) and 75 were male (32.3%).Their mean age was 21.8.In terms of the ethnicity, the majority were Caucasians (n = 214, 90.3%), followed by Asian (n = 11, 4.6%), Hispanics (n = 5, 2.1%), others (n = 4, 1.7%), and American Indian (n = 1, 0.4%).Among the 207 Korean participants, 117 were female (56.5%) and 90 were male (43.5%).Their mean age was 23.5.
College students were selected for this study because heterosexism and homophobia are powerful forces that have pointed effects on students; researchers have identified their many faces on college campuses (Fine 2011).In addition, the Gay, Lesbian, and Straight Education Network's 2011 National School Climate Survey reports over a third of LGBT students faced physical harassment and about 20 percent were assaulted in 2012 (Bonauto and Robbins 2013).Nevertheless, many aspects of heterosexism and homophobia remain unexplored in educational institutions.
This study employed a 2 × 2 × 2 between-subject factorial design.The three factors were cultural orientation (collectivism vs. individualism), gender (male vs. female), and types of gay-themed ads (gay male imagery vs. lesbian imagery print ads).To test the proposed hypotheses, separate analyses were run.For instance, ANOVA was run to test the proposed H1 and H3, whereas MANCOVA was conducted to test the proposed H2, H4, and H5, with three covariates such as brand familiarity, brand involvement, and brand knowledge.In the current study, tolerance toward homosexuality was measured as a continuous variable (as a dependent variable).Later, this variable was median-split into a low tolerance group and a high tolerance group to see if there was any difference between these two groups in terms of attitude toward the ad and attitude toward the brand.
This study was conducted online and an online survey website was created through Qualtrics.Participants were given an informed consent notice and then directed to either the gay male imagery print ad condition or to the lesbian imagery print ad condition.Before participants were exposed to their respective stimuli, their cultural orientation and tolerance of homosexuality were measured in order to prevent any bias likely to be derived from the gay male or lesbian female print ads.Finally, participants were asked to answer demographic questions pertaining to age, gender, sexual orientation, religiosity, and years in college.

Experimental Stimuli and Pre-Test
For experimental stimuli, this study selected hotel print ads because a large portion of print ads in gay and lesbian magazines fall within the hotel category (Rivendell Media 2014).A gay male imagery print ad was selected from a popular gay magazine and a lesbian imagery print ad was selected from a popular lesbian magazine.Both ads were designed for the same hotel brand, thus sharing the same headline and body copy.Korean versions of the gay male imagery and lesbian imagery print ads were electronically manipulated after a back-translation into Korean was made of the headline copy.The gaymale imagery print ad featured two men looking into each other's eyes in a hotel room; the lesbian imagery print ad looked the same except for the substitution of two women looking into each other's eyes.
Thirty college students participated in a pre-test.The pre-test was conducted to ensure that subjects perceived the two types of print ads as the ads were intended to be perceived (either a gay male imagery or lesbian imagery print ad).Subjects were given a statement that declared an opinion about the print ad (i.e., "In my opinion, this print ad is clearly targeting gay male or lesbian female consumers"), which was measured on a 7-point scale anchored with "strongly disagree" and "strongly agree."The one sample t-test results showed that the gay male imagery print ad was perceived as a gay-targeting ad (M = 6.03,SD = 0.81; t(29) = 13.77,p < 0.001) and the lesbian imagery print ad was perceived as a lesbian-targeting ad (M = 5.6, SD = 0.67; t(29) = 12.99, p < 0.001).It was thus clear that the manipulation was successful.

Measures
Cultural orientation: Individualism and collectivism.To measure subjects' cultural orientation, researchers employed a 17-item scale (α = 0.75) from Hui (1988).Hui reported a consistent pattern of construct validation.One measure that researchers widely use to gauge feelings, beliefs, intentions, and behaviors consistent with either an individualistic or a collectivistic cultural orientation is the individualism-collectivism scale (INDCOL; Hui 1988).The scale asks respondents to indicate on a five-point, Likert-type scale their level of agreement or disagreement with each item (1 = strongly disagree to 7 = strongly disagree).The present study used an overall mean score of the participants to determine their general level of collectivism or individualism.A score of more than 4 suggested that a respondent could be categorized as an individualist; a score of less than 4 suggested that a respondent could be classified as a collectivist; and a score of 4 indicated neutrality.
Tolerance toward homosexuality.Tolerance of homosexuality was measured by 9 items, modified from the 64 in Herek (1984) condemnation-tolerance (C-T) scale.The C-T scale captures the essence of definitions of homophobia.These nine items, selected based on their high factor loadings in Herek's research, were all in excess of 0.70.The reliability of the scale-selected C-T items in this study was high (α = 0.90).Subjects were asked to express their agreement with the statements on a 7-point scale, anchored by strongly disagree and strongly agree.
Attitude toward the ad and the brand.Advertising affects consumers' attitudes toward an ad and toward a brand (Teng and Laroche 2006).Attitude toward the ad was measured using three, 7-point semantic differential scales: very bad-very good, and very unfavorable-very favorable, like very much-dislike very much (α = 0.92, MacKenzie and Lutz 1989).Attitude toward the brand (Ab) was measured using five, 7-point semantic differential scales: unappealing-appealing, bad-good, unpleasant-pleasant, unfavorable-favorable, and unlikable-likable (α = 0.91; Spears and Singh 2004).
Covariates: Brand familiarity, product involvement, and product knowledge.Covariates were measured to control for extraneous variation in the data using analysis of covariates.Brand familiarity, product knowledge, and involvement were included as potential covariates because research has suggested that these factors may have an effect on brand evaluation (Chebat et al. 2001;Kardes 1988;Sawyer and Howard 1991).The respondents indicated, on a single bipolar scale, whether they were familiar or unfamiliar with the brand.Brand familiarity was measured using the following item: "The brand is . . .extremely unfamiliar (1)-extremely familiar (7) to me." Product knowledge was measured on four, 7-point scales (know very little-know very much, inexperienced-experienced, uninformed-informed, novice buyer-expert buyer; α = 0.92).Product involvement was measured on four, 7-point scales (unimportant-important, irrelevant to me-relevant to me, means nothing to me-mean a lot to me, not needed-needed; α = 0.93).Since product involvement (Wilks' Lambda = 0.99, F = 1.21, p < 0.05) did not significantly affect the dependent variables, this covariate was dropped from further analyses.As shown in Table 1, brand familiarity (Wilks' Lambda = 0.97, F = 5.19, p < 0.01) and brand knowledge (Wilks' Lambda = 0.95, F = 5.49, p < 0.01) had a statistical significance on the dependent variables and thus remained in further analysis.

Cultural Orientation: Individualism vs. Collectivism
This study assumed Koreans to be collectivistic and Americans individualistic.To ensure the correctness of this assumption, the study employed Hui (1988) INDCOL scale to measure subjects' cultural orientations.Of the 232 U.S. subjects, 165 (71.1%) were found to be individualistic, while 67 (28.9%) were found to be collectivistic.Of the 207 Korean subjects, 46 (22.2%) were found to be individualistic, while 161 subjects (77.8%) were found to be collectivistic.The chi-square result indicates that the Americans were more likely to be individualistic while the Koreans were more likely to be collectivistic.The difference was statistically significant (χ 2 = 104.78,p < 0.001).For further analyses, the researchers ruled out the 67 U.S. subjects who identified as collectivists and the 46 Korean subjects who identified as individualists.
H1 posits that collectivists (Koreans) will have less tolerance of homosexuality than will individualists (Americans).Thus, H1 proposes that collectivists (Koreans), when exposed to gay-themed ads, will have less favorable attitudes toward the ad and the brand than will individualists.ANOVA was run to test the proposed H1.As Table 1 indicates, cultural orientation has a statistically significant effect on tolerance of homosexuality [F(1, 328) = 183.35,p < 0.001].Tolerance was significantly lower for collectivists (M = 3.30, SD = 1.95) than for individualists (M = 5.74, SD = 1.26).As shown in Table 2, there was a significant main effect of cultural orientation (Wilks' Lambda = 0.82, F = 35.62,p < 0.001).The main effect indicates that cultural orientation has substantial independent effects on attitude toward the ads and attitude toward the brand.As seen in Table 3, univariate analyses indicate that cultural orientation had an effect on attitude toward the ad (F = 54.89,p < 0.001) and on attitude toward the brand (F = 59.65, p < 0.001).Table 4 shows cell means for dependent variables.An examination of the mean differences between the United States (individualism) and Korea (collectivism) suggested that U.S. subjects have more favorable evaluations of the advertising (Mean U.S. Subjects = 4.50 vs. Mean Korean Subjects = 3.56) and the brand (Mean U.S. Subjects = 5.07 vs. Mean Korean Subjects = 4.16).Thus, H1, H2a, and H2b were supported in this study.Furthermore, as can be seen in Table 4, significant two-way interactions occurred between cultural orientation (individualism vs. collectivism) and ad types (gay male print ad vs. lesbian female print ad; Wilks' Lambda = 0.93, F = 11.62,p < 0.001).As can be seen in Table 3 and Figures 1 and 2, univariate analyses indicated that two-way interaction between cultural orientation and ad types affected advertising evaluation (F = 15.27,p < 0.001) and brand evaluation (F = 6.32, p < 0.05).Furthermore, as can be seen in Table 4, significant two-way interactions occurred between cultural orientation (individualism vs. collectivism) and ad types (gay male print ad vs. lesbian female print ad; Wilks' Lambda = 0.93, F = 11.62,p < 0.001).As can be seen in Table 3 and Figures 1 and  2, univariate analyses indicated that two-way interaction between cultural orientation and ad types affected advertising evaluation (F = 15.27,p < 0.001) and brand evaluation (F = 6.32, p < 0.05).H3 proposes that males will have less tolerance of homosexuality than will females.Thus, H3 posits that males, when exposed to gay-themed ads, will have less favorable attitudes toward advertising and brand than will females.As Table 5 indicates  4.2.Effect of Gender in Tolerance toward Homosexuality, Attitude toward Ad, and Brand H3 proposes that males will have less tolerance of homosexuality than will females.Thus, H3 posits that males, when exposed to gay-themed ads, will have less favorable attitudes toward advertising and brand than will females.As Table 5 indicates, consumers' gender has a statistically significant effect on tolerance of homosexuality [F(1, 328) = 6.11, p < 0.05].Tolerance was significantly lower for males (M = 4.20, SD = 2.04) than for females (M = 4.76, SD = 2.02).As Table 2 indicates, there was no significant main effect of gender (Wilks' Lambda = 0.99, F = 0.17, p < 0.05).Univariate analyses also indicate, as shown in Table 3, that gender does not have effect on advertising evaluation (F = 0.261, p > 0.05) and brand evaluation (F = 0.009, p > 0.05).Thus, H3 was supported while H4a and H4b were not.H5 posits that gay male imagery print ads will lead to less favorable attitudes toward ads and brand than will lesbian imagery print ads.As Table 2 indicates, there was a significant main effect of ad types (Wilks' Lambda = 0.98, F = 3.90, p < 0.05).As shown in Table 3, univariate analyses indicate that ad type had an effect on advertising evaluation (F = 18.43, p < 0.05) and brand evaluation (F = 6.44, p < 0.05).Examining the mean differences between the gay male imagery and lesbian imagery print ads suggested that the lesbian imagery print ads lead to more favorable advertising evaluation (Mean gay male print ad = 3.85 vs. Mean lesbian female print ad = 4.20) and brand evaluation (Mean gay male print ad = 4.53 vs. Mean lesbian female print ad = 4.70).Thus, H5a and H5b were supported in this study.

Discussion
The objective of the present study has been to investigate the effects of cultural orientation, gender, and types of gay-themed ads on attitudes toward gay-themed advertising and the advertised brand.The study findings highlight the importance of cultural orientation in engendering tolerance of homosexuality and shaping attitude toward gay-themed ads, and the brand advertised.The study found that Korean college students (i.e., collectivists) had lower tolerance of homosexuality than did U.S. college students (individualists).It is noteworthy that, when it comes to tolerance toward homosexuality, the gap between Korean subjects and U.S. subjects was far greater than expected.On a seven-point scale, U.S. subjects scored 5.88, which can be viewed as a fairly high tolerance of homosexuality; Korean subjects scored a fairly low 3.36.This trend was also revealed in individuals' attitudes toward gay-themed ads and its advertised brand, suggesting that Korean subjects had less favorable attitudes than their American counterparts toward gay-themed ads and the advertised brand.
The results suggest that it may be premature for the Korean advertising industry to roll out gay-themed ads.In fact, gay-themed advertising could bring on a backlash effect for the advertised brand.
Regarding the effects of gender, the study yielded mixed results.As posited in this study, females had higher tolerance of homosexuality than males.The result supported the notion that gender-role beliefs lead males to demonstrate a lower tolerance of homosexuality (Herek 1988;Oakenfull and Greenlee 2004;Kite and Whitley 1996).However, gender did not influence attitudes toward the gay-themed advertising or the advertised brand.No statistical significances were found regarding advertising and brand evaluation between males and females.The study results suggest that an individuals' level of tolerance of homosexuality does not necessarily influence their evaluation of gay-themed ads or its advertised brand.
The results of this study imply that lesbian imagery print ads could lead to greater tolerance of homosexuality and more favorable evaluations of the advertising and brand than could gay male imagery print ads.The study findings support the prior literature suggesting that heterosexuals tend to have more positive attitudes toward lesbians than toward gay males (Oakenfull and Greenlee 1999).It is carefully surmised that this attitude toward gay males and lesbians may have been reflected in the reactions to the gay male and lesbian print ads.Participants may have perceived the lesbian print ad, as has been suggested in the literature, as erotic (Louderback and Whitley 1997).Thus, heterosexual consumers will more positively evaluate a lesbian imagery print ad than a gay male imagery print ad.It is worth noting, also, that a lesbian imagery print ad could be perceived as a normal ad featuring intimacy between two females.

Limitations and Areas for Future Research
The results of this study should be considered in light of several limitations.First, this study limited its responses to college students in the United States and Korea.University students may, as an aggregate, have different perceptions of homosexuality than the general population.For future research, broadening samples demographically by including various age groups could be useful in investigating how consumers perceive gay-themed advertising in the United States and Korea.
To make the study results more representative and generalizable, it is essential to use the general population in the future.To enhance the generalizability of the study's findings, future researchers might consider using Amazon Mechanical Turk (Mturk, Seattle, WA.United States).
Second, this study used a real brand name to lend a sense of realism to subjects participating in the study, instead of using a fictitious brand.However, the use of a real brand name could result in consumer bias in processing the advertising message.For a future study, in order to control consumers' preexisting biases toward a brand, it may be desirable to use a fictitious brand.In addition, the product category used in this study might be less relevant to study participants.If researchers use college student samples for future research, it is advised that they consider using a unisex product category related to college students.
A further limitation is the stimuli employed in this study.The gay male imagery and lesbian imagery print ads were selected from gay and lesbian magazines published in the U.S. The print ads were then manipulated for Korean samples.Models used in the print ads are Caucasians.Social identity theory suggests that an individual's identification with and membership in a social group allows them to enhance their group's social standing at the expense of another (Tajfel and Turner 1986).In an advertising context, U.S. college students may have higher identification with models in print ads than Korean college students.This could thus lead to favorable advertising and brand evaluation.In addition, since gay-themed advertising can be frequently found, not only in gay media but in mainstream media in the United States, U.S. college students are more acclimated to gay-themed advertising than are Korean college students.The LGBT population in Korea is too small to target.The LGBT market is not as profitable in Korea as it is in the United States.That is why Korean audiences have far less exposure to the LGBT community compared with their counterparts.This media exposure could lead to different responses to gay-themed advertising.
Finally, the current study examined the effects of gay-themed ads only among young heterosexual consumers in Korea and in the United States.It might be more difficult to recruit young LGBT consumers than young heterosexual consumers in both countries.However, it would be worthwhile to compare young heterosexual consumers' responses with those of their LGBT counterparts when it comes to evaluating gay-themed ads.Previous literature suggests that LGBT consumers are likely to have more favorable attitude toward the ads and brands when the ads are targeting them (Lyi and Park 1999;Tuten 2005;Angelini and Bradley 2010).

Figure 1 .
Figure 1.Interaction between cultural orientation and ad type on advertising evaluation.

Figure 1 .
Figure 1.Interaction between cultural orientation and ad type on advertising evaluation.

Figure 2 .
Figure 2. Interaction between cultural orientation and ad type on brand evaluation.4.2.Effect of Gender in Tolerance toward Homosexuality, Attitude toward Ad, and Brand , consumers' gender has a statistically significant effect on tolerance of homosexuality [F(1, 328) = 6.11, p < 0.05].Tolerance was significantly

Figure 2 .
Figure 2. Interaction between cultural orientation and ad type on brand evaluation.

Table 1 .
ANOVA Results: Effect of Cultural Orientation on Tolerance toward Homosexuality.

Table 2 .
MANCOVA Results: Effects of Cultural Orientation, Gender, and Ad Types.

Table 3 .
Tests of Between-Subject Effects.

Table 4 .
Subjects' Means and Standard Deviations for Ad Evaluation, and Brand Evaluation.

Table 4 .
Subjects' Means and Standard Deviations for Ad Evaluation, and Brand Evaluation.

Table 5 .
ANOVA Results: Effect of Gender on Tolerance toward Homosexuality.Types of Gay-Themed Ads: Gay Male Imagery Print Ads vs. Lesbian Female Imagery Print Ads