Introduction

In the last few centuries, women in Western countries have achieved revolutionary advancements in terms of civil and social rights. Nevertheless, women continue to be underrepresented in leadership positions, and this major issue needs to be tackled. This “unfinished revolution” is very apparent in the field of politics. In Italy, the country where the present research was conducted, there have been some significant steps forwards: in 1948, the first elected parliament presented only 5% of female deputies, whereas the election of March 2018 saw a threshold of 35% of female deputies being achieved. However, despite this noteworthy improvement, women in politics still tend to be found in marginal or low-prestige positions: throughout the first 70 years of the Italian Republic, women have held only 5% (78 out of over 1500) governmental offices, and no woman has ever been nominated as prime minister (Andreuccioli et al. 2018). Barriers to gender equality in political representation are present in numerous countries throughout the world (GAP 2017), and several explanations—from structural to individual—have been advanced to comprehend and overcome them. As regards structural issues, the lack of robust gender-friendly policies to bridge the deep-rooted socioeconomic disadvantages between women and men (Coffé 2013; Frazer and MacDonald 2003) as well as the traditional division of labour—wherein women are more likely than men to assume domestic and familial responsibilities—are still worrisome issues in Western countries (Eagly and Carli 2007; Campbell and Winters 2008). As regards the individual obstacles, research has shown that women present lower levels of political interest than men (Lawless 2012; Pacilli et al. 2012) even after controlling for education as well as material resources (Fraile and Gómez 2017). Last but not least, as Dolan (2003) argued, electing women to office takes not only women who will stand as candidates but also voters who will perceive them positively and then vote for them. As regards perceptions of female and male politicians, media framing—that is, the way objects of interest are reported and emphasised by news media (Weaver et al. 2004)—constitutes a relevant factor. Research has shown that along with higher coverage of men than women (Sensales et al. 2018), there exists widespread sexist and stereotypical communication about politicians (Carlin and Winfrey 2009; Heldman et al. 2005; Sensales and Areni 2017). For instance, Gibson and Heyse (2010), examining the 2008 Republican National Convention, showed that speakers rhetorically celebrated the hegemonic masculinity of John McCain as well-suited for the U.S. presidency while denigrating feminine traits. Although the crucial role of gender stereotypes in social perception has been consistently shown (Ellemers 2018; Heilman 2001; Rudman and Glick 2001), research examining the impact of gender stereotypes and ideology regarding gender relations on voting intention has provided inconsistent evidence. While gender stereotypes were not significantly related to voting for women Democratic candidates for the House of Representatives in 2010 (Dolan and Lynch, 2014), as regards the United States 2016 presidential elections, greater sexism predicted more positive attitudes towards Trump and less favourable attitudes towards Clinton, even after controlling for political ideology (Ratliff et al. 2017). Thus, whether and how gender stereotypes can promote or hinder voting intention towards a woman is still an issue to be examined.

Gender Stereotypes and Politics

Stereotypes are general expectations about the characteristics and behaviours of members of social groups (Ellemers 2018). According to the stereotype content model (Fiske et al. 2002), two fundamental dimensions are active in human perception: warmth and competence. While high-status groups are generally perceived as high in competence and low in warmth, low-status groups are considered as low in competence and high in warmth. This pattern also applies to gender: men are usually stereotyped as competent, assertive, confident, and independent, while women are described as sympathetic, sociable, interdependent, and relationship orientated (Deaux and Lewis 1984; Eagly and Steffen 1984; Langford and MacKinnon 2000; Williams and Best 1982). The social outcome is that men are considered more appropriate than women to fill public roles and responsibilities, and women are perceived as more suited to domestic roles or care professions, with men thus taking charge and women taking care (Hoyt 2010). Gender stereotypes are not just descriptive of how men and women are, but also prescriptive of how men and women should be (Eagly and Karau 2002; Rudman and Glick 2001). According to role congruity theory (e.g., Eagly et al. 2000), a group member will be positively evaluated when their characteristics are perceived to align with the requirements of the group’s typical social roles. In contrast, when people deviate from stereotypical expectations, they are perceived as violators of gender norms and encounter social and economic penalties (i.e., backlash, Moss-Racusin et al. 2010; Rudman and Phelan 2008; Cucchi and Cavazza 2020). For this reason, female leaders experience prejudice in stereotypically masculine domains because the social role of a good leader is inconsistent with being a good woman (Eagly and Carli 2007). Moreover, since gender stereotypes legitimise men’s privilege relative to women, they have been mainly examined when targeted at atypical (agentic) women (Rudman 1998). Nevertheless, social and economic penalties affect men as well. For instance, men encounter a backlash when they violate gender stereotypes associated with the chronic exhibition of strength along with the avoidance of weakness. In line with this, Moss-Racusin et al. (2010) have shown that modest men (vs. modest women) applying for a managerial job were perceived as less agentic and less likeable.

Politics is a field where stereotypically male features such as assertiveness, decision-making ability, independence of judgement and competence are considered among the foremost requisites (e.g., Norris 2004). As a consequence, in tandem with the robust congruence between stereotypical male qualities and the characteristics of the ideal political candidate, the incongruence between stereotypical female qualities and politics can severely undermine women’s political participation at all levels. For instance, as regards women’s aspirations to pursue a political career, research has shown that despite comparable credentials, background, and experience, women are less likely than men to consider themselves qualified to seek office (Fox and Lawless 2011), and they are less likely to become candidates when the representative is chosen by an election (Kanthak and Woon 2015). As regards whether gender stereotypes affect how people perceive women and men as being fit for politics, research has provided inconsistent evidence. Since gender stereotypes of women are at odds with voters’ expectations of politicians as tough and aggressive individuals, one would expect that stereotypes harm female candidates. Nevertheless, even if this result has been confirmed by some studies (Fox and Smith 1998; Rosenwasser and Seale 1988), other studies have shown that stereotypes have little effect on evaluations/descriptions of female candidates (Brooks 2013; Dolan 2014; Huddy and Capelos 2002; Sapiro et al. 2011), or they can even help female candidates when social welfare (Dolan 2010; Huddy and Terkildsen 1993) or health issues (Fridkin and Kenney 2009) are considered.

In addition, some conditions seem to be necessary for the gender stereotype to be activated: working on profiles of political candidates, Bauer (2015) showed that it is not gender per se but gender stereotype activation that influences the evaluation of candidates and hinders the intention to vote for women. In particular, she highlighted that female stereotype reliance is not automatic but requires activation, but once activated, it reduces support for female candidates. However, in this work, Bauer manipulated only the salience of feminine stereotypes, without considering the masculine ones. In the present study, we reasoned that if politics is still conceived as a “masculine thing”, then male (vs. female) target characteristics could activate a stereotype of the politically committed individual, making someone appear worthy of a vote. This is important, because the prototypical politically committed individual may be seen as an implicit norm for gaining a consensus and being politically efficacious. A shared prototype mainly characterised by male features could hinder women’s motivation to participate.

Overview

The aim of the present studies was to examine the role of gender stereotypes in affecting the intention to vote for women and men. We focus on the extent to which men and women are seen as representative of the ‘politically committed person’, as a function of their congruent or incongruent gender-stereotyped characterisation. More precisely, we were interested in examining whether making gender stereotype salient for a male/female target (Study 1) would affect the perceived probability that the target would be involved in political activities (stereotype activation) and affect the willingness of voting for him/her in case of candidacy (stereotype application). In addition, we wanted to test whether this effect would be due to the congruence between the sex of the target and the gender stereotype made salient or to the mere masculine characterisation of the target, irrespective of his/her sex (Study 2).

Study 1

In this study, we aimed at testing whether the salience of gender stereotype (vs control condition) in the description of a person makes a man appear more (and a woman less) representative of the politically involved individual (Hp1-stereotype activation) and a man more (and a woman less) worthy of a vote (Hp2-stereotype application). Moreover, we predicted that the prototypicality of the target as a political involved person (i.e., the heightened estimate of the male target to participate in politics) when described as a real man, in respect to the other conditions, mediated participants’ willingness to vote for him in case of candidacy, In other words, we expected a moderated mediation in which the interaction between being a man and being described in masculine terms should influence the willingness to vote for him through the perceived prototypicality of the man as a political involved person (Hp3, see Fig. 1).

Fig. 1
figure 1

The tested model of moderated mediation (Study 1)

Full size image

Method

Participants

An a priori power analysis was conducted for sample size estimation (Faul et al. 2007). With alpha = 0.05 and power = 0.80, the projected sample size needed to detect a medium effect size (f = 0.25, Cohen 1988) is approximately n = 128 for a between-groups comparison (ANOVA with four groups). We advertised the study through Facebook contacts and enrolled all individuals who responded and volunteered to participate in this online procedure, after giving their informed and explicit consent. Thus, we recruited a total of 436 Italian people (347 women, aged 18–70 years, M = 25.29, SD = 6.77) who voluntarily completed an online questionnaire presented as a study on everyday person perception.

Of these participants, 367 passed the manipulation check (i.e., they could correctly remember at least one out of the two adjectives manipulated in the target presentation—see below) and the rest (20.41%) were excluded from the analysis.Footnote 1 Of the remaining participants, 291 were women, and 75 were men, and their mean age was 24.92 (SD = 6.19). They were resident in the North (137, 37.3%), in the Center (79, 21.5%) and in the South (149, 40.6%) of Italy. Two hundred and fifty-four (69.2%) were students, 21 (5.7%) unemployed, and the others were workers. As for their educational level, 10 (2.7%) declared a junior high-school completion accreditation, 189 (51.5%) declared a senior high-school completion accreditation, and 165 (44.9%) had a university education. Three participants (2.7%) did not indicate their educational level.

Design and Procedure

On the first page of the questionnaire, participants were invited to carefully observe a picture of an individual along with a brief description. Participants were randomly distributed to a 2 (sex of the target) × 2 (stereotype activation vs control) between-participant design.

In all conditions, the description of the target included the name (Anna vs. Alberto), age, familial status, and two main traits (i.e., spontaneous and attentive). Finally, two other adjectives were manipulated so as to make a stereotypical sex description salient: for the man, a sentence specified that acquaintances described him as authoritative and a “real man”; in contrast, the woman was described as kind and feminine. In the control condition, both the targets were described as conscientious and sincere. The six adjectives were selected from the Bem inventory in the Italian version (Bem 1974; De Leo et al. 1986) as typically associated to men/women or both to the same extent.

After viewing the target, participants were invited to fill in a series of measures which included manipulation checks, estimate of the probability that the target was involved in political activities, likelihood of voting for him/her in case of candidacy, and the scale of benevolent and hostile sexism (Glick and Fiske 1996, in the Italian version of Manganelli Rattazzi et al. 2008). Finally, participants were invited to report their interest in politics (four 7-point items, α = 0.85), political orientation on the left–right continuum (range 1–11) and some socio-demographic information.

Dependent Measures

Stereotype Activation—Likelihood of the Target Participating in Politics

Stereotype activation was captured by a typical experimental task used for the study of the representativeness heuristic (Kahneman and Tversky 1982). We asked our participants to estimate the probability (on the 11-point scale anchored at 0–100%) of the target being involved in a series of activities. In this way, we could capture the gender stereotype connotation that participants associated with the person committed to politics (i.e., stereotype activation), avoiding the social desirability of a direct question on such a matter.

The questionnaire included a list of 10 activities. Five political activities were embedded in the list: being involved in politics, writing for an online magazine of political culture, participating in protest demonstrations, being a member of the municipal council, and managing a Facebook page on current political issues. The remaining filling activities were: writing poems, being a lawyer, being a teacher, being part of a drama group, managing a fashion blog. An explorative factor analysis with Varimax rotation including the 10 items sorted three factors, the first of which collected the five political activities (explained variance = 24.46%, eigenvalues = 2.45, factor loadings > 0.35).

From the response means about the 5 political activities, we computed an index of the estimated likelihood of the target participating in politics (α = 0.78).

Stereotype Application—Willingness to Vote for the Target

One further question concerned the willingness to vote for the target in case of her/his political candidacy (on the same 11-point scale anchored at 0–100%).

The dataset is available at https://osf.io/93dbk/.

Results

Table 1 reports descriptive statistics and intercorrelations for the measures and control variables.

Table 1 Descriptive statistics and correlations among the study variables. Study 1
Full size table

Effects of Target Gender and Gender Stereotype Salience

The first 2 (gender of the target) × 2 (gender stereotype salience) ANOVA on the perceived likelihood of the target participating in politics yielded the main effect of the target gender, F(1, 363) = 16.27, p < 0.001, η2 = 0.04. Not surprisingly, given the persistent gender gap in politics, respondents rated a higher probability for the men (M = 5.40, SD = 1.73) in respect to the women (M = 4.55, SD = 1.84) to be involved in political activities. Most germane to our primary concerns, the main effect of target gender was qualified by the significant predicted interaction with the gender stereotype salience manipulation, F(1, 363) = 8.77, p = 0.003, η2 = 0.02.

As depicted in Fig. 2, when estimating the likelihood of the target being involved in politics, participants did not differentiate according to the stereotype salience in case of a woman, F < 1, whereas the same estimate varied when the target was a man: in line with Hp1, respondents imagined a greater probability that he was involved in politics when he was depicted as a “real man” in respect to the control condition, F(1, 162) = 12.00, p = 0.001, η2 = 0.07.

Fig. 2
figure 2

Perceived likelihood for the target to participate in politics (stereotype activation) and willingness to vote for the target (stereotype application) as a function of their gender and stereotype salience (Study 1)

Full size image

A similar pattern of results emerged from the same ANOVA on the willingness to vote for the target in case of their candidacy. Contrary to what we expected (Hp2) but in line with previous experiments performed in Italy (e.g., Cavazza and Guidetti 2014; Cavazza 2016), participants reported a higher willingness to vote for the women (M = 7.03, SD = 2.30) than for the men (M = 6.38, SD = 2.24), F(1, 363) = 8.77, p = 0.003, η2 = 0.02. In addition, the target’s sex significantly interacted with the gender stereotype salience in influencing the likelihood to vote, F(1, 363) = 4.65, p = 0.032, η2 = 0.01. The men acquired a slight consensus from the stereotype salience, F(1, 162) = 3.28, p = 0.07, η2 = 0.02, whereas the consensus for the women was unaffected, F(1, 201) = 1.50, p = 0.22, η2 = 0.007 (Fig. 2).

These effects were not moderated by participants’ sex, interest in politics, self-reported position on the 11-point left–right continuum, nor by their benevolent or hostile sexism orientation. In addition, those control variables were not correlated with the mediator, with exception of interest in politics, nor with the dependent variable (see Table 1). Including the interest in politics as a covariate in the above ANOVAs did not change the pattern of results.

Finally, the gender stereotype salience manipulation did not affect any of the other five filling activities included in the questionnaire (i.e., writing poems, being a lawyer, being a teacher, being part of a drama group, managing a fashion blog) or interact with the target gender.

Moderated Mediation

We used PROCESS, the SPSS macro provided by Hayes (2013), to test a model in which the heightened estimate of the male target participating in politics when described as a real man, with respect to the other conditions, mediated participants’ willingness to vote for him in case of candidacy (model 7, with 5000 bootstrap resamples). The result of the moderated mediation analysis confirmed that this indeed was the case, in line with Hp3 (Table 2).

Table 2 Conditional indirect effect of the stereotype salience and target gender on participants’ stereotype application (willingness to vote for the target) through stereotype activation (estimated likelihood for the target to participate in politics). Study 1
Full size table

Discussion

In this experimental study, we found that, irrespective of respondents’ sex and level of sexism, a man is considered more likely to fit the category of the politically committed person when described as a real man, whereas the gender stereotype salience does not affect the representation of a woman as fit for politics. However, in this study, the manipulation was limited to the attribution of two gender-specific qualities to the target, thus comparing a gender-congruent stereotype activation with a control condition. We did not include the gender-incongruent stereotype (i.e., women with masculine qualities and men having feminine qualities), which might have differently promoted or hindered the perception of the male and female targets. This limitation prevents us from concluding that the observed effects are indeed due to the mere salience of gender stereotypes or rather to the congruence between the gender of the candidate and the salient stereotype. Furthermore, in a world in which visual information—in particular, the physical appearance of political candidates—has such a crucial role (e.g., Bauer and Carpinella 2018; Carpinella and Johnson 2016), we cannot rule out that an automatic activation of stereotype was also at work in the control condition based on the target image alone. This is why we devised a second study aimed at overcoming these two limitations.

Study 2

We ran Study 2 with a similar procedure to Study 1, but we included some significant differences to overcome the limitations outlined above. First, to check whether the observed effects were due to the congruence between the sex of the candidate and the salient stereotype or merely to the salience of the male stereotype characterisation, we ran a full 2 (sex of the target) × 2 (masculine vs feminine stereotype characterisation) factorial design. In this way, we compared the stereotype activation in the evaluation of a woman and a man as representative of the ‘politically involved individual’ when they are depicted as congruent or incongruent with their gender stereotype.

Moreover, given that the visual stimuli included in Study 1 might have automatically activated gender stereotypes also in the control condition, in the following study we did not include the image of the target.

Method

Participants

The same a priori power analysis conducted for sample size estimation for Study 1 was applied to Study 2 with the same experimental design (ANOVA with four groups). Through a snowball sampling, we enrolled all individuals who responded and volunteered to participate in this online procedure. Thus, we recruited a total of 258 Italian people (179 women, aged 18–56 years, M = 22.93, SD = 6.56) who completed an online questionnaire presented as a study on everyday person perception, after giving their informed and explicit consent.

Of these participants, 206 passed the manipulation check (see below) and the rest (20.15%) were excluded from the analysis.1 Of the remaining participants, 143 were women, and the participants’ mean age was 22.82 (SD = 5.96). They were resident prevalently in the Center of Italy (144, 69.9%) and were prevalently students (152, 73.8%). As for their educational level, 6 (2.9%) declared a junior high-school completion accreditation, 157 (76.2%) declared a senior high-school completion accreditation and 41 (19.9%) had a university education. Two participants (1%) did not indicate their educational level.

Design and Procedure

We followed the same procedure employed in Study 1 and the same material, except for the description of the target, who was only described as characterized by 4 stereotypical feminine adjectives (affectionate, sensitive to others’ needs, warm and kind) or masculine adjectives (competitive, willing to take risks, influential and ambitious). These eight adjectives were selected from the Bem inventory in the Italian version (Bem 1974: De Leo et al. 1986) as typically associated to men/women.

Participants were randomly distributed to a 2 (sex of the target) × 2 (feminine vs masculine stereotype characterisation) between-participant design. After reading the target’s description, participants were invited to fill in the same questionnaire of Study 1.

For the manipulation check, we asked respondents to recognise the adjectives used to describe the target among a list of eight adjectives—i.e., the four used in the masculine stereotype characterisation and the four used in the feminine stereotype characterisation conditions. To avoid an excessive decrease of the signal-to-noise ratio of the data set (Oppenheimer et al. 2009), we retained those respondents who remembered a majority of correct adjectives.

Dependent Measures

Stereotype Activation—Likelihood of the Target Participating in Politics

As in Study 1, we performed an exploratory factor analysis with Varimax rotation including the 10 items. It sorted two factors, the first of which collected the five political activities (explained variance = 37.64%, eigenvalues = 3.76, factor loadings > 0.62). Therefore, we calculated an index of the probability estimation for the target being involved in politics through five political activities (α = 0.86).

Stereotype Application—Willingness to Vote for the Target

As in Study 1, one question concerned the willingness to vote for the target in case of her/his political candidacy (on an 11-point scale anchored at 0 to 100%).

The dataset is available at https://osf.io/93dbk/.

Results

Table 3 reports descriptive statistics and intercorrelations for the measures and control variables.

Table 3 Descriptive statistics and correlations among the study variables. Study 2
Full size table

Effects of Target Gender and Gender Stereotype Characterisation

The first 2 (sex of the target) × 2 (gender stereotype characterisation: feminine vs. masculine) ANOVA on the perceived likelihood of the target participating in politics only yielded the main effect of gender stereotype characterisation, F(1, 201) = 46.39, p < 0.001, η2 = 0.19. Irrespective of the sex of the target, respondents rated a higher probability of involvement in political activities for the individual described in masculine terms (M = 5.54, SD = 2.10) in comparison to the one described in feminine terms (M = 3.74, SD = 1.68), see Fig. 3. The inclusion of the interest in politics as a covariate left this pattern of results unaltered.

Fig. 3
figure 3

Perceived likelihood for the target to participate in politics (stereotype activation) and willingness to vote for the target (stereotype application) as a function of their gender and gender stereotypical traits (Study 2)

Full size image

The same ANOVA performed on the willingness to vote for the target did not yield any significant effects. In particular, being characterised by feminine (vs. masculine) traits did not reduce the willingness to vote for the target in case of candidacy (see Fig. 3), even when controlling for the left–right respondents’ self-placement that correlated with the dependent variable (see Table 3).

Moderated Mediation

Following the procedure described by Hayes (2013) for estimating indirect effects, we tested model 7 using bootstrapping with 5,000 resamples to compute 95% confidence intervals. Our results indicated that the male stereotyped characterisation of the target increased willingness to vote for him/her in case of candidacy through the increased likelihood estimation for the target to be involved in politics (Table 4). These effects were not moderated by participants’ sex, interest in politics, self-placement on the 11-point left–right continuum, nor by their benevolent or hostile sexism orientation.

Table 4 Indirect effect of the stereotype characterization (feminine vs masculine) on participants’ stereotype application (willingness to vote for the target) through stereotype activation (estimated likelihood for the target to participate in politics). Study 2
Full size table

Discussion

Study 2 was performed to overcome the limitations of Study 1 and provided consistent findings. In particular, it clarified that gender congruency or incongruency per se did not promote or hinder the estimated probability of being involved in politics and the willingness to vote for the target. Instead, it confirmed that the salience of the male characterisation was enough to infer a higher probability of being involved in politics irrespective of the sex of the target, and, indirectly, a higher probability of attracting a consensus. Convergent results in respect to Study 1 were observed, although the target was only described to prevent a possible stereotypical activation by images.

General Discussion

Although gender equality in political participation is a crucial aspect of a fair and democratic society, the achievement of equal opportunities for women’s full participation in all spheres of political life in Western countries is still far from being reached. Previous research has carefully examined the factors that hinder the participation of women in politics, considering the role of hard and soft barriers (Lawless 2012). As regards this last point, there has been contradictory evidence regarding the role of gender stereotypes in affecting the intention to vote for male and female politicians. In the present research, we relied on a typical experimental task documenting the use of the representativeness heuristic, whereby people make judgements of likelihood by an assessment of the similarity of a sample to a population. Focusing on the intention to vote as the outcome, we aimed to examine the role of gender stereotype salience in affecting the extent to which women and men are viewed as representative of ‘the politically involved individual’ prototype, and the intention to vote for a male or female target. In Study 1, we found that a man is considered more likely to fit the category of the politically committed person when he is described as a real man, whereas the gender stereotype salience does not affect the representation of a woman as being fit for politics. Indeed, the presentation of a male target according to the gender stereotype increased the willingness to vote for him, while the presentation of a female target according to the gender stereotype slightly decreased the willingness to vote for her (although this last result was not significant). Moreover, we found that the increased perceived likelihood of the male target participating in politics when described as a real man, in respect to the other conditions, mediated participants’ willingness to vote for him in case of candidacy. Study 2 replicated Study 1’s findings, confirming that presenting male stereotypical characteristics activates a representativeness heuristic of the likelihood of the target participating in politics, which in turn makes the target more likely to receive votes, regardless of the target’s gender. Interestingly, in line with Sensales and colleagues (2018), our pattern of results was not moderated by participants’ sexism, showing that the idea that men (or male characteristics) are more suited to politics than women (or female characteristics) is widespread and present also in those people who do not endorse traditional views of gender relations. Our finding that a stereotypical description of a female target (Study 1) and of a subset of female stereotypical characteristics (Study 2) did not activate a corresponding ‘politically involved individual’ prototype is consistent with the sub-typing theory (Schneider and Bos 2014) and the leader-not-ladies theory (Brooks 2013), which show that male politicians share substantial stereotype content with the superordinate group of men, while female politicians share little stereotype content with the superordinate group of women. Since the stereotype content for female politicians has been shown as distinct from the stereotype content of women, and no traits that are stereotypical of women in general have been shown as strong descriptors of female politicians (Schneider and Bos 2014), future studies should examine whether people would still prefer to vote for men, or a target with stereotypical male characteristics, while also examining the subset of characteristics considered typical for female politicians, such as, for instance, being well-educated, well-spoken, and hardworking. In addition, our findings showed that the attributes we included in the two studies as most commonly associated to men (i.e. authoritative, competitive, willing to take risks, influential and ambitious) are those driving the perception of a target as fit for politics. However, we cannot definitely know what other male and female attributes are necessary for a stereotypical male or female job. This is why, further studies should test the applicability of these results with different descriptions—since we measured ratings of hypothetical individuals under minimal information conditions—and also consider whether gender stereotypes transcend party affiliation, as previous research has ascertained (Sanbonmatsu and Dolan 2009).

An original aspect of our research is that, in contrast to previous studies (e.g., Dolan 2010, 2014), we did not investigate whether gender stereotype salience affects the evaluation of competence and/or warmth of the target. Indeed, we showed that the perception of the likelihood of political commitment based on the representativeness heuristic was enough for men (or for male stereotypical traits) to support the ability of a social target to attract consensus. The normativeness of male characteristics could constitute an obstacle for women’s motivation to participate in political activities, or to be included on candidate lists. Moreover, a strength of the present study is that the experimental task employed allowed us to capture the activation of the gender stereotype while avoiding the social desirability to which a direct measure would have led. In sum, more than to evidence stereotypical women disadvantage as prospective political candidates, these findings reveal that stereotype activation could benefit male prospective candidates. Thus, this result is relevant not only for women but also for men who do not conform to gendered norms of hegemonic masculinity. It would also be interesting to understand if the same stereotypical male characterization would extend its positive influence on political consensus when applied to targets beyond the gender binary, such as genderqueer.

Although both studies were performed in Italy, we are confident that our results are generalisable to other contexts in which a gender gap in political participation is still evident. However, this must be confirmed by future studies.

Our research indicates that politics is still conceived of as a “real man” thing, since the stereotypical male description activated the stereotype of the politically committed individual worthy of a vote. It also contributes to the understanding of how gender stereotypes continue to influence political representation and decisions. Extensive efforts should be undertaken to enrich the visibility of female models—as well as counter-stereotypical male models—in politics in order to work against the idea that the prototypical politically committed individual is a man or a person with stereotypical male characteristics. This change would increase not only equal opportunities in politics for men and women but also change the way in which politics is conceived and realised.