Objecting to Objectification: Women’s Collective Action against Sexual Objectification on Television
Media often portray women as mere sexual objects, but to date no known research has explored relations between exposure to such media content and willingness to engage in collective action. In the present study, Italian participants (78 men; 81 women) were exposed to a nature TV documentary (Control video), a television clip portraying women as sexual objects (SO video), or to the same sexually objectifying television clip including a commentary against such degrading depiction of women (Critique SO video). After exposure to the Critique SO video, women, but not men, reported greater collective action proclivity and behavioral intention to support a protest against female sexual objectification, as compared to the Control condition. Importantly, results further demonstrated that anger was the mechanism underlying women’s collective action proclivity, as well as intention to react. These findings suggest that media literacy messages in the form of critique videos may be valuable tools to promote more active and critical media consumption and that media specialists, concerned citizens, and social media activists may use such messages to motivate women to collectively take action against sexual objectification.
KeywordsMass mediaObjectificationCollective actionMedia literacy
In many western countries we are accustomed to be exposed to media images of undressed and sexy bodies often used as decorative objects or instruments to attract new consumers. Female bodies are the more common targets of such representation. Content analyses have shown that women are more likely than men are to be sexually objectified in advertisement, magazines, films, television (TV), and music videos (Aubrey and Frisby 2011; Conley and Ramsey 2011; Fouts and Burggraf 2000; Hatton and Trautner 2011; Smith et al. 2013; Vandenbosch et al. 2013). Although this trend is not novel, and had already been documented in Archer and colleagues’ early work (Archer et al. 1983), research has only recently begun to examine the consequences of such media portrayals of women.
Exposure to sexually objectifying media content may have serious negative effects. According to objectification theory (Fredrickson and Roberts 1997), living in a context in which bodies, especially women’s, are hyper-sexualized by media contributes to sexual objectification. When sexually objectified, a woman is treated as a mere sexual object deprived of individuality and personality, as if her body (or sexual body parts) could represent her entire person (Bartky 1990; Fredrickson and Roberts 1997). Sexually objectifying media may influence the way in which women are treated and perceived by others and by themselves. Recent studies in U.S. and European contexts, indeed, demonstrate that both men and women dehumanize sexually objectified female targets in magazines (Puvia and Vaes 2013; Vaes et al. 2011) and that men exposed to sexually objectifying images are more likely to harass women, to endorse traditional masculinity ideology, and to legitimize anti-equality attitudes and violence (Galdi et al. 2014; MacKay and Covell 1997; Malamuth and Check 1981; Milburn et al. 2000; Ward et al. 2006). Furthermore, American women exposed to sexually objectifying media are more likely to experience self-objectification, concerns over their body image, and eating disorders (Abramson and Valene 1991; Aubrey 2006, 2007; Hargreaves and Tiggemann 2004; Holmstrom 2004; see Grabe et al. 2008, for a review). Pioneering research has also shown that American women’s experience of sexual objectification may not only increase self-objectification, but also disrupt women’s social activism on gender equality (Calogero 2013).
Given the serious consequences of media sexual objectification, media literacy has been proposed as an intervention strategy to break the vicious cycle of sexual objectification (American Psychological Association [APA] 2010; Calogero and Tylka 2014; Tylka and Augustus-Horvath 2011; Zanardo 2011). The available evidence supports this claim, showing that, at least in western countries, media literacy messages might work as a buffer by reducing women’s internalization of beauty ideals and body concerns (Halliwell et al. 2011; Irving et al. 1998; Watson and Vaughn 2006). However, to the best of our knowledge, no research to date has addressed viewers’ collective reactions toward media objectification and media literacy messages.
The present study, therefore, had two main aims. Our first goal was to test whether mere exposure to sexually objectifying media would elicit viewers’ engagement in gender-related collective action, or whether, alternatively, an added critique against such degrading portrayals of women is necessary to stimulate media awareness and thus motivate people to participate in collective action. According to many scholars, media literacy messages should “enhance viewers’ criticism, by increasing their knowledge of the media, awareness of media influence, and ability to assess the realism of the media representation of reality,” thus ultimately reducing “the impact of media on audiences’ beliefs, attitudes, norms, and behaviors” (Jeong et al. 2012, p. 455). Our second goal of the present study was to test the role of emotional and cognitive reactions toward objectifying media as predictors of individuals’ projected engagement in collective action.
Sexually Objectifying Media and Collective Action
As highlighted in the 2010 report of the APA Task Force on the Sexualization of Girls in the United States, women are the common target of sexual objectification in visual media. Images of scantily dressed women taking sexy poses have increased over the years in magazines (Hatton and Trautner 2011), music videos (Aubrey and Frisby 2011), and top-grossing films (Smith et al. 2013). This trend is growing even faster for teenagers, with over one-half of female adolescents represented in visual media in a sexually objectifying manner (Smith et al. 2013).
Italian television fits well with the trend just described. For example, within the European project “Women and Media in Europe,” the Italian Center of Social Studies and Investments (Centro Studi Investimenti Sociali [CENSIS] 2006) analyzed the content of 598 television programs from the seven most popular Italian broadcast networks, finding that women were mostly depicted as “showgirls,” such as actresses (56.3%), singers (25%), and models (20%). According to CENSIS, women were more likely to be associated with fashion, entertainment (31.5%), or physical violence (14.2%), but they were rarely represented in the context of politics (4.8%), business (2%), or culture (6.6%). Further, in Italian TV programs the host is often a man (63%), whereas women are typically relegated to decorative roles, often scantily dressed (36.9%), with cameras frequently focusing on their bodies in a voyeuristic way (30%), instead of highlighting their artistic abilities (15.7%). Overall, Italian TV tends to show women in marginal roles and as sexual decoration.
Concerned with the degrading portrayal of women in Italian TV, a well-known diversity management expert, Lorella Zanardo, produced a powerful documentary (Zanardo et al. 2009) including a stream of clips from popular TV programs, accompanied by a personal commentary. The commentary invites viewers to critically question the sexualized portrayal of women and to become aware of the technical choices (e.g., positioning of video camera) involved in women’s objectification. Hence, the documentary may be classified as a media education tool, intended to increase viewer’s media literacy. Although in recent years gender equality campaigns have grown globally (e.g., “If not now, when?”; “HeForShe” United Nations” 2014, campaign), to this date little is known about the efficacy of such campaigns on women’s and men’s willingness to participate in collective action and gender activism. Therefore, in the present study we tested viewers’ collective action proclivity in response to sexually objectifying media and in response to a criticism of the same media content.
Collective action can be defined as actions (petitions, public protests, boycotts, etc.) aimed at improving the condition of a disadvantaged social group (van Zomeren and Iyer 2009). Different social sciences have investigated collective action because it is considered as one of the most effective ways for disadvantaged group members to gain social equality and achieve social change and justice (van Zomeren and Iyer 2009; Wright and Baray 2012; Wright and Lubensky 2009). According to the Dynamic Dual Pathways model (van Zomeren et al. 2012), two different coping strategies (emotion- vs. problem-focused) may be undertaken by group members facing social disadvantage.
In the emotion-focused approach, collective action is promoted by anger arising from the perception of unfairness and external blame for the disadvantaged situation (e.g., government, high status group). In the problem-focused approach, the motivation to participate in collective action derives from perceived group efficacy, which is closely related to the perception that other group members are willing to engage in collective action as well (action support). The model posits that contexts suggesting greater group efficacy and action support would encourage a problem-focused approach, whereas contexts eliciting stronger group-based anger (e.g., external blame of unfairness of the situation) would encourage an emotion-focused approach (van Zomeren et al. 2012). Therefore, as a second goal of the present study, we investigated whether the exposure to sexually objectifying media and to a critique of such media content would predict an emotion-based or problem-focused collective action pathway.
To date, research has mainly addressed low-status group members’ willingness to join collective action. Nonetheless, some studies have also investigated high status groups (Iyer and Ryan 2009; Mallett et al. 2008; Postmes and Smith 2009), showing that perspective-taking, group-based guilt and, again, group-based anger may be crucial predictors of collective action by advantaged group members towards improving the disadvantaged group’s condition (Iyer and Ryan 2009; Leach et al. 2006; Mallett et al. 2008). Because an additional interest of the present study was to explore the reactions of men (out-group members) toward female sexual objectification in the media, we also investigated the role of perspective-taking and guilt in predicting individuals’ collective action responses.
Gender and Collective Action
In the western world, women are recognized as a socially disadvantaged group because of their lower status, power, and opportunities, which contribute to overall gender inequality (Barreto et al. 2009). It is therefore important to investigate the factors that could prevent or motivate individuals to take collective action aimed at improving women’s social condition. In general, compared to women, men are less prone to support feminist goals (Williams and Wittig 1997). Other research, however, has highlighted a growing involvement of men in activism toward gender equality, especially antiviolence activism (e.g., White Ribbon Campaign; Flood 2001, 2005). Closer to the present study, Bongiorno et al. (2013) found that using sexually objectified female targets to advertise PETA (i.e., People for the Ethical Treatment of Animal Organization) may reduce men’s intention to support the organization, as compared to non-objectifying advertisement.
More is known about women’s collective action targeting gender disparities. Unsurprisingly, women with a feminist identity are more likely to engage in collective action (Liss et al. 2004; Nelson et al. 2008), whereas self-objectification seems to be a barrier to collective action. When in a state of self-objectification (i.e., activated by asking participants to remember an experience in which they felt treated as sexual objects), women were more likely to support the gender status quo and were less willing to engage in social activism (Calogero 2013). Similarly, benevolent sexism undermines women’s engagement in collective action, whereas exposure to blatant hostile sexism increases it because women become less inclined to justify the gender system (Becker and Wright 2011). In addition, Ellemers and Barreto (2009) found that women perceive old-fashioned, but not modern, sexism as a form of inequality. This, in turn, elicits anger, support for collective action, intention to protest, and collective protest behavior. However, no known study to date has explored whether women’s proclivity to react would occur in the face of mere exposure to sexually objectified portrayals of women in the media, or whether, in addition, a critical point of view is necessary to trigger, both for women and men, collective action aimed at stopping the widespread sexual objectification of girls and women in the media.
The Present Study
The present study was aimed at investigating the effects of exposure to sexually objectifying media, with or without a reasoned critique of such media content, on viewers’ (a) collective action proclivity and (b) behavioral intentions to participate in a public rally against such degrading representations of women. We also explored whether media content would affect perspective-taking, guilt, anger, action support, and group efficacy, which have been proposed as antecedents of collective action (e.g. van Zomeren et al. 2004; Mallett et al. 2008). To meet these goals, participants were randomly exposed to watch one of three videos: (a) a video clip of sexually objectified TV programs (Sexual Objectification [SO] video condition) in which women are presented as sexual objects, (b) the same clip of sexually objectifying TV including background comments against the degrading portrayal of women on TV (taken from the original documentary “Women’s body”, Zanardo et al. 2009; Critique Sexual Objectification [Critique SO] video condition), or (c) a nature TV documentary (Control video condition). We hypothesized that, after exposure to the Critique SO video (vs. SO or Control video), participants, especially women, would express stronger willingness to engage in collective action and greater behavioral intentions to support the cause.
We also tested whether the Critique SO video would increase male and female viewers’ comprehension of women’s situation (perspective-taking), responsibility for the situation in which women are relegated in the media (guilt), anger, perception that other in-group members would fight for the same cause (action support), and appraisal of in-group members’ efficacy to achieve social change (group efficacy). In line with proposals on media literacy interventions to raise a critical view on sexualized media (e.g., APA 2010; Calogero and Tylka 2014; Tylka and Augustus-Horvath 2011), we expected that a reasoned and assertive point of view on such media content would trigger people’s (both women’s and men’s) reactions. Furthermore, drawing from collective action models (Mallett et al. 2008; van Zomeren et al. 2012), we investigated whether and which of the collective action antecedents would mediate the relation between experimental video condition and participants’ collective action proclivity and behavioral intentions to support the cause.
Finally, we speculated about the role that social dominance orientation might play in affecting viewers’ reactions. Social dominance orientation (SDO) is the tendency to believe that some groups are inherently superior or inferior to others and to approve such inequality between social groups (Pratto et al. 1994). Given that SDO is linked negatively with support of women’s rights and positively with a view of women as sexual objects (Pratto et al. 1994; Pratto et al. 2000; Pratto et al. 1997), we investigated its relationship with collective action antecedents, participants’ proclivity to engage in collective action, and their projected behavioral reactions against women’s sexual objectification in the media. In line with previous literature, we expected higher levels of SDO to correlate negatively with all collective action antecedents, as well as collective action proclivity and intended behavioral support for the cause. In other words, we hypothesized that the more participants would approve inequality among groups, the less they would experience anger, guilt, action support, group efficacy, and perspective-taking toward women’s condition, and the less they would plan to engage in collective action against degrading TV representations of women.
One of two female experimenters recruited 159 residents of Northern Italy (78 men; 81 women) either at different University libraries and study rooms or among neighbors and acquaintances (see Galdi et al. 2014, for a similar procedure). The experiment was run using a laptop computer in quiet rooms at the university, libraries or at participants’ home. The sample (Mage = 32.50 years, SD = 12.33 years) comprised 43 (27%) university students, 47 (30%) blue-collar workers, 44 (28%) white-collar workers, and 25 remaining participants (15%) including housewives, unemployed, and professionals. Gender distribution was similar across the three main categories (university students, blue-collar and white-collar workers), χ2(2,Ν = 134) = 1.40, p = .50. Moreover, female and male participants were similar in age, t(157) < .11, p > .90. All participants took part in the study voluntarily without monetary compensation. The procedure of the experiment and the main dependent variables were administered in the same order as presented in the following.
Procedure and Materials
Participants were informed that the study was aimed at investigating mass media communication and that the main task would be to evaluate a brief video clip. Participants were first invited to complete a paper-and-pencil questionnaire including demographic information, television viewing habits (i.e., Exposure to Sexist and Non-sexist TV programs), and a scale allegedly measuring personal characteristics (which was in reality the SDO scale). Then, participants were randomly assigned to one of the three video conditions: Critique Sexual Objectification (27 men, 26 women), Sexual Objectification (26 men, 27 women), Control (25 men, 28 women). To support the cover story, after exposure to the clip, participants evaluated the video and rated their current mood.
At the end of the task, the experimenter asked participants to participate in an allegedly unrelated experiment on attitudes and the effectiveness of communication via internet. All participants agreed. Therefore, they were invited to complete a questionnaire on social perception (i.e., Collective Action Antecedents scale and Collective Action Proclivity scale). Immediately afterwards, participants were given a leaflet proposing an online petition promoted by a (fictitious) non-profit association, allegedly fighting against the objectification of women in society. After reading the petition, participants were instructed to indicate whether they would support the cause of the association. At the end of the experiment, participants were fully debriefed and thanked for their participation.
Exposure to Sexist and Non-sexist TV Programs
To assess participants’ habitual exposure to televised sexist and non-sexist programs, we used a list of 12 popular Italian TV programs, six pre-tested as being sexist and six pre-tested as neutral (See online supplement for pretest information). Participants were asked to report how often they watched each program on 4-point scales from 1 (never/I don’t know the program) to 4 (always).
Social Dominance Orientation
The SDO scale, originally developed by Pratto et al. (1994), is the most common measure used to assess individuals’ belief that some people or groups are inherently superior to others, as well as the degree of approval of unequal group relationships. Participants filled out an Italian adaptation of the scale (Aiello et al. 2005) composed of nine items related to the approval of inequality (e.g., “Some groups are simply more worthy than others”) and nine items related to the approval of equality among social groups (e.g., “It would be nice if there was equality among all social groups”). None of the items referred directly to gender. Responses were provided on 5-point scales ranging from 1 (not at all) to 5 (very much). An averaged index of SDO was calculated after reverse-coding the nine items indicating approval of equality (Cronbach’s α = .89). Higher values reflect higher social dominance-oriented beliefs.
Three video clips (Sexual Objectification, Critique Sexual Objectification, Control) were employed. For the Critique Sexual Objectification (Critique SO) and the Sexual Objectification (SO) video condition, a brief extract of the Italian video-documentary “Women’s body” (Zanardo et al. 2009; also available online with English subtitles) was used. The only difference between the Critique SO and the SO conditions was that the SO video did not include the commentary by the author, which was replaced with pop music. Both videos include scenes from popular Italian TV programs showing provocatively dressed and posed women or scantily clad female assistants, who allegedly help male presenters conduct the show.
Can a woman crawl under a Plexiglas table, pretending she is the leg of the table, spend a long time under there, pretending that it’s only a silly game? Can this be done without leaving a scar somewhere in her body? And what should people who are watching the program feel about it? On TV there is a woman, and a man is using her as the leg of a table. (Zanardo et al. 2009)
Commenting on the woman hanging as a prosciutto ham, Zanardo said: “Why don’t we do something about it? Why can’t we show our own truth? Why do we keep accepting this constant humiliation? Why don’t we fight to protect our rights? What are we afraid of?” (See online supplement for complete transcript.)
To summarize, the visual component was identical in the Critique SO and SO videos, the only difference was the audio: the Critique SO video included the original critique comments by Lorella Zanardo, whereas in the SO video Zanardo’s comments were replaced with pop music. Finally, the Control video condition included scenes of a nature TV documentary accompanied by soft music. All videos were presented using a laptop and were approximately 3-min long.
Evaluation of the Video and Mood
To support the cover story, after watching the video clip, participants rated how interesting, pleasant, and well edited they considered the video on scales ranging from 1 (not at all) to 7 (very much). A score of Video Evaluation was calculated by averaging the responses on the 3 items (Cronbach’s α = .71). Participants also reported their mood on a continuum line ranging from 0 (very good) to 14 cm (bad).
Collective Action Antecedents
Participants filled out a scale that comprised five well-known collective action antecedents: perspective-taking, guilt, anger, action support, and group efficacy (see Mallett et al. 2008; van Zomeren et al. 2004). Specifically, an 11-item scale was used in which some items were adapted for male participants (see brackets). The scale assessed: Perspective-taking (2 items; “I can understand Italian women’s feelings for their condition of discrimination” and “I can understand the feeling of frustration and humiliation of Italian women for their social status”; female participants: r = .85, p < .001; male participants: r = .42, p < .001), Guilt (2 items; “Women [Men] are partially responsible for the discriminatory condition that they [women] live in our society” and “Women [Men] should feel guilty about the sexist attitudes against women”; female participants: r = .69, p < .001; male participants: r = .67, p < .001), Anger (3 items; “I feel angry for how women are regarded in Italy”, “The portrayal of women in Italian television makes me angry” and “I am embittered for women’s condition in the Italian society”; female participants: α = .87; male participants: α = .90), Action Support (2 items; “I think that most women [men] would be inclined to act in order to change the general social condition of their group [women]” and “I think that among women [men] there is a widespread discontent for the discrimination of their group [women]”; female participants: r = .65, p < .001; male participants: r = .70, p < .001), perception of group’s efficacy to achieve social change (Group Efficacy; 2 items; “I think that women [men] together can change the general social condition of their group [women]” and “I think that women [men] can counteract discrimination against their group [women]”; female participants: r = .84, p < .001; male participants: r = .79, p < .001). Participants were instructed to reflect on the present condition of women in Italy and to indicate how much they agreed with each item on scales ranging from 1 (not at all) to 7 (very much). Items were averaged for each of the five measures so that higher scores indicated greater perspective-taking, guilt, anger, action support, and group efficacy.
Collective Action Proclivity
Following van Zomeren et al. (2004), we assessed participants’ collective action proclivity using three items: “I would participate in a demonstration against the actual condition of women in Italy”, “I would do something together with other women [men] to protest against the condition in which we are relegated”, and “I would participate in collective action to stop discrimination of Italian women”. Participants responded on 7-point Likert scales ranging from 1 (not at all) to 7 (very much). Given the good reliability of the scale (female participants: α = .95; male participants: α = .97), an average index of Collective Action Proclivity was calculated.
To obtain a behavioral intentions’ measure of the effects of the three video clips, participants were presented with a leaflet showing an on-line petition promoted by a (fictitious) non-profit association (“Not Just Dolls”), allegedly fighting “against the widespread objectification of women in society.” After giving a short description of the main purpose of the association and providing website information, the petition concluded: “We are tired of viewing soubrettes and girls treated like showpieces on TV. We say ENOUGH to this use of women. Not all of us are like that, we are not dolls! Give us our dignity back!” After reading the petition, participants were asked to respond “yes” (coded 1) or “no” (coded 0) to three questions: (a) “I am going to sign the web petition promoted by the association,” (b) “I will participate in the rally scheduled for next week,” and (c) “I will become a member of the association”. The sum score of Behavioral Intentions thus ranged from 0 (support for none of three parts of petition) to 3 (support to all three parts of petition).
Descriptive statistics and correlations among study variables by participants’ gender
2. Video Evaluation
8. Action Support
9. Group Efficacy
10. CA Proclivity
11. Behavioral Intentions
Despite a strict randomization procedure, participants’ indices of SDO varied across conditions, F(2,153) = 3.25, p = .04, ηp2 = .04 (M = 2.20, SD = .76, for Critique SO condition; M = 2.54, SD = .71, for SO condition; M = 2.37, SD = .62, for Control condition). However, the interaction between Gender and Condition on SDO was not significant (F < .40, p > .50). Given these results, a multiple moderation model (using PROCESS; Hayes 2013) was conducted for each of the main dependent variables (i.e., Perspective-taking, Guilt, Anger, Action Support, Group Efficacy, Collective Action Proclivity, and Behavioral Intentions). Condition, using dummy coding (Dummy 1: Critique SO = +1, SO = 0, Control = 0; Dummy 2: SO = +1, Critique SO = 0, Control = 0), was included as predictor, whereas Gender (female = 1, male = 0) and SDO (centered) were entered as moderators. The inclusion of the three-way interaction among SDO, Condition (Dummy 1 and Dummy 2), and Gender did not lead to a significant improvement in the explained variance (max ΔR2 = .02, p > .08), thus disconfirming a potential moderating role of SDO. Nonetheless, SDO was included as covariate in all the subsequent analyses.
Evaluation of the Video and Mood
A two-way ANCOVA was conducted on participants’ scores of Video Evaluation, with Gender (male, female) and Condition (Critique SO, SO, Control) as the between-participants variables and with SDO as a covariate. Results showed a significant effect of Condition, F(2,152) = 6.16, p = .003, ηp2 = .08: Post hoc tests with a Bonferroni correction showed that participants liked the Control (M = 4.07, SD = 1.23) more than the SO video (M = 3.14, SD = 1.40; p = .001), whereas no difference emerged between the Control and the Critique SO (M = 3.71, SD = 1.36) or between the Critique SO and SO conditions.
A Gender x Condition interaction also emerged, F(2,152) = 7.04, p = .001, ηp2 = .08. Simple effect analysis revealed a significant effect of Condition for women, F(2,77) = 13.84, p < .001, ηp2 = .26, but not for men (p = .80): Women liked the SO (M = 2.63, SD = 1.33) less than the Critique SO (M = 3.95, SD = 1.14; p < .001) and the Control (M = 4.39, SD = 1.30; p < .001) videos, whereas no difference emerged between the Critique SO and the Control clips. Moreover, simple effect analyses on the effect of Gender within Condition showed that men liked the SO video more than women did, F(1,152) = 8.56, p = .004. Importantly, no gender differences were found in the Control and Critique SO conditions. Given that the evaluation of the video was included to support the cover story, and it is not relevant to the purpose of the present study, we will not discuss it further.
A two-way ANCOVA was then conducted on participants’ scores of Mood, with Gender and Condition as the between-participants variables and with SDO as a covariate. As shown in Table 1, women reported higher levels of negative mood as compared to men, F(1,132) = 16.21, p = .001, ηp2 = .11. A significant effect of condition was also found, F(2,132) = 11.84, p = .001, ηp2 = .15: Compared to the Critique SO (M = 8.37, SD = 3.20) and the SO (M = 7.18, SD = 4.14) conditions, participants felt better after exposure to the Control clip (M = 4.92, SD = 3.66; ps < .008). Importantly, no interaction effect emerged between Gender and Condition, thus indicating that the reported results were not affected by participants’ mood as a function of their gender and experimental condition. Therefore, we will not discuss Mood further.
Collective Action Antecedents
A MANCOVA was conducted on participants’ scores of Perspective-taking, Guilt, Anger, Group Efficacy, and Action Support, using Gender and Condition as independent variables and with SDO as a centered covariate. SDO was a significant covariate, Pillai’s Trace = .42, F(5,148) = 21.82, p < .001, ηp2 = .42. A significant effect of Gender was also found, Pillai’s Trace = .19, F(5,148) = 6.89, p < .001, ηp2 = .20. Univariate analyses revealed that women reported higher levels of Guilt, F(1,152) = 9.69, p = .002, ηp2 = .06, and Anger, F(1,152) = 24.19, p < .001, ηp2 = .14, as compared to men (see Table 1). No gender differences were found on participants’ scores of Perspective-taking, Group Efficacy, and Action Support (Fs < .40, ps > .50).
Descriptive statistics for dependent measures as a function of video condition and participants’ gender
Critique SO video
Compared to women, men showed lower scores of Perspective-taking in the SO and Control conditions (ps < .02), whereas no gender differences were found on the Critique SO condition. Women and men also reacted differently depending on the experimental condition: For female participants, Perspective-taking was higher after exposure to the Critique SO than after exposure to the Control video (p < .001), with the SO video occupying an intermediate position that did not reliably differ from the other two conditions. Conversely, men showed a decline in Perspective-taking in the SO condition as compared to the Control condition (p = .01), with the Critique SO video occupying an intermediate position.
Compared to men, women reported greater levels of Guilt after exposure to the Critique SO (p < .001), whereas no gender differences emerged in the Control and SO conditions. Moreover, as shown in Table 2, men’s scores of Guilt were not affected by video condition. Conversely, for female participants, the Critique SO video elicited greater Guilt, as compared to both the SO and the Control video condition (ps = .003).
Compared to men, women showed higher scores of Anger after exposure to the Critique SO and the SO videos (ps < .001), whereas no gender differences emerged in the Control video condition. Men’s scores of Anger were not affected by video condition. Conversely, women reacted differently depending on the experimental condition: After exposure to the Critique SO and the SO video clip, female participants reported greater Anger than after exposure to the Control video (ps < .008).
Collective Action Proclivity
An ANCOVA was performed on participants’ indices of Collective Action Proclivity, using Gender and Condition as the between-participants factors and with SDO as a covariate. SDO was a significant covariate, F(1,152) = 90.87, p < .001, ηp2 = .37. A main effect of Gender also emerged, F(1,152) = 19.62, p < .001, ηp2 = .11: Women showed greater Collective Action Proclivity than did men (see Table 1). Condition was also a significant factor, F(2,152) = 3.19, p = .04, ηp2 = .04. As reported in Table 2, participants showed greater proclivity to be involved in collective action after exposure to the Critique SO as compared to the Control video (p = .02), with the SO video occupying an intermediate position that did not reliably differ from the other two conditions.
Importantly, the condition main effect was qualified by a significant Condition x Gender interaction, F(2,152) = 5.20, p = .007, ηp2 = .06: Compared to women, men showed lower Collective Action Proclivity after exposure to the Critique SO and the SO clip (ps < .001), whereas no gender differences were found in the Control video condition. Simple effect analysis revealed also a significant effect of Condition for women, F(2,152) = 8.39, p < .001, ηp2 = .10, but not for men (p > .79). As shown in Table 2, for female participants, Collective Action Proclivity was higher after exposure to the Critique SO than the Control video clip (p < .001), with the SO video occupying an intermediate position that did not reliably differ from the other two conditions.
Similar to Collective Action Proclivity, we conducted an ANCOVA, with Gender and Condition as the between-participants variables and with SDO as a covariate. Again, SDO was a significant covariate F(1,152) = 18.63, p < .001, ηp2 = .11. No main effect emerged. Although the interaction between Gender and Condition did not reach conventional statistical significance, F(1,152) = 2.52, p = .08, ηp2 = .03, it is worthwhile to notice that after exposure to the Critique SO women showed a tendency to take action against media objectification of women to a greater extent, as compared to the Control condition (p = .05). Moreover, a difference between men and women emerged in the SO video condition (p = .04), with men expressing lower behavioral intentions.
Moderated Mediation Analyses
Two moderated mediation models were conducted to test whether collective action antecedents (i.e., Perspective-taking, Guilt, Anger, Action Support, and Group Efficacy) mediated the relation between Condition and either Collective Action Proclivity or Behavioral Intentions, also considering participants’ Gender as a moderator. Using PROCESS computational tool for conditional process analysis (Hayes 2013), Collective Action Proclivity was used as the criterion variable in the first model. Given that video Condition was a categorical variable with three levels, we created two dummy-coded variables. Specifically, Dummy 1 tested for the effect of the Critique SO (coded +1) versus SO condition (coded 0) and Control condition (coded 0). Dummy 2 tested for the effect of the SO condition (coded +1) versus Control (coded 0) and Critique SO condition (coded 0). Dummy 1 and Dummy 2 were simultaneously entered in the moderated mediation model as predictors (see Hayes 2013 for testing multiple IVs in PROCESS), with Gender (male = 0, female = 1) as moderator and SDO (centered) as controlling covariate. Perspective-taking, Guilt, Anger, Action Support, and Group Efficacy were modeled as centered parallel mediators. For all models Variance Inflation Factors showed no multicollinearity problems (max VIF = 3.30).
In the present study we investigated the effects of exposure to objectifying TV images, either by itself or combined with a critique aimed at raising people’s awareness, on individuals’ collective action proclivity and behavioral intentions to take action against the degrading depiction of women in the media. Several important results emerged. First, in line with our predictions, after exposure to the critique video (vs. control video), female participants were more prone to recognize the disadvantaged position of women in Italian society (perspective-taking), and they felt angrier and guiltier for the way in which Italian media and society treat women. After exposure to the critique video clip, women were also more willing to show a proclivity toward collective action, as compared to women in the control condition. Importantly, the same pattern of results emerged at the level of behavioral intentions. Women were more likely to support the cause against the widespread objectification of women in the media (e.g., project signing a petition and participating in a rally) after exposure to the critique (vs. control) video, whereas the experimental video condition did not affect men, who, compared to women, showed lower action tendencies after exposure to the blatant sexually objectifying video.
An even more important result of the present study comes from the moderated mediation analyses, which shed light on the mechanism underlying the effects of video exposure. For women, but not men, anger was found to be the unique mediator of the positive effect of the critique video (vs. sexually objectified and control videos) on collective action proclivity, as well as on behavioral intentions to react. Overall, these findings are consistent with previous research (Ellemers and Barreto 2009; Iyer and Ryan 2009; van Zomeren et al. 2004) and with the dynamic dual pathway model (van Zomeren et al. 2012), both of which indicate anger as one crucial path through which collective action responses of disadvantaged groups unfold. Consistently, in the present study women, who are the target of sexual objectification in the media, showed greater anger for their disadvantaged situation after exposure to critique messages compared to exposure to either the un-critiqued sexualized images or the control video. Female anger, in turn, triggered higher collective action proclivity and behavioral intentions to support the cause against media sexual objectification.
Two further important findings emerged from the present study. First, we investigated the role played by social dominance orientation in affecting participants’ collective action proclivity. SDO was negatively related to both men’s and women’s level of collective action predictors, collective action proclivity, and to participants’ behavioral intention to support the cause against female sexual objectification in the media. Thus, in line with previous studies (Pratto et al. 2000; Pratto et al. 1994), the present findings suggest that the endorsement of inequality beliefs may reduce people’s willingness to engage in collective action.
Second, habitual exposure to sexually objectifying TV was generally associated with lower levels of collective action proclivity, as well as with women’s lower behavioral intentions to support the cause. These results extend previous research showing that frequent exposure to sexualized media increases endorsement of stereotypical gender roles and the view of women as sexual objects (Peter and Valkenburg 2007; Ward 2002; Ward and Friedman 2006). In sum, the overall pattern of results suggests that the chronic exposure to objectifying media might lead to the dangerous assumption that such female portrayal is the norm, thus further reducing people’s likelihood to react.
Going back to objectification theory (Fredrickson and Roberts 1997), constant exposure to sexualized images by media in everyday life may have important negative repercussions for women’s psychological and cognitive well-being (Moradi and Huang 2008, for a review). Recent research has proposed media literacy as a crucial intervention strategy that might help temper the negative outcomes of sexual objectification (APA 2010; Calogero and Tylka 2014; Tylka and Augustus-Horvath 2011). Furthermore, some research has shown that media literacy intervention aimed at challenging the unrealistic beauty ideals proposed by media may reduce not only the internalization of such ideals, but also women’s body dissatisfaction (Halliwell et al. 2011; Irving et al. 1998; Watson and Vaughn 2006). However, to date little is known about the potential efficacy of such strategies on people’s willingness to take action. Therefore, the positive effects produced by the critique video used in the present study are encouraging. Our results not only respond to the recent APA report (2010) pointing to the exponential increase of sexualized images proposed by media, but also have important implications for implementing and testing the efficacy of intervention programs. We demonstrate that sensitizing campaigns, such as Zanardo’s (Zanardo et al. 2009), could represent, at least for women, a powerful tool to raise awareness and to motivate individuals to engage in collective action aimed at improving media portrayals of women.
Limitations and Future Research Directions
The reaction of men in the present study is complex and deserves closer analysis. First, when simply exposed to sexually objectifying TV without any reasoned critique, men expressed less support for women’s cause than women did. This result is in line with previous research showing that exposure to sexually objectifying media increases men’s endorsement of masculine gender role norms and proclivity to sexual harassment (Galdi et al. 2014). The present results are also in line with findings by Vaes et al. (2011) who showed that men tend to dehumanize sexually objectified women when sexually attracted by them. Therefore, not surprisingly, in the current study, after exposure to sexually objectified depiction of women, men showed lower intention to take part in collective action fighting for gender equality. At the same time, men did not manifest any willingness to participate in collective action even after the exposure to a reasoned critique of such sexualized TV portrayals. The present results indicate that exposure to comments against the degrading TV portrayals of women may be effective to motivate women, but not men, to take action.
Clearly, further research is needed to investigate potential factors that may increase men’s engagement in social activism to improve women’s condition. As a case in point, in the present study the background voice of the critique video was female and addressed specifically media sexual objectification of women. This may have suggested to male participants that sexist and objectifying media are mostly a female problem and that men cannot do much to improve the situation. Future studies should test whether a male critique voice, possibly combined with evidence that men may also suffer from media objectification, would enhance men’s involvement in the issue of sexual objectification, thus making it an across-gender cause.
Addressing men explicitly, as Emma Watson did in her famous UN speech in September 2014 as part of the “HeForShe” campaign (United Nations 2014), may be an effective strategy to raise men’s awareness and willingness to take action. These suggestions would also be in line with Calogero and Tylka (2014, p. 765), who have recently argued that sexual objectification could be more effectively harnessed if its rejection was framed “as endorsed by and for the betterment of the broader society.” Showing that media sexual objectification is harmful for the entire society and portraying men as part of the solution, rather than the problem, might therefore provide a promising approach to involve men in collective action.
An additional limit of the present study concerns potential demand characteristics, given that we did not explicitly test for suspicion. Thus, we cannot exclude that expectancy effects may partially be responsible for gender differences, with men displaying more reactance to influence attempts of the critique video (Klein et al. 2012). Moreover, we did not explore which specific aspects of the critique participants remembered better, which would allow us to identify the most effective characteristics of the critique message. Future research is needed to address both of these issues.
Although the moderated mediation model clearly supports the mediating effect of anger in the relation between the critique (vs. sexually objectified and control) video exposure and women’s behavioral intentions, the non-significant effect of condition in the univariate analysis merits further consideration. We argue that because the behavioral measure adopted in the present study is in itself a critique against sexual objectification, it may have reduced the difference between conditions, thus undermining the effect of the critique condition. In addition, except for guilt, univariate analyses supported a positive effect of the critique against the control (but not against the pure sexually objectifying) video, thus suggesting a more general reactiveness of women against sexually objectifying contents regardless of the presence or absence of a critical point of view. However, the moderated mediation models dispel this alternative interpretation, showing that it is the unique contribution of the critique video that is responsible for increasing women’s anger and consequently their collective action responses.
Another aspect that deserves attention in future studies is the potential role of guilt. Women (but not men) felt more responsible for the situation after having been exposed to the critique video than in the other conditions. This might have occurred because only the critique condition specifically proposed a reflection about female representation in the media, which might have elicited an assumption of responsibility by women. Nonetheless, guilt played no mediating role in the present study. Although guilt does not seem to be a driving force in collective action, it may well drive other reactions to objectifying media, a question to be investigated in future research. Future studies might also include shame in the model because, contrary to guilt, it has been shown to predict avoidance tendencies especially when the discrimination is perceived as other-caused (Schmader and Lickel 2006).
A final potential limitation of the present study is its ecological validity: The main findings of the study are based on a 3-min video, representing an artificial concentration of real Italian TV programs. However, because TV in Italy (and possibly in other western and non-western countries) portrays highly sexist and objectifying scenarios very frequently, our results may well be an underestimation of the real effects of daily exposure to such degrading TV depictions in western countries. Nevertheless, we deem it necessary for future research to extend the present findings to different cultural contexts and to also test other media literacy manipulations and stimuli.
The present findings are especially important for their practical implications. Because the present study is, to our knowledge, the first that tested the efficacy of a critique video on women’s collective reaction, additional research is needed before general conclusions can be drawn. However, our findings are encouraging because they suggest that media literacy messages in the form of critique video clips may be valuable tools to promote more active and critical media consumption, as well as to encourage proactive reactions (Tylka and Augustus-Horvath 2011). We do not claim that this approach is the only, or even the most efficient, intervention to mobilize media consumers. Nonetheless, the current approach provides a possible tool that media specialists, concerned citizens, and social media activists may use to motivate women to take action in favor of a more balanced portrayal of men and women in the media.
The present study provides novel evidence that exposure to media literacy messages, such as a critique aimed at sensitizing people and raising their awareness of sexually objectifying practices in the media, increases women’s proclivity to take action and willingness to stand against such objectifying and degrading portrayals. Importantly, the present study indicates that anger is the mechanism underlying women’s proactive responses. Hopefully our findings will stimulate further research to test the efficacy of campaigns and interventions to promote a critical approach toward the media, such as Zanardo’s (2011) project, “A new look at the media,” which aimed at training adolescents and educators to approach the media with a critical eye.
This work was supported by PRIN grant number 20123X2PXT_003 (2012) from the Italian Ministry of Education, University and Research. We thank Lorella Zanardo for her support on this project.