Effects of gender reactions to stereotype advertisements – Case study in Bulgaria

This investigation has two aims: to establish the presence of stereotypes in advertisements offered in Bulgaria and to identify which stereotypes achieve positive and negative effects on consumers of both genders. Taking into account Eisend's theoretical formulation of gender roles in advertising (2019), as well as research on the cross-gender effects of stereotypes in advertising according to the most recent studies of Akestam et al. (2021), the author examines the influence achieved through three of the components – trait descriptors, physical characteristics and role behaviors. The results of the analysis of variance (ANOVA) show that according to trait descriptors for women, as well as role behavior for men, the highest influence was achieved in the investigated advertisements. When measuring cross-gender influences, no statistically significant differences were found between men and women in the influence of advertising containing images of the same gender. Conducted Student-Fischer’s T-tests evidenced influence achieved for advertisements with male and female images, as well as with female product images. This study can serve organizations offering products and services in Bulgaria, and its originality is expressed in the mixed consumer reactions found.

consumers of both genders. Taking into account Eisend's theoretical formulation of gender roles in advertising (2019), as well as research on the cross-gender effects of stereotypes in advertising according to the most recent studies of Akestam et al. (2021), the author examines the influence achieved through three of the componentstrait descriptors, physical characteristics and role behaviors. The results of the analysis of variance (ANOVA) show that according to trait descriptors for women, as well as role behavior for men, the highest influence was achieved in the investigated advertisements. When measuring cross-gender influences, no statistically significant differences were found between men and women in the influence of advertising containing images of the same gender. Conducted Student-Fischer's T-tests evidenced influence achieved for advertisements with male and female images, as well as with female product images. This study can serve organizations offering products and services in Bulgaria, and its originality is expressed in the mixed consumer reactions found.

Introduction
Although advertising researchers have conducted a number of studies on the gender role of stereotypes, there are many ambiguities such as: what are the main stereotypes of both genders; which stereotypes are falling away and which are confirmed; as well as what is expressed in the high evaluation of stereotypes in advertising by consumers. According to Desvaux, Devillard-Hoellinger and Baumgarten, if precise reasons for these features are not formed, the phenomenon of "stereotype in advertising" will continue to exist and be studied (Desvaux, Devillard-Hoellinger and Baumgarten, 2007, p. 44). All this necessitates a content analysis regarding the influence of stereotypes, as well as establishing the achieved positive and negative impact on both genders. The portrayal of male and female figures in advertising, the following of certain stereotypes, as well as the imposition of certain norms of behavior for the genders, cause different reactions of consumers for the two genders. The traditionally repeated stereotypical image has one advantageit makes consumers' thinking more lazy, because "stereotypes are created by the continual, extended exposure of consumers to patterns of imagery" (Paek, Nelson & Vilela, 2011, p.197). Advertisers portray men and women differently in advertisements, but these different images lead to intended and unintended effects (Baeza, 2011). Internationally, advertising conforms to traditional gender images (Paek, Nelson & Vilela, 2011, p.193), that is, not mixing of the genders, demonstrations with same-sex couples and marriages were observed. Even if one media can afford an advertisement with a gay couple, the trend does not continue in other issues of the media, which has the mechanisms to confirm stereotypical behavior (Phillips & McQuarrie, 2011). Therefore, advertisers are hesitant to what extent they can use stereotypical images to achieve influence. Paek, Nelson and Vilela state that "men are represented in prominent visual and auditory roles, while women are portrayed in stereotypical ways" (Paek, Nelson and Vilela, 2011, p.192). Offering new current research on the stereotypical presence of women and men in advertisements, it is expected to shift the perspectives of both users and researchers (Eisend, 2019;Akestam, Rosengren, Dahlen, Liljedal & Berg, 2021), and according to the researcher of the present work, to cause different psychological reactance (Brehm, 1966;Thorbjørnsen & Dahlén, 2011) in the users. Therefore, the different images, the implied images of the two genders, as well as their combined presentation with different products and services, can determine some of the final results in the supply and demand of the advertised goods.
The present article, which aims to explore crossgender responses to stereotypical advertising influence as part of consumer reactance, is organized as follows: the first section reviews previous literature on stereotype terminology and understanding. In accordance with this review, the hypothesis regarding the influence of gender on stereotypical advertising is formed. It then describes the methodology used in the study and its main findings on stereotypical influence measured by gender. Finally, a discussion of the obtained results and a summary of the conclusions of the study are made.

Theoretical Framework
In order to outline the social effects on consumers and society, researchers find that the relationship between gender roles (Eisend, 2019;Hatzithomas, Boutsouki & Ziamou, 2016), stereotypes in advertisements (Taylor & Stern, 1997;Knoll, Eisend & Steinhagen, 2011) should be explored, but also the role of brand and mediarelated social effects (Eisend, 2019;Wolin & Korgaonkar, 2003). According to the new advertising trends, the effect of gender stereotypes is studied, that is, these are binary definitions of gender according to which study the effect of gender stereotypes on consumers of the same (binary) gender (Eisend, 2019, p.78). Difficulties in research come from the fact that the concept of gender becomes more nuanced in different advertisements and societies.
According to researchers such as Akestam, Rosengren, Dahlen, Liljedal and Berg (2021), such studies "suffer from common terminology" or use the terminological framework of other studies such as: personality traits (Lee & Ashton, 2004;Markey & Markey, 2009;Johnson, 2014), the concepts of idealization (Richins, 1991), body size (Bian & Wang, 2015), the concept of attractiveness (Bower, 2001), which shift the perspective from gender portrayals in media and advertisements to consumer attitudes and behaviors.
In order to form a terminological framework of research, the concepts of "stereotype" and how it is present in advertising must be determined. Stereotypes facilitate the perception of advertising (Macrae, Milne & Bodenhausen, 1994), allowing the perceiver to rely on previously stored knowledge rather than new information. According to Taylor and Stern (1997), stereotype is related to cultural context, the most widely applied cultural model in advertising research being Hofstede's (2001) cultural dimensions. The understanding of a stereotype also includes "a widely accepted belief about the personal qualities of members of a social category, such as gender, ethnicity, or sexual orientation" (Akestam, 2017, p.16). Other difficulties in stereotype research arise from their dynamics, because stereotypes have the ability to change over time (Eisend, 2010), and also different stereotypical advertising images can reinforce the role of public stereotypes (Eisend, Plagemann & Sollwedel, 2014). In addition, Knoll, Eisend and Steinhagen emphasize the social role of stereotype as "a set of concepts relating to social category" (Knoll, Eisend and Steinhagen, 2011, p.869). At the heart of the stereotypical division of the genders is Ashmore and Del Boca's (1981) claim that certain characteristics distinguish men from women.
As a result of logical and substantive summary, the following research questions can be identified, which are the subject of consideration by the researchers (Table 1). Table 1.

Research questions
Stereotypical influence / Authors, Year 1. How are gender stereotypes perceived from advertising and how does this affect information processing?
Gender stereotypes function as a type of heuristic. Stereotypes in advertising facilitate perception because they require less cognitive processing effort (Pratto & Bargh, 1991).
2. What are the negative effects of exposure to gender stereotypes?
Anxiety of women and girls as stereotypical images of women tend to portray weak women (Halliwell & Dittmar, 2004). Stereotypical images of men have similar negative effects on men's body satisfaction, self-esteem, and self-esteem because they portray athletic and muscular men (Lorenzen, Grieve, Thomas, 2004). 3. Is gender stereotyped in advertising and does the degree of stereotyping decrease or increase over time?
In the stereotyping of gender in advertising are offered attributes that differentiate males and females (Ashmore & Del Boca, 1981). These attributes may be trait descriptors, physical characteristics, role behaviours or occupational statuses (Deaux & Lewis, 1984).

Are depictions of gender roles deviating from equality and what is happening to gender roles?
The portrayal of gender roles deviates from equality because there are different standards of equality for men and women. (Eisend 2010;Knoll, Eisend & Steinhagen, 2011). Changes related to gender values are observed (Eisend, 2010;Matthes, Prieler & Adam, 2016). 5. What are the social effects of gender stereotypes in advertising?
Gender stereotyping through advertising reduces women's professional performance, achievement aspirations, and positive selfperception and thus challenges the provision of equal opportunities for both sexes in society (Davis, 2002). The changing role of men from strong masculinity to femininity, as well as the reverse change for women, leads to a mixing of gender roles. Negative social effects for the sexes have been reported due to the adoption of different roles as well as the mixing of gender functions (Maher & Childs, 2003). Source: own elaboration.
According to Deaux and Lewis (1984), as well as Eisend (2019), gender stereotypes have four distinct components that are distinguishable at their core: trait descriptors (e.g., assertiveness, concern for others), physical characteristics (e.g., hair length, body height), role behaviors (e.g., leader, childcare), and occupational status (e.g., truck driver, elementary school teacher, homemaker) (Deaux & Lewis 1984;Eisend, 2019, p.73). Current research focuses on trait descriptors, physical characteristics, and role behaviors, delineating several problem areas that need to be clarified. It is assumed that women will have more pronounced negative reactions to the presentation of stereotypical images of women, derived from the mass advertisements on the Internet, offered to consumers in Bulgaria for evaluation. The assumption is possible because in the research circles the opinion is formed that the woman in advertisements is more often placed in a subordinate role, and when she shows power through a career, she copies the behavior of the man. In an analysis of media influence, it is found that this is due to the presentation of stereotypical images, which lead to higher levels of perceived negative influence on other usersmen and women, which determines the reactance of users to advertising. (Akestam, Rosengren, Dahlen, Liljidal & Berg, 2021, p. 67). On the other hand, men, as a less sensitive media audience (Spasova & Taneva, 2021), are shaped as consumers with less pronounced levels of negative affect from internet advertisements presenting stereotypical images of men in the same companies' advertisings. The main theoretical explanations for men's and women's negative advertising response to stereotypical versus non-stereotypical presence in advertisements are related to theory of psychological reactance (Brehm 1966;Thorbjørnsen & Dahlén, 2011), which posits that people are predisposed to maintain and restore their personal freedom when it is threatened. A similar threat is observed in both genders as a reactance against a stereotypical advertising image that limits consumers. In addition, the research should track respondents' identification with one or the other gender in relation to their final reaction to stereotypical images of women or men. The research community lacks additional meta-analyses showing aggregated information on the importance of both genders (Eagly, 2009;Shield, 2016) in the perception of stereotypical images in advertising. Wolin (2003) found that advertisements have become less stereotypical in recent years, and selectivity hypotheses suggest that women process advertisements differently compared to men (Spasova & Taneva, 2021). This statement is consistent with recent research showing that consumers, regardless of their gender, respond more positively to nonstereotypical gender representation in advertising (Baxter, Kulczynski & Illicic, 2016;Liljedal, Berg & Dahlen, 2020).
The parameters of the stereotypical representation of images of men and women in advertisements that are investigated in the present paper need to be delineated. The researcher makes a completely targeted selection of mass advertisements on certain social platforms -Facebook and Instagram, depicting images of men and women with saturated sexuality, attractiveness, good muscles, strongly associated with products intended only for men or only for women. Several stereotypical variables were selected to analyse the content of advertisements in the indicated social platforms: trait descriptors -attachment to the other gender, closeness, intimacy in stereotypical bisexual advertising; protection over the neighbour in stereotypical bisexual advertising; commitment and fidelity in stereotypical bisexual advertising; physical characteristics -length and position of the woman's hair in a mass advertisement for an Internet service; musculature and physical attractiveness of the man and the woman in a perfume advertising; sexiness of body position in perfume advertisements of famous brands; role behaviourscouple roles in stereotypical bisexual media advertising; a loving couple in an advertisement for famous brand perfumes. The variables mentioned above are compared, and advertisements in which only one of the genders is present are also offered in a completely targeted manner. The main criteria by which advertisings are measured is the presence of one or both genders, establishing the degree of impact on users who identify themselves as one or the other gender.
According to these previous researches, the researcher puts forward the following hypotheses: Hypothesis 0: Consumer gender does not determine positive and negative advertising responses to stereotypical images. Hypothesis 1: Consumer gender determines positive and negative advertising responses to stereotypical images.
In view of previous research on the effectiveness of advertisements achieved through stereotypic influence, this article seeks to empirically demonstrate the results of gender and crossgender consumer responses. The reason for conducting the study is that research on the influence of the studied reactions as part of consumer reactance is extremely limited, and the results of the studies would supplement the scientific information on the researched topic.

Methodology
The methodological basis of the research is the general scientific principles for systematizing and summarizing the results of research on the achieved effects of stereotypical influence. The purpose of the present study is to examine the influence of gender on consumer susceptibility to stereotypical images, as well as the resulting positive and negative consumer reactions. In the study, an Internet computer-based survey management survey with the following modules was implemented: 1) first module -demographic characteristics of the respondents; 2) second modulephysical characteristics with images of men and women in advertising -10 statements; 3) third module -trait descriptors with images of men and women in advertising -10 statements; 4) fourth modulerole behaviour with images of men and women in advertising -10 statements. Respondents must accept or reject statements about male, female, or both-gendered ads. This scale contains 30 items, organized in 3 subscales (10 items in each subscale with good validity and consistency), corresponding to the stereotypical images in the advertisement: physical characteristics (Cronbach α 0.56), trait descriptors (Cronbach α 0.69), role behaviour (Cronbach α 0.61). The stated Cronbach α values can be considered acceptable for scales up to 10 items (Ganeva, 2016). The validity of the whole questionnaire (Cronbach α 0.71) also shows good reliability of the scale for the sample studied. The respondent answered a total of 30 questions on a five-point Likert-type scale, which includes grades from 1 -I do not agree, to 5 -I agree.

Results and Discussion
The survey was conducted from the beginning of 2021 to the beginning of 2022. Self-reported data was collected from a total sample of 356 students from Bulgarian, ensuring a size for a representative 95% (being e = ± 5%; p = q = 0.50). Respondents were randomly selected from the general population according to their selfreported gender. All respondents were asked by e-mail to complete an online questionnaire. About 2% of the students did not indicate their gender: "prefer not to say", and their answers were not included in the study, because according to the requirements of stereotypical influence studies, gender is a determining factor. The total sample comprised 46.91% are men (167 people) and 53.09% are women (189 people), aged 18 to 65 years old (M=32.41).
Data were processed with the statistical analysis package SPSS 19.0. The following basic statistical analyses were used: One-way analysis of variance ANOVA to examine the influence of gender on gender stereotypes as well as crossgender responses to advertising influence. Student-Fischer t-test for the statistical significance of the differences between the means for the groups formed by gender and between the variables comprising the stereotypic influence scales.
In order to establish the achieved influence of gender stereotypes in advertising on the groups of respondents, one-factor analysis of variance was applied (Table 2). The results show that gender as an independent variable affects the gender stereotypes in the advertisements studied, as well as the general susceptibility to stereotype influence. For the sample formed, there are statistically significant gender differences in susceptibility to gender stereotyping in advertisements. The highest influence of trait descriptors was found in women (F = 4.65; p <0.03; Means(female) =4.21), and the highest influence of role behaviour in men (F = 3.24; p <0.00; Means (male)=3.41). On the other hand, the weakest influence was derived for role behaviour in women (F = 3.24; p < 0.00; Means(female) = 3.37), and the weakest influence of trait descriptors in men (F = 4.65; p < 0.03; Means(male) =2.76). Therefore, men declare the most positive reactions to advertisements with stereotypes based on role behaviour, and in contrast, women show positive reactions to advertisements with stereotypes based on trait descriptors. The explanation of these results is in support of other empirical evidence, which states that when confirming stereotypical behaviour (Phillips & McQuarrie, 2011), men in advertisements are more often depicted in role behaviour than women, which reinforces the role of societal stereotypes (Eisend, Plagemann & Sollwedel, 2014). In addition, the obtained result does not indicate that gender deviates from equality, because women responded positively to trait descriptors that examine concern for a close person, affection, and others.
In order to derive evidence regarding the crossgender influence of advertisements on both genders of users, one-factor analysis of variance was applied, looking for users' reactions to images of the opposite gender (Table 3). Gender as an independent variable was again an influencing factor in achieving the cross-gender effect through advertising, with users of both genders having high values regarding images of the opposite gender: for men (F = 5.69; p < 0.00; Means(male) = 4.02), and in females (F = 4.21; p < 0.00; Means(female) = 4.10). Therefore, the presence of opposite-gender images does not elicit reactance against a stereotypical advertising image (Brehm, 1966;Thorbjørnsen & Dahlén, 2011). This empirical evidence supports the views that advertisements aimed at only one of the genders would achieve a greater effect among consumers in Bulgaria than advertisings aimed at both genders (Spasova &Taneva, 2021).
Another important indicator of establishing the presence or absence of a negative reaction is the low values obtained from the one-factor analysis of variance regarding the respondents' advertising response to advertisements containing images of their gender: in men (F = 4.21; p < 0.00; Means (male) =2.38), and in females (F =5.69; p <0.00; Means(female) =1.28). The obtained results are empirical evidence of the achieved weak influence of stereotypical advertising intended for the same gender. Many researchers of stereotype influence in advertising find negative reactions of women to stereotypes of the female gender (Akestam, Rosengren, Dahlen, Liljidal & Berg, 2021, p.67). One of the reasons why women react negatively to stereotypical advertising images of women is that they believe that others are negatively affected by such images (Akestam, 2017, p. 71). Although women were initially considered to be the only target audience for advertisements featuring women, in recent years the viewpoints have shifted (Eisend, 2019) because the standards of equality between men and women have also changed (Eisend, 2010;Knoll, Eisend & Steinhagen, 2011). The results show that consumers of both genders respond equally to advertisements depicting male and female images, the effect achieved being statistically significantly higher for males (F = 3.40; p < 0.01; Means(male) =3.89) than for females (F = 3.40; p < 0.01; Means(female) = 2.99). In addition to confirming the trend of higher male sensitivity and lower female sensitivity to advertisements with binary images of both genders, the comparison of results in Table 3. This empirical evidence can be used by marketers and advertisers because advertisements depicting both genders are used to persuade mostly male target groups and less often for advertisements aimed at women (Table 4). In the present study, these indicators for the evaluation of advertisements are not taken into account, but the Student-Fischer t-test for the statistical significance of the differences between the means for the groups, formed by gender and between the variables comprising the stereotypic influence scales, show mixed results. Applying a stereotypical (with images of men and women) versus a non-stereotypical advertisement (with images of products without people present) affects affect statistically significantly more for advertisements with images of men and women (Mean 3.99) as well as with images of products for women (Mean 3.71), versus product images for men (Mean 3.40). When taking into account the simultaneous influence of the studied variables, a higher achieved influence was found for images of women in advertising, as well as products intended for women (Mean 3.71; t2,5 = 3.44; p < 0.00), and also images of women as well as images of men and women (Mean 3.16; t2,3 = 4.26; p < 0.02). The results support the claims of other researchers that men are presented in prominent visual and auditory roles, while women are portrayed in stereotypical ways (Paek, Nelson & Vilela, 2011, p.192), that is, higher achieved stereotypical influence is found in women through stereotypical advertisements, compared to men.
The shift in viewpoints was found to affect men (Eisend, 2019), because comparing the values showed that men reacted negatively to advertising depicting men and products for men (Mean 3.02; t1,4 = -4.52; p < 0.01). Therefore, advertisements aimed at men evoke more pronounced psychological reactance (Brehm, 1966;Thorbjørnsen & Dahlén, 2011) in consumers compared to advertisements aimed at women. Therefore, it can be assumed that the practical implementation of more images of men in advertisements aimed at men will be more persuasive in the single advertisement depicting men and women when the gender characteristics of the target groups are not taken into account. It is possible that this hypothesis can be tested in future studies.

Conclusions
The present study offers some empirical evidence for the mixed reactions typical of both genders of the audience that are affected by cross-gender influences. In addition, studies on the influence of stereotypes in advertisements on consumer beliefs are entirely dependent on other additional factors such as cross-gender influence, that is, the attitude of one gender towards a certain stereotype can largely determine the attitude of the other gender. Other studies prove this claim through positive effects of using nonstereotypical images of women and men in advertisements (Liljedal, Berg & Dahlen, 2020;Bian & Wang, 2015;Janssen & Paas, 2014;Antioco, Smeesters, Le Boedec, 2012).
For the formed sample, statistically significant gender differences, were found in relation to susceptibility to gender stereotypes in advertisements, with the highest achieved influence observed in men through the role behavior indicator such as caring for children, being a leader, and in women -trait descriptors such as self-affirmation, concern for others. Therefore, some new social effects are reported, due to the adoption of different roles as well as mixing of gender functions (Maher & Childs, 2003). Findings also demonstrate the high effectiveness of advertisements that focus on consumers of one gender, but not both (Johnson, 2014;Darley & Smith, 2013;Wolin, 2003). In this case, gender deviates from equality only for men who are depicted as part of the family, and there is no deviation for women because they are often depicted in a role of concern for others (Eisend, 2010;Knoll, Eisend & Steinhagen, 2011).
The achieved stereotypical influence depends to a large extent on the cross-gender effect, which is measured in the present work. Similar results reported for both genders, that is, the presence of images of the opposite gender does not cause reactance against a stereotypical advertising image in respondents (Brehm, 1966;Thorbjørnsen & Dahlén, 2011). The explanations for this result are complex: types of advertisements, the ways in which stereotypical images are presented, the predisposition of both genders to stereotypical images, and others. Empirical evidence supports the claim that gender stereotypes lead to positive selfperceptions and induce equal opportunities for both genders in society (Davis, 2002). In addition to this statement is the similarity in results, when reaching influence from advertisements intended for both genders.
The application of stereotypical (with images of men and women) versus non-stereotypical advertising (with images of products without the presence of people) achieves a significant effect for advertising aimed at both men and women, as well as advertising with images of products for women. Therefore, the author subscribes to the statement that the presence of stereotypical images of both genders does not cause strong reactance in the opposite gender, but rather stimulates some mixed reactions in consumers self-identifying to one of the genders. Stronger psychological reactance (Brehm, 1966;Thorbjørnsen & Dahlén, 2011) have been reported in male consumers under mixed influence achieved by advertisements depicting men and products for men. The thoroughness of the research requires conducting additional studies to deduce the cause-and-effect relationships for the achieved results. Future research, aimed at cross-gender influences in advertising, should take into account other demographic or socio-psychological factors causing mixed reactions among consumers of products and services in Bulgaria.