1 Introduction

Political advertising is a major tool in political marketing. Growing feelings of political cynicism and a low involvement in politics by a significant portion of the electorate dispirit any active interest in news sources, which, in turn encourages and glamourises political advertising as a critical form of communication between candidates and voters (Kaid 1999; O’Cass 2005; Biocca 2013). Through advertising, the political candidate seeks to persuade or remind voters of his/her candidacy and, chiefly, to influence them positively in order to win their votes (Lees-Marshment 2009; Shen 2012).

The sponsor of political advertising varies between countries, depending on the electoral system. For instance, in countries like the U.S.Α. and France, voters have to choose between candidates who are standing for president as well as for other representatives, whereas in countries like Greece or the Netherlands, voters have to choose between the different political parties and their parliamentary candidates (Negrine and Papathanassopoulos 1996; Cheibub and Limongi 2002). Hence, in presidential systems political advertising campaigns focus on the image of presidential candidates, whereas in parliamentary systems (such as in the Greek electoral system) political parties develop national political campaigns. In addition, as far as the parliamentary systems are concerned, candidates who wish to enter the parliament develop political advertisements, usually aired by the local media, in order to influence voters in their constituencies.

Unlike other sources, such as political interviews and debates, advertising is an unmediated form of communication, which means that the content of the message is under the absolute control of the sponsor (Kaid 1997; Samaras and Papathanassopoulos 2006; Strach et al. 2015). This allows the politician to carefully construct a desired image, and thus promote his/her candidacy with optimum effectiveness to particular target groups (O’Cass 2002). Considering though that many people find political information complicated and difficult to process, the use of schemas may facilitate their processing and retention of information about candidates (Chang and Hitchon 2004). In this context, the candidate selects messages, verbal or visual, that are inter alia compatible with the convictions and expectations of the voters he or she is addressing (O’Cass 2002; Robinson 2010; Arendt et al. 2015). Gender schemas, or gender stereotypes, that represent a category of traits, attributes and behaviours commonly associated with men and women (Kittilson and Fridkin 2008; Wagner et al. 2017) fall among the general schemas. In that sense, by capitalising on the existing gender stereotypes, a candidate may relate to the positions, characteristics, behaviour and even the appearance of the ideal politician (Kahn 1996; Sanbonmatsu 2002; Turska-Kawa and Olszanecka-Marmola 2016). The present study examines this general proposition, by endeavouring to fill the following research gaps.

The use of gender stereotypes in advertising has received significant academic attention, counting more than 50 years of research (Gauntlett 2002; Wolin 2003; Zotos and Tsichla 2014). Most research focuses on women’s portrayals, propelled by the feminist movement and the evolution of gender roles in contemporary western societies (Stern 1999; Zotos and Tsichla 2014). Several studies investigate men’s portrayals together with women’s as a frame of reference (Zotos and Lysonski 1994; Hatzithomas et al. 2016); few studies though concentrate exclusively on men’s representations (e.g. Kervin 1990; Patterson and England 2000).

Gender portrayals in the political advertising arena have received only limited attention by the academic community. A few exceptions include the studies by Kahn and Goldenberg (1991), Huddy and Terkildsen (1993), Sanbonmatsu (2002), Fridkin and Kenney (2009) and Sanghvi and Hodges (2015) that provide evidence of the use of gender stereotypes in U.S. politics, and the studies by Fox and Smith (1998), Gordon et al. (2003), Fox and Lawless (2004), Smith et al. (2005) and Lee (2014) that concentrate on the effects of gender stereotypes on voters’ advertising perceptions and attitudes. Given the fact that millions of dollars are spent annually on political advertising (Hoegg and Lewis 2011), the need for further investigation into the field of gender stereotypes in political advertising should be stressed (Johnston and Kaid 2002; Fridkin and Kenney 2009). Moreover, the majority of prior studies were of North American or Western European origin (Lee 2014) and as such research is needed in other countries.

There is an ongoing debate among scholars over whether or not candidates should carry gender-congruent traits and issues (Lee 2014). A stream of research suggests that congruity between candidates’ gender and their personality traits can increase voters’ favourable evaluations (Herrnson et al. 2003; Banwart and McKinney 2005; Lee 2014), whereas a second group of studies supports the notion that gendered information does not affect voters’ attitudes (Bystrom and Kaid 2002; Bystrom et al. 2004). Though adequate research has been devoted to the effects of gendered information on voters’ attitudes, there is no evidence on how parliamentary candidates use gender stereotypes (in a congruent or incongruent manner) to make inferences about their issue positions.

Large-scale societal transitions and technological developments influence the long-term changes of political advertising campaign practices (Magin et al. 2017). The growing number of non-voters, swing voters, populist parties as well as the increase of party competition significantly influenced the content of advertising (Biocca 2013; Magin et al. 2017). Despite broad consensus in the literature that political advertising is constantly changing (Blumler and Kavanagh 1999; Blumler 2016), there is a dearth of longitudinal research primarily designed to understand these alterations. Hence, a longitudinal approach is expected to yield important insights regarding the possible changes in the frequency of the use of different stereotypes projected over time (Carlson 2001).

This study seeks to fill the void in the literature, by providing evidence from the Greek context. The purpose of the research is to investigate the manifestation of stereotypical portrayals in print advertisements of male parliamentary candidates in Greece over the period 1993–2009. Particularly, by analysing the status quo and monitoring potential changes in the stereotypical patterns, the study seeks to answer the following two questions: (a) what are the common stereotypes employed in print advertisements of male parliamentary candidates, and (b) what are the changes in the depiction of these stereotypes over the period studied. A better understanding of the interplay between societal alterations and the use of gender stereotypes in political advertising would provide researchers, politicians and advertisers with a more complete and detailed picture of trends in gendered political advertising.

The structure of the paper is as follows: Firstly, this paper begins with a review of the existing literature on gender stereotypes in political advertising. Next, the methodology of the study is thoroughly outlined, followed by the presentation of the findings and their relevance to the existing literature. The paper concludes with a summary of the key insights of the study, the acknowledgement of the limitations and the provision of directions for future research.

2 Literature review

2.1 Gender stereotypes and advertising

A stereotype is a representation of a social group emphasising a characteristic or a set of characteristics that may project a standardised image, whether it be correct or false, of that group (Moriarty et al. 2009). Stereotypes use easily understood, symbolic signs that are accepted by the majority of the members of a community, and they refer to a specific social context and time frame. In that sense, they represent social constructs that affect how the members of a society perceive other individuals or social groups (Gauntlett 2002).

Gender stereotypes are general beliefs about gender-related qualities and roles associated with psychological characteristics and behaviours of men and women (Browne 1998). According to Van Zoonen (1996), the meaning of gender is not constant, but varies along with specific historical and cultural conditions. In other words, the concepts of masculinity or femininity are socially constructed and dynamic, that is, susceptible to change and alteration over time. Prior research suggests that masculinity is the product of historical, social and cultural processes, and a response to changing roles of femininity (Kimmel 1987).

Gender stereotypes serve as mental structures integrating personal characteristics, behaviour and convictions into the social categories of “man” and “woman”. The classification is multilevel, relating to four different components: aspects of physical appearance (e.g. hair, figure, grooming, etc.), role-related behaviours (e.g. child rearing, staying at home, etc.), professional life (e.g. choice of occupation, professional commitment, etc.), and trait descriptors such as concern for others, self-perception, and means of social advancement and recognition (Deaux and Lewis 1984). The stereotyping of an individual by the other members of a society usually leads to oversimplifications and superficial – and consequently often mistaken – assessments (Eisend 2010). Real facts and actions may be overlooked, and opinions may be formed on the basis of an incomplete picture.

Since the 1950s, research interest in the investigation of gender portrayals in advertising has been paramount. Early studies document the existence of stereotypical depictions of men and women in commercial advertisements: Women used to be presented in domestic contexts (as housewives and mothers), dependent upon men, or in decorative roles (Zotos and Tsichla 2014). On the contrary, men used to be depicted in the public domain, in professional roles, occupied with activities outside of the house or as authority figures (Lysonski 1985). From the 1990s on, a number of studies argue in favour of a greater pluralism in the way the sexes are depicted, and suggest a shift towards more egalitarian portrayals (Klassen et al. 1993; Piron and Young 1996; Hatzithomas et al. 2016). At the same time, a new type of masculinity appears to emerge, as recently men seem to be much more concerned with their physical appearance than before (Patterson and England 2000).

Even though the social roles of men and women have dramatically evolved, advertising continues to keep a masculine perspective by portraying men and women in roles that do not reflect their attributes and individuality and by rarely depicting men in household settings or in family roles (Gentry and Harrison 2010; Zayer 2010; Tsichla and Zotos 2016; Verhellen et al. 2016). Longitudinal evidence from the Greek context aligns with the abovementioned picture: Data from commercial advertisements comparing the period 1982–1983 and 1987–1988 (Zotos and Lysonski 1994), as well as 1998–2002 (Doulkeri 2003) illustrate that men were usually projected as strong, dominant and professionally oriented.

2.2 Gender stereotypes in political advertising

The manifestation of gender stereotypes in political advertising is a critical and multilayered phenomenon. Gender stereotypes affect the manner in which media, party recruiters and voters assess men’s and women’s electoral prospects (Flammang 1997; Fox 1997; Niven 1998). When political candidates are concerned, voters who lack any information other than the candidate’s gender may infer the presence (or the absence) of certain traits and competences on the basis of readily available information (Gordon et al. 2003). In that sense, gender stereotypes may serve as “information shortcuts” for voters (Sanbonmatsu 2002, p. 21). Even in cases where a candidate is well-known, voters are likely to form opinions on the basis of stereotypical views of gender, quite possibly disregarding personality, achievements or political position (Huddy and Terkildsen 1993; Kahn 1996; Fox and Smith 1998; Sanbonmatsu 2002).

Gender stereotyping in campaign themes is usually signified in terms of personality traits and issue competence, paving the way to the formation of gender-congruent expectations (Huddy and Terkildsen 1993; Lee 2014). Traditionally, men have been largely associated with competence, rationality and agency (Broverman et al. 1972; Eagly and Steffen 1984), while women have long been associated with warmth, expressiveness and communion (Broverman et al. 1972; Eagly and Steffen 1984). Since people’s perceptions about gendered personality traits are not flexible (Gordon et al. 2003), voters expect female candidates to possess stereotypically “feminine traits” and thus being more attractive, emotional, sensitive, understanding, compassionate, honest, sincere, ethical, decent, polite, friendly and people-oriented than their male counterparts (Huddy and Terkildsen 1993; Koch 1999; Sanbonmatsu 2002). Conversely, male candidates tend to be associated with stereotypically ascribed “masculine” traits and therefore be seen as more success-oriented, competent, tough, dynamic, active, aggressive, assertive, rational, knowledgeable, and as possessing more leadership qualities and decision making abilities than women (Alexander and Andersen 1993; Huddy and Terkildsen 1993; Fridkin and Kenney 2009; Lee 2014).

In terms of issue competence, perceptions of voters are found to be associated with gender stereotypes about male and female candidates’ traits as well (Huddy and Terkildsen 1993). Particularly, women are perceived to be better qualified in dealing with “compassion” issues like poverty, education, health care, children rearing and family issues, whereas men are perceived as more suitable for “tougher” issues like defense, foreign policy, military, economy and crime (Shapiro and Mahajan 1986; Leeper 1991; Huddy and Terkildsen 1993; Panagopoulos 2004). In addition, female candidates are considered more suitable to deal with issues of special interest to women, such as sexual harassment, abortion and women’s rights (Huddy and Terkildsen 1993; Kahn 1996; Dolan 2005). It is interesting that the traits and issues voters tend to associate with female candidates are central to the traditional domain of the family (the private domain), whereas the policy expertise and characteristics linked to male candidates tend to be apparent in the public domain (Lawless 2004).

Candidates face two options regarding the incorporation of gender in the construction of their image and the tone of their communication and advertising campaigns: They can either reinforce gender-based expectations by reflecting the underlying stereotypes in all means of political communication, or they may choose to compensate for their gender-based expectations, so contradicting the existing stereotypes (Kahn 1993, 1996). For instance, female candidates were found to emphasise “masculine” traits in TV and internet self-presentations (Banwart and Kaid 2002), thus conveying a similar candidate image to their men opponents. However, given that gender is a highly visible attribute many researchers claim that political candidates have little choice but to respond to gender-based norms (Fox 1997; Panagopoulos 2004; Lee 2014).

Research evidence suggests that voters tend to favour “masculine” characteristics in their elected leaders in both national and local levels of political office, which culminates in a predisposition towards male over female candidates (Rosenwasser and Dean 1989; Huddy and Terkildsen 1993). Indeed, stereotypically “masculine” characteristics including self-confidence and assertiveness are vital for campaigning and fundraising (Leeper 1991). Conversely, the possession of “feminine traits” is not considered to be of particular advantage in the political realm (Huddy and Terkildsen 1993). In essence, since male politicians are perceived as more competent, tough and knowledgeable, and as possessing more leadership abilities than their female counterparts, it appears that gender-consistent messages will benefit male candidates more than messages that contradict prevailing stereotypes. Indeed, research findings demonstrate that male candidates tend to reinforce gender-based stereotypes in political advertisements (Panagopoulos 2004).

On the other hand, studies on stereotype violation indicate that counter stereotypical behaviour may produce feelings of dislike towards the stereotype violator (Rudman 1998; Rudman and Fairchild 2004). Similarly, according to Miller and Burgoon (1979) people have particular expectations about the appropriate communication behaviour for men and women, and violations of these expectations may affect persuasion. Hence, the following hypothesis is advanced:

  • H1: Male parliamentary candidates are portrayed more frequently as possessing stereotypically “masculine” traits (i.e. as successful and active/dynamic politicians), and less frequently as possessing stereotypically “feminine” traits (i.e. as attractive, friendly and honest politicians) in Greek print advertising.

2.3 The Greek context

The period between 1993 and 2009 was a period of major changes and developments on social, communication and political levels in Greece which affected the way politicians communicated with the electorate (Kotzaivazoglou 2011). In terms of Mancini and Swanson’s (1996: 19) descriptors, it was a time when the Greek society largely shifted from the “traditional archetype” to the “modern archetype”. The changes and developments that took place included (a) a change in the political culture, whereby the traditional forms of political communication based on partisanship, nepotism and personal relations gradually weakened and gave way to new forms of mass communication including political advertising, and (b) the changing position of women in Greek society, which redefined the roles and relation of the two sexes (Kotzaivazoglou and Zotos 2009). During that period, the educational level of Greek women superseded that of Greek men, whereas both the women’s employment rate and women’s participation in top executive ranks increased significantly (Koskina 2009). The evolution in women’s status brought about a corresponding evolution in egalitarian attitudes as well: In a study about gender inequality in the EU, Greece (together with Cyprus) records the greatest change in the EU as respondents report that gender inequality is far less widespread today than 10 years ago (Eurobarometer 2009). Also, in the same study, Greek respondents report the highest levels of optimism regarding the future portrayal of women stereotypes by the media and one of the highest scores regarding perceptions about the improvement of women’s rights, their income as well as their presence in traditionally masculine professions in the future.

Furthermore, the narrowing of the ideological distance between the parties of government (Loulis 1999; Spourdalakis and Tassis 2006) and the increasing public indifference to politics (Loulis 2003) resulted in an intensive use of marketing techniques (Yannas 2002, 2005). More importantly, political discourse seemed to fade in favour of the “rhetoric of image”, moving away the emphasis from the political party to the image of the candidate (personalisation of politics) (Papathanassopoulos 2000; Samaras and Papathanassopoulos 2006). It is evident that this shift in the focus of communication leaves considerable room for the diffusion of gender stereotypes. Political discourse became assimilated into and absorbed by the modes and techniques of commodity discourse (Heretakis 2002; Bouzas-Lorenzo 2010). Politicians were portrayed as commodities and advertised as if they were products, through advertising campaigns that emphasised their attractive personality characteristics (Heretakis 2013). Considering that advertising is a system of visual representation that both reflects and contributes to culture, (Hall 1980; Albers-Miller and Gelb 1996; Zotos and Tsichla 2014) it seems reasonable to assume that a more egalitarian depiction of male candidates will be found in Greek political print advertising.

Besides, the growing participation of Greek women in the electoral race for a seat in the Parliament should be noted. Regarding the early years of the period under investigation, only 18 women were elected in the 1993 elections and similarly, 19 women were elected in 1996. However, these numbers have grown significantly after the millennium, as 31 women entered the Greek Parliament in 2000, 39 in 2004 and 48 in 2007. In 2008, equal treatment legislation established for the first time a quota for candidacies available to women in political parties’ pool of eligible candidates, resulting in the election of 52 women in 2009 (Moschovakou 2015). This evolution is expected to influence the tone of political campaigns of their men counterparts, as the sex of the opponent is found to exert an influence in the tone of the candidates’ campaign messages as well (Banwart and McKinney 2005). Relevant studies illustrate that men running against female candidates were constrained in their behaviour and adapted the content of their campaigns in order to attract women voters (Fox 1997, 2000). Similarly, in a study of campaign data from the US context, Panagopoulos (2004) revealed that male candidates who run against women try to compensate for gender stereotypes by placing more focus on children, education, social security and the environment in their advertisements. In addition, in such cases men tend to portray themselves as “protectors” more often than when running against men (Panagopoulos 2004). Hence, the following hypothesis is formulated:

  • H2: Stereotyping in print advertising of male parliamentary candidates changes over the years: a) The portrayal of men as possessing stereotypically “masculine” traits (i.e. successful and active/dynamic politicians) shows a decrease, whereas b) the portrayal of men as possessing traditionally “feminine” traits (i.e. attractive, friendly and honest politicians) presents an increase.

3 Research methodology

3.1 Sample and research procedure

The objective of this study is to investigate the predominant gender stereotypes in print political advertisements of male candidates in the Greek parliamentary elections held between 1993 and 2009. The sample of the advertisements was drawn from newspapers. In Greece, during the period under examination, newspapers were the traditional medium of communication used for campaign advertising by parliamentary candidates (Doulkeri and Panagiotou 2005). Newspaper advertisements represent a relatively economical and well-targeted means of promotion and, due to legal restrictions, essentially the only means of mass communication available to candidates for advertising purposes. According to law 3023/2002, on funding of political parties and candidates, parliamentary candidates are not permitted to use TV, radio and outdoor media in their political advertising campaigns. It should be noted that in the earlier years of the period under scrutiny (1993–1999), social media advertising was underdeveloped and rarely used by male parliamentary candidates in Greece. Even in the second period (2000–2009), only a small minority of Greek parliamentary candidates rely upon social media to communicate their messages to citizens (Hermans and Vergeer 2013). Further, as law 3023/2002 imposes, TV spots tend to promote the party or the leader of the party and not parliamentary candidates.

Twelve national (Apogevmatini, Avgi, Eleftherotypia, Eleftheros Typos, Express, Imerisia, Kathimerini, Rizospastis, Ta Nea, To Ethnos, To Pontiki, To Vima) and eight local (Aggelioforos, Eleutheri Thraki, Empros, Macedonia, Patris, Peloponissos, Thessalia, and Thessaloniki) newspapers were collected based on their political orientation and readership over the period in question and analysed. The selection was confined to advertisements appearing in the last 30 days before each election, which in Greece is the “official” campaign period and is the preferred time for candidates to launch their advertisements. After removing duplicate advertisements, a total of 863 different ads for a variety of candidates were examined.

The sample selection process was multi-levelled (see Krippendorff 2004). First, the population was limited to male candidates, because they by far outnumber female candidates in Greece (Lappas et al. 2008). In particular, less than 17.3% of the total parliamentary seats were occupied by women for the period from 1993 to 2009 (Moschovakou 2015). Second, a set of six election campaigns, i.e. 1993, 1996, 2000, 2004, 2007 and 2009, covering a period of seventeen years was selected. The particular time span was considered reasonable for drawing conclusions that may reflect broader changes in the political communication. This was a period of significant social and political change in Greece (Kotzaivazoglou 2011). Prior studies (Viser 1997, 1999), also, have indicated that a period longer than 10 years is sufficient to measure any changes in advertising design.

3.2 Coding procedure

In order to address the purpose of the study, content analysis was used. This method was selected as particularly suitable for the systematic analysis, quantification and comparison of important communications data appearing in the content of different texts and messages (Weber 1990; Hansen et al. 1998; Bertrand and Hughes 2005), this being the case in this study. According to Hansen et al. (1998: 123), content analysis is the ideal method for research investigating how and how far different social and cultural phenomena, such as stereotypes, are reflected in communication messages, including news bulletins and advertisements, this being the case here (Hansen et al. 1998).

The content analysis instrument is adapted from Cwalina et al. (2005) (Table 1). According to the instrument, the image of politicians comprises two dimensions: leaders’ abilities and sociability. The element of leader’s abilities includes individual attributes such as being successful, active/dynamic and attractive, whereas the aspects of sociability are attributes such as being friendly and honest. In line with prior studies (Huddy and Terkildsen 1993; Holtz-Bacha and Kaid 1995; Schweiger and Adami 1999; Sanbonmatsu 2002; Fridkin and Kenney 2009) the categories of successful and active/dynamic politician, express mainly traditionally masculine stereotypes, whereas those of attractive, friendly and honest politician, are considered feminine stereotypes.

Table 1 Content analysis categories
Full size table

The use of political gender stereotypes drawn from the international literature is based on the assumption that these stereotypes are applicable to the Greek context, an assumption that is not unfounded considering the Americanisation of political communication in Greece during the period in question (Negrine and Papathanassopoulos 1996; Samaras 2008).

To ensure the reliability and validity of the research, the recommendations of Weber (1990) and Krippendorff (2004) were followed. A sample of 100 advertisements was selected and coded independently by two trained researchers, one a man, the other a woman. The study used Perreault and Leigh’s reliability index (1989), which takes into account the proportion of agreement, the number of categories codified and the number of persons taking part in the codification. The intercoder reliability for the variables used in this analysis averaged .83: namely .82 for successful, .83 for active/dynamic, .85 for attractive, .84 for friendly and. 82 for honest. Chi-square analysis is used in order to identify statistical differences in men’s portrayals over time.

4 Results and discussion

Table 2 illustrates overall percentages over the period 1993 to 2009, the percentage changes and the statistically significant differences in the proportional appearances of men’s stereotypes. Chi-square analysis was used to compare the frequencies of men’s stereotypes across the years, indicating that significant changes were documented in the use of stereotypes over time.

Table 2 Changes in the frequency of appearance of gender stereotypes in the advertisements of male parliamentary candidates over the period 1993–2009
Full size table

In particular, the stereotype of the successful politician is the one that appeared most frequently in the period under consideration, accounting for 40.7% of all cases (Table 2). In the advertisements where this particular stereotype is dominant, the most prominent elements in the candidate’s image are his professional capacity, political work, experience, education and social activity. The emphasis of the message is on “who he is”, and on projecting his abilities in order to persuade the voters that he deserves to be elected. A set of non-verbal elements concerning the candidate’s overall appearance – dress, posture, gaze – reinforce this message. Candidates in this category almost always appear wearing dark suits, conservative ties and white or pale blue shirts, a choice signifying success, authority, reliability and credibility. For instance, Akis Tsochatzopoulos’s advertisement for his parliamentary candidacy in 1993 (Panhellenic Socialist Movement), showed him dressed in a black suit, listed his achievements and emphasised the slogan “With confidence, strength and responsibility for the future of Greece”.

According to Doulkeri and Panagiotou (2005), elements of this stereotype occur in almost all political advertisements. This can be explained because the elements that characterise a candidate’s personality, such as his professional career, experience, intelligence and ability to succeed, are an integral part of his image and key selection criteria for voters (Schweiger and Adami 1999; Capelos and Huddy 2002). The frequent appearance of this particular stereotype seems to have an additional, equally important cause. The stereotype of the successful politician is related to a broader perception of masculinity that was dominant in the Greek society of the “traditional archetype”, and to the stereotype of the ideal man (Zotos and Lysonski 1994; Doulkeri 2003): the successful, independent, dominant male, the personification of toughness and leadership, a self-confident achiever in total control of his emotions.

The frequency of appearance of the stereotype of the successful politician gradually decreased over the period, in a statistically significant manner (X2 = 20.71, p < .001), although remaining the most frequently appearing stereotype in all the campaigns. In 1993 (47.7%) and 1996 (52.2%) the stereotype of successful politician was used in almost half of the advertisements, whereas in the campaigns of 2000, 2004, 2007 and 2009 its frequency of appearance dropped to 41.7%, 37%, 35.9 and 27.8%, respectively. The emerging discussion on the need for “feminisation” of politics, “with the insertion and integration of women both in numbers and ideas” into the political process (Lovenduski 2005, p. 12) seems to have paid off, leading to a less masculine portrayal of parliamentary candidates.

The second most frequently used stereotype, representing 28.5% of the total ads (Table 2), is that of the active/dynamic politician. Candidates in this category present themselves as fighters, active, assertive, defending their country and protecting their fellow citizens. For instance, the parliamentary candidate Sotiris Kouvelas (member of the New Democracy Party in 1993) used as a slogan the statement “In the battle ahead”, in order to symbolise his role as a fighter for the rights of people. There seem to be two reasons for the high frequency with which this particular stereotype appears in the advertisements of the period in question. One is related to the public’s keen interest in a better life: in their effort to send an “enticing” message to voters, the candidates promise to go all-out for their country and their fellow citizens. The other is related to the common perception of masculinity, of which assertiveness, vigour, courage, even aggressiveness are key components (Badinter 1994; Doulkeri 2003). Huddy and Terkildsen (1993) reported that this is the most representative male stereotype in the political world. Male politicians stress toughness and aggression as prominent characteristics of their image, linking them indirectly to their ability to handle difficult situations decisively and effectively (Fridkin and Kenney 2009). In a similar vein, Conroy (2007) supports that masculine candidates tend to receive more votes that feminine candidates, regardless of their sex. Especially during hard economic and political times hegemonic masculine “protectors” can be fully trusted to lead the “frightened masses” (Messner 2007). The stereotype of the active/dynamic politician, however, gradually appeared less frequently over the same period. As Table 2 shows, the rate of appearance of this stereotype slipped from 33.8% (1993) to 31.1% (1996), 32.1% (2000), 27.8% (2004), 24.8% (2007) and 19.1% (2009). The differences among the years are marginally statistically significant (X2 = 9.51, p < .09) (Table 2).

According to the research findings, the stereotype of the attractive politician was the third most popular stereotype in the period under consideration, occurring 16.9% of all cases (Table 2). In the relevant advertisements, candidates focus on their youth and attractiveness and offer a vision or merely some hope for a better future, using emotional language. They aim, in other words, to influence voters mainly by evoking positive feelings, an approach which departs from the typical image projected by male politicians and resembles the style of female candidates. Particularly, it concentrates more on appearance, youth, personal interest and emotional content (Huddy and Terkildsen 1993; Capelos and Huddy 2002). For example, a promising and handsome politician, Stavros Kalafatis (New Democracy Party), as a candidate for the parliamentary elections of 2000, highlighted his youthfulness and attractiveness by using the slogan “Greece will be more beautiful tomorrow”. The stereotype of the attractive politician rose from 9.9 and 11.2% in the 1993 and 1996 campaigns respectively, to 16.7% in 2000, 20.4% in 2004, 21.4% in 2007 and 22.6 in 2009 (X2 = 14.45, p < .013). Recent studies (Stockemer and Praino 2015) indicate that low involved voters use physical appearance of political candidates as a primary criterion in their vote choice. The growing number of uninformed and politically unknowledgeable voters may exert an increasing effect on the use of the stereotype of the attractive politician. In the same vein, there is a thriving discussion about the penetration of the political sphere by celebrity culture (Street 2012). Politician celebrities, such as Barack Obama, celebrity politicians, such as Schwarzenegger, and other attractive famous faces try to seduce audiences and attract votes (Marsh et al. 2010). Hence, politics is being “celebritised” and shifted from substance to surface.

The stereotype of the friendly politician appeared in this study with a frequency of 10.3% (Table 2). The message of the advertisements in which this was the dominant stereotype targets the ordinary, average voter. The candidate dresses casually, has the image of the ordinary man-in-the-street, and is usually pictured as smiling, likable, approachable and friendly. He is presented as being just the same as his fellow citizens, implying that he can empathise with their needs and daily lives. For instance, Haris Kastanidis, a parliamentary candidate of the Panhellenic Socialist Movement (2000), smiling and dressed casually, supported the slogan “For people, for their hopes”. With this particular type of portrayal the candidate seeks to create the image of a politician who cares about people, in order to endear himself to the voting public and build up a sort of mediated familiarity. Like the previous stereotype, this style of projection diverges from the standard approach of male politicians. As noted above, sensitivity and an interest in the ordinary citizen and his daily life are characteristics more traditionally associated with women (Huddy and Terkildsen 1993; Sanbonmatsu 2002; Fridkin and Kenney 2009).

The appearance rate of the friendly politician rose in a statistically significant fashion (X2 = 51.84, p < .001) from 4% in the 1993 campaign and 2.2% in 1996 to 7.7% (2000), 9.9% (2004), 15.2% (2007) and 26.1% (2009). This growing tendency can be explained by the increased, ongoing personalisation of politics in the last decades (Corner and Pels 2003; Karvonen 2010): voters decide on the basis of their view of politicians rather than their evaluation of the party or the platform (Manin 1997). Hence, contemporary politicians try to create a bond with their target audience through communication techniques that make them seem almost a part of the audience (i.e. using the element of homophily) (Bronstein 2013).

In this sample population, the stereotype of the honest politician was dominant in only 25 advertisements (2.9%, see Table 2). More than a third of the total occurred in the 2004 elections, possibly because the issues of integrity and consistency were high on the political agenda of that particular pre-election period. Candidates seem to have chosen this type of portrayal in order to differentiate themselves from other candidates and reflect the voters’ desire for ethics and consistency in politics. Honesty, transparency, sincerity, consistency, morality and the honouring of promises made are elements that voters want to see in candidates, since politicians are often accused of hidden dealings, corruption, intrigue, exaggerated pre-election promises and inconsistency (Capelos 2005). For this reason, some candidates choose to advertise their integrity and moral character, as well as their consistency of word and action (Schweiger and Adami 1999). For example, Stelios Papathemelis (member of the Panhellenic Socialist Movement in 2000) emphasised his integrity and honesty by using the slogan “He speaks the truth, he deserves your vote”. No statistically significant differences in the stereotype of the honest politician were recorded in the period under consideration (X2 = 3.21, p < .668).

According to our analysis, hypothesis 1 is supported. Μale parliamentary candidates are portrayed more frequently as successful and active/dynamic politicians, and less frequently as attractive, friendly and honest in Greek newspaper advertising. Hypothesis 2a is also supported, since the stereotypes of successful and active/dynamic politician show a statistically significant decrease, whereas hypothesis 2b is partially supported since only the portrayals of attractive and friendly politician present a statistically significant increase.

5 Conclusions, limitations and suggestions for future research

The purpose of this study was to examine the dominant gender stereotypes in print political advertisements of male parliamentary candidates in Greece over the period 1993–2009. The study sought to answer two questions: (a) what was the frequency of appearance of the specific stereotypes over the period in question, and (b) what changes occurred in the frequency of appearance of those stereotypes over time.

Overall, the two most frequently appearing stereotypes were those of the successful and the active/dynamic politician, stereotypes closely associated with traditional “masculine” traits. These stereotypes are considered to be the most characteristic male stereotypes in politics (Huddy and Terkildsen 1993; Doulkeri and Panagiotou 2005; Fridkin and Kenney 2009) and align to the image projected by political candidates in other countries such as the U.S.A. (Fox 1997; Carlin and Winfrey 2009). However, the study documented significant changes in the projected stereotypes over time. While the stereotypes of the successful and the active/dynamic politician remain the most frequent, a statistically significant decrease was documented over the period in question. On the contrary, the depiction of politicians as attractive and friendly showed a statistically significant increase.

This change seems to be mainly due to the related convergence observed between men’s and women’s stereotypes in politics (Huddy and Terkildsen 1993). It can be interpreted as the result of a wider change in the position and the role of the two sexes in the society that have shaped the perception of masculinity (Badinter 1994; Gauntlett 2002). The new masculinity wants the male politician less tough, rational, dynamic and assertive, and more attractive, emotional, friendly, decent, honest, sensitive and forward-looking (Badinter 1994; Clare 2000; Gauntlett 2002). The recent emergence of attractive and young new politicians such as Alexis Tsipras (Prime Minister of Greece, Syriza party), Stavros Theodorakis (the leader of the political party “the River”), and Nikos Androulakis (secretary of the Panhellenic Socialist Movement) appears to be in line with the findings of the present study.

Moreover, another interpretation is that the focus of political communication appears to have shifted gradually from the candidate persona to the reason why voters would benefit from his election and from quasi-rational arguments to emotionally charged messages. In other words, advertisements tend to focus more on the voter (Robinson 2010), moving away from candidate-oriented pitches, where the dominant element was the characteristics of the candidate (e.g. his knowledge, experience, capability, energy and fighting spirit) to voter-oriented messages focusing principally on the needs and wants of the voter, such as the vision of a better life or an interest in the daily life of the average citizen.

In addition, these findings may be attributed to the greater participation of women in politics that fuelled an adaptation of men’s political campaigns in order to combat their strengths and attract women voters. In particular, the increased portrayal of traditional women’s stereotypes in the political advertisements of male candidates may constitute an attempt to enrich their unchallengeable male identity with characteristics, positions and attributes that have a resonance with the electorate and in this way to expand the electoral base targeted, adding more groups of voters, such as young people and women.

This study contributes to the existing literature by developing an understanding of the use of gender stereotypes in political advertising over the years, as it is the first such research referring to the Greek context. In the contemporary political environment, the role of advertising that capitalises on existing myths and symbols and relies on frequent exposure by the mass media has become paramount in the formulation of the ultimate political communication strategy (Richardson 2003). The diffusion of gender stereotypes by political advertising may exert considerable influence to the electorate, as gender-trait stereotypes affect the perceived competency of candidates on a wide range of issues (Huddy and Terkildsen 1993), while even high levels of political information do not lessen the likelihood of gender stereotyping (Lawless 2004).

The findings of the study also bear managerial implications, in that they provide political candidates with an opportunity to understand the status quo for a seventeen year time span, and manage effectively their personal characteristics and political convictions in order to shape or renew an image that endows them with a competitive edge. According to Davies and Newman (2006), image is one of the most important factors affecting electability and voter sentiment, while the aestheticisation of politics (Beck 1994) underlines the preoccupation of voters with image, presentation and stylishness.

On the other hand, the reinforcement of a traditional masculine persona by men and accordingly the reinforcement of traditional women’s traits by women seems to have positive effects in the short term, but eventually it leads to the perpetuation of gender stereotypes (Kahn 1992). Particularly, the persistence of gender stereotypes has considerable social implications for female candidates. Voters’ perception of gendered personality traits is one of the most discouraging factors for women’s occupation in politics (Abdela 2005). Considering that the participation of Greek women in the Parliament accounts for 23.8% (2015), a percentage significantly below the mean value of the European Union countries that equals to 37.3% (European Parliamentary Research Service 2017), social policy concerns are raised.

This study concerns Greek male parliamentary candidates in the period 1993–2009. Even though the findings align with similar studies in the U.S.A., since this is a research endeavour that employs a cultural outlook, care should be taken regarding the generalisability of the findings. The results could perhaps be enriched by using material drawn from other media, such as advertisements on personal websites, in social media and in other forms of self-presentation. The results of this study could stimulate further research that would allow for relative comparisons. Future research could examine gender stereotypes projected in the advertisements of female candidates so as to determine whether they also follow traditionally ascribed gender roles and monitor changes in their representation over time.