Introduction and state of the question

Women are facing diverse challenges in exercising and communicating their leadership styles. Leadership—particularly, the political leadership—has been culturally conceived as based on attributes that have been chosen to characterise the stereotype of masculinity. Traits that are commonly associated with the leadership role, such as power, authority, efficiency, and achievement, are also more frequently assigned to the male gender role (Morales & Cuadrado, 2011). This phenomenon is evident when analysing some of the frames used by the media to refer to women's political leadership. There are frequent "frames" such as "the newcomer", which underestimates women's capacities to hold political positions, or "women leaders as agents of change", which suggests that women are only necessary at exceptional points in time or areas where certain transformations are required (Norris, 1997). Women political leaders also receive more negative coverage, with a predominance of stereotypical themes and traits (Van-der-Pas & Aaldering, 2020). During election campaigns, media coverage is also unequal and biased: not only do male candidates receive more invitations and mentions in the media, but coverage of female candidates is characterised by strong stereotyping (Baxter, 2017; Winfrey & Schnoebelen, 2019), mainly focused on their physical appearance and domestic roles related to care work (García-Beaudoux et al., 2020). While there is a changing trend in most current policy arenas in the direction of parity, gender-biased coverage of women's leadership has seen little changed (Wagner et al., 2019).

According to role congruity theory (Eagly & Karau, 2002; Eagly, 2005), women sometimes choose to exercise and communicate styles aligned with stereotypically male standards in order to overcome the bias that exists against their leadership because of the cultural role expectations held about both the female gender and leadership. To this end, they reinforce the communication competency of hard leadership skills, such as technical skills, or talents related to managerial, executive, and strategic planning tasks. This behaviour is often socially penalised, as stereotypically masculine conduct is punished and scolded when it is engaged in by women (Eagly & Karau, 2002). Waters et al. (2019) have warned that media coverage of women politicians reinforces female stereotypes, and that this poses a threat to them: women leaders who are not perceived as "feminine" because they tend to speak their minds or publicly state their ambition, risk being criticised for defying the gender role imposed on them.

In short, a woman who exhibits behaviour regarded as male will face social disapproval because she fails to meet the expectations associated with her gender. However, if she exercises and communicates a form of leadership characterised by strong soft skills, such as empathy, emotional intelligence or interpersonal intelligence, the quality and extent of her hard skills and her leadership ability will be questioned (García-Beaudoux et al., 2020). This study also addresses the phenomenon of personalisation, as it is one of the communication tools that helps to convey government messages and actions. The presence of the political leader turned into a political symbol representing a party has been a trend since television became a mass medium. The humanisation, celebrification, and hyper-personalisation of political figures have increased with the growing use of social media such as Facebook, Twitter, and Instagram (Berrocal-Gonzalo et al., 2022; Mazzoleni & Bracciale, 2019), in which the charm of high-profile political figures is used to communicate political content (Dader, 2012). Moreover, this tendency to place the leader at the centre of the political process is not only characteristic of the activities of candidates during an election campaign (Moreno-Díaz, 2022; Ferré-Pavia & Codina, 2022), but also of some political leaders in office. Personalisation as a communication strategy entails a somewhat simplified understanding and interpretation of political facts (Rebolledo, 2017) and is part of the phenomenon known as politainment (Schultz, 2012), which is understood as the blending of political information and entertainment. Two communication tactics that are particularly suited to both the logic of personalisation and of simplification are storytelling and “storydoing”. Whereas leaders use storytelling to facilitate electoral or governmental political communication (D'Adamo & García-Beaudoux, 2016), they use “storydoing” to use images that show them playing a key role in actions in various scenarios. In this way, a visual record is created that portrays consistency between saying and doing, with the aim of increasing their credibility and legitimacy (Sarasqueta, 2020).

Personalisation and political leadership style are particularly evident in social networks because of their strong visual component. Instagram is a platform of particular interest for two reasons. While its visual nature and applications facilitate the personalisation of politics (Gómez-García et al, 2019), it is also used as an information medium, especially by young people, which provides fertile ground for exploring the styles of political leadership to which youth are exposed. Digital media have been recognised as a new agent of socialisation, particularly in relation to the political socialisation that may be taking place within social media, especially regarding young people (De-la-Garza-Montemayor et al., 2019). Smith (2016) noted that Instagram is the second most engaged social network, as 60% of its users’ log in daily and the biggest population group among Instagram users is between 18 and 24 years old (Fondevila-Gascón et al., 2020).

Political communication has become a necessary element not only to win elections, but also to promote and legitimise government actions (Rubio, 2011). The use of social media as a means of communication allows political leaders to connect with citizens, encourage their participation, promote dialogue, and leverage the feedback, comments or opinions received (Herrera-Aguilar & O'Quinn-Parrales, 2019). Social networks create spaces for deliberation and accountability where there were none before. New technologies have transformed politics and the way citizens and politicians communicate with each other.

Material and methods

This paper explores the characteristics of the leadership styles communicated through the official Instagram profiles of fourteen women elected for executive government positions. They come from four continents: Oceania, Asia, Europe and America. Seven are, or were, heads of government, six are or were mayors, and one is the president of an autonomous region in Spain. This is a sample of women political leaders who served in executive political roles during the health emergency resulting from the COVID-19 pandemic. The sample is diverse in terms of variables such as age, geographical origin, political ideology, interests, education and experience.

The analysis is based on the premise that, in relatively general terms, the media frame the presentation of women's leadership using headlines leaning towards homogenising genders, such as "the female leadership style"; "a new style of leadership"; "is the female leadership style more successful?". ¹ The results presented in this article question this expectation of uniformity.

This study examines variations in the variable " leadership styles as conveyed" by women leaders in relation to four other categories: political ideology, generational affiliation, level of government, and preferred techniques for communicating government actions. The relationship between the leadership style conveyed in social networks and the degree of personalisation exhibited by each of the women leaders in their publications is also explored. The purpose of investigating the relationship between these two variables (leadership style as conveyed in the media and personalisation) is to find out how the personal image is used when communicating government actions. The study is descriptive and exploratory in nature and was conducted over a cross-sectional time horizon. The method used was comparative analysis (Nohlen, 2013) combined with content analysis (Krippendorff, 1990; Neuendorf, 2002; Igartua, 2006). The content posted on the feeds of the verified Instagram accounts of fourteen elected female politicians in government executive positions was systematically compared.

The sample was selected according to three criteria. They were all women leaders who: (a) held elected governmental executive positions; (b) governed a population of over 800,000 people; and (c) headed national or local governments during the COVID-19 pandemic. As the States and Autonomous Regions they led are very diverse, the intensity of the crisis faced, and their communication needs are likely to have been different as well. However, the COVID-19 pandemic provided some communication scenarios that were common to all of them, such as the need to communicate measures related to the use of masks, rules for isolation, use of public space, and vaccination protocols, among others.

Several studies have shown differences in the ways that political leaders used communication during the pandemic. Watkins and Clevenger (2021) found that some focused on the dissemination of useful information and specific crisis management measures, while others downplayed the crisis or provided ineffective information. Waylen (2021) argued that the effectiveness of the new rules and measures communicated by some leaders to deal with the crisis was negatively affected in the cases of those with a hypermasculine leadership style, such as Boris Johnson. Drylie-Carey et al. (2020) compared the public communication of Boris Johnson, Emmanuel Macron, Pedro Sánchez and Giuseppe Conte during the COVID-19 health crisis and found that the image each of them portrayed on Twitter was vastly different from one another.

The sample for this study consisted of the following fourteen women leaders: seven heads of government, six mayors and one president of a Spanish autonomous region. They were Angela Merkel (Germany), Jacinda Ardern (New Zealand), Sanna Marin (Finland), Mette Frederiksen (Denmark), Erna Solberg (Norway), Katrin Jakobsdottir (Iceland), Tsai Ing-Wen (Taiwan), Anne Hidalgo (Paris), Virginia Raggi (Rome), Ada Colau (Barcelona), Claudia López (Bogotá), Claudia Sheinbaum (Mexico City), London Breed (San Francisco) and Isabel Díaz Ayuso (Madrid Autonomous Region). The following table details the type of government, year of birth, generation to which they belong, political affiliation, and region that they govern.

The study analyses the communication posted on their verified Instagram accounts. Each publication was considered a unit of analysis. The comparison required a total of 2,330 units of analysis to be reviewed. These units constituted the total number of publications made by the fourteen women leaders selected during the period from 1 January to 30 June 2021. The review covered all the posts they made, including photos, videos, albums, and badges, with their respective texts on their Instagram feeds. The decision about coding the posts in the feeds and not those shared through the stories was based on permanence: the stories had a maximum duration of 24 hours, while the content posted in the feed makes it possible to outline and communicate a profile that will be permanent. Selecting six months as a cut-off period was intended to allow an evaluation to be made of how they communicated their leadership style in a context shared by all of them, namely, the health crisis associated with the COVID-19 pandemic and the start of mass vaccination, an unprecedented, highly complex operation which also constituted a unique communication challenge.

The choice of Instagram as the platform to analyse was based on the fact that, in the political arena, it is considered a platform with the ability to improve the exchange that takes place between leaders and citizens; to set agendas and trends; and to mobilise and enhance the construction of leadership, as it is a network that facilitates personalisation (Towner & Lego-Muñoz, 2017; García-Beaudoux & Slimovich, 2021; Pineda et al., 2020).

The hypothesis underpinning this study is that the leadership styles communicated by women leaders are not homogeneous in terms of gender, but vary according to variables such as their positions on the ideological spectrum, the level of government they exercise and their generational affiliation. Similarly, it is expected that there will be differences in the degree of personalisation of their communication and in the preferred techniques for communicating government actions to citizens. The research objectives were as follows:

• O1. To investigate whether there were statistically significant differences between women leaders in the type of leadership according to their ideological tendency.

• O2. To examine whether there were statistically significant differences in exercising leadership between the women leaders selected according to the generation to which they belonged.

• O3. To observe whether there were statistically significant differences in the type of leadership between the women leaders selected according to the level of government.

• O4. To determine whether there were statistically significant differences in the type of leadership between the women leaders selected according to the degree of personalisation.

• O5. To establish whether there were statistically significant differences in the type of leadership between the women leaders selected according to the narrative technique used.

The variables and their indicators, included in the research design, were selected after an extensive literature review, and resulted from categories created and used by the authors of previous similar research (D'Adamo et al., 2015; D'Adamo & García-Beaudoux, 2016; García-Beaudoux, 2017), which have also been cited and used by other researchers (Quevedo-Redondo & Portalés-Oliva, 2017; López-Rabadán & Doménech-Fabregat, 2018).

Variable 1: Generational affiliation.

• Conceptual and operational definition: According to their date of birth, the women leaders were classified into one of the following three generations: "Baby Boomer" (people born between 1949 and 1968), "Generation X" (people born between 1969 and 1980), "Millennial" (people born between 1981 and 1993).

Variable 2: Ideological self-positioning.

• Conceptual and operational definition: Position of the women leaders selected in the ideological spectrum from left to right, according to how they defined themselves in media interviews, social networks and election campaign information.

Variable 3: Level of government.

• Conceptual and operational definition: This variable could have one of two values: national level (for presidents and prime ministers) or local level (for mayors and the president of an Autonomous Region).

Variable 4: Leadership style as conveyed in the social networks.

• Conceptual definition. This was defined as a variable that could have two dimensions: a style characterised by the communication of hard skills and a style characterised by the communication of soft skills.

1) Hard leadership skills are those skills that focus on communicating competencies, technical and/or strategic skills.

2) Soft skills are those that focus on conveying communication skills, interpersonal and social skills, emotional intelligence, and the ability to cooperate and/or work in a team (García-Beaudoux et al., 2020).

• Operational definition.

(1) Hard leadership skills were operationally defined by the presence of the following indicators in the images or texts: efficiency, management skills, administrative skills and/or strategic planning skills of the women leaders.

(2) Soft leadership skills were operationally defined by the presence of the following indicators in the images or texts: aspects of personal lives, showing emotion, relationships with their work team, and/or direct interactions with citizens (García-Beaudoux et al., 2020).

(3) A leadership style was considered to be mixed when the images and texts analysed showed an equal proportion of hard leadership indicators (efficiency, management skills, administrative skills, strategic planning skills), and soft leadership indicators (aspects of personal lives, showing emotion, relationships with their work team, and/or direct interactions with citizens).

Variable 5. Personalisation of communication.

• Conceptual definition. This is communication in which organisations, public institutions or political measures are not presented in and for themselves but are represented by political leaders who give them a face and a voice (Marcinkowski & Greger, 2002).

• Operational definition. Presence of the women leaders selected in the images and/or videos posted. This variable could have only these two values.

Variable 6. Thepreferred technique for communicating government actions.

• Conceptual definition. The communication technique most frequently used by women leaders to publicly share government information, decisions and/or actions.

• Operational definition. Two indicators were used: communication of hard data, statistics and arguments that appeal to logic; communication through storytelling techniques (narration of own- or third-party stories, individual or community cases, or anecdotes) (D'Adamo & García-Beaudoux, 2016) or “storydoing” (narrative based on the active participation of the leader as a key player in political, social or community events) (Sarasqueta, 2020).

Analysis and results

Regarding the political ideology variable, the result of a chi-square test showed that the women leaders who defined themselves as being on the right of the political spectrum communicated a leadership style characterised by hard skills, while those who positioned themselves on the left of the ideological spectrum communicated a leadership style that focused on soft skills.

A chi-square test was also carried out in order to assess whether there were differences in the leadership styles as communicated on Instagram according to the generational affiliation of the women politicians selected. The results showed that baby boomer leaders tended towards communication characterised by hard skills, millennial leaders communicated more soft-skilled leadership, and Generation X leaders had a balanced mix of hard and soft skills communication.

A chi-square test was also conducted which showed that the women who held executive positions at the local level (mayors and the president of an autonomous region) communicated a leadership style characterised by hard skills, while soft skills predominated in the communication by those who governed at the national level.

Regarding the preferred techniques used for communicating government actions, the chi-square test showed that the female leaders who were inclined to use storytelling and “storydoing” generally communicated predominantly using soft skills, while those who used statistics and logical arguments used leadership styles with a predominance of hard skills in their communication.

A chi-square test was then carried out which showed that the women leaders in the sample who did not resort to personalisation in their communication exhibited either a hard or a mixed leadership style, while women leaders who used personalisation were characterised by a soft leadership style in their public communications.

Thus, it was observed that the tendency to communicate hard or soft leadership styles was not uniform but varied according to the political ideology of the women leaders, their generational affiliation, and the level of the position they held (local or national). Additionally, although all of them used Instagram as a platform to communicate their government actions to citizens, the degree of personalisation was not consistent.The main communication characteristics identified for each leader are outlined below:

• Ada Colau: soft leadership style, 61.6%; personalisation of communication, 75.6%; and argumentation, 15.1%.

• Angela Merkel: hard leadership style (66.7%); personalisation of communication (92.2%); and “storydoing” (82.9%).

• Anne Hidalgo: soft leadership style (50%) and hard leadership style (50%); personalisation of communication (71.1%); and “storydoing” (60.5%).

• Claudia López: hard leadership style (61.1%); personalisation of communication (69.5%); and argumentation, 34.7%.

• Claudia Sheinbaum: hard leadership style (69.4%); no personalisation of communication (73.4%); and argumentation (20.7%).

• Erna Solberg: soft leadership style (75%); personalisation of communication (77.4%); and “storydoing” (69.8%).

• Isabel Díaz Ayuso: soft leadership style (67.3%); personalisation of communication (87.8%); argumentation (2%); and “storydoing” (2%).

• Jacinda Ardern: soft leadership style (53.2%); personalisation of communication (55.6%); “storydoing” (52.4%).

• Katrin Jakobsdottir: soft leadership style (68%); personalisation of communication (87.2%); and “storydoing” (53.2%).

• London Breed: hard leadership style (64.2%); personalisation of communication (52.2%); and argumentation (55.2%).

• Mette Frederiksen: soft leadership style (62.9%); personalisation of communication (82.1%); storytelling, 51.2%.

• Sanna Marin: hard leadership style (53.2%); personalisation of communication (96.2%); storydoing, 59%.

• Tsai Ing-wen: soft leadership style (49.2%); personalisation of communication (67.7%); “storydoing” (50%).

• Virginia Raggi: hard leadership style (53.6%); no personalisation of communication (60.3%); and argumentation (47.3%).

Discussion and conclusions

The above results lead to the conclusion that the leadership style communicated by the sample of women politicians elected to government executive positions varied according to the values of other variables, such as their generational affiliation, their political ideology and the level of government where they held a position. Regarding the variations in the communication of a hard or soft leadership style, it can be assumed that the characteristics and expectations of the different types of the target audiences had some influence on the decision, depending on the national or local level of the public position held. Local audiences mainly demand displays of managerial skills, which reinforces the need to convey a leadership style marked by hard skills.

As for the degree of personalisation in the communication of their government actions, those women leaders who leant towards a hard leadership style did not show a tendency to personalise; while those who were characterised by communicating a soft leadership style tended to personalise when relaying government measures and decisions. Instagram's rationale favours the phenomenon of the personalisation of politics, promotes the use of anecdotes and narrative techniques such as storytelling or “storydoing” for the communication of political government actions and encourages the posting of simple visual messages in the form of photos, albums, selfies, videos and collages that seek to humanise and give prominence to high-profile politicians (Berrocal-Gonzalo et al., 2022; Ekman & Widholm, 2017). However, according to the results of this analysis, the trend toward personalisation of communication is not homogeneous. In fact, the different techniques chosen to communicate government messages and actions, as well as the varying degree of personalisation of communication, was found to be strongly related to the leadership style of the women politicians in the sample.

Regarding the role of Instagram and other social networks such as TikTok as a political socialisation tool for young people, it would be interesting to design a study to investigate whether this segment of the population tends to prefer the softer leadership styles communicated by political leaders who are closer in age to them, or whether young people have a greater preference for leadership styles. Women leaders from three different generations are represented in the sample analysed, Sanna Marin being the only millennial. Her Instagram profile exhibits a high degree of personalisation. Her personal image is often associated both with the communication used in her representative and governmental tasks and with aspects of her private life.

The study shows that not all women politicians have the same style when it comes to leading a government project. The communication of both male and female leaders in the context of the COVID-19 pandemic was heterogeneous (Palau-Sampio & Carratalá, 2022). The leadership styles communicated by the women leaders analysed varied according to their generational affiliation in the type of (hard or soft) skills they chose to convey, as well as in the degree of personalisation and the techniques they prefer to use when disseminating their government’s decisions and actions. The assumption that women politicians share a gender-defined leadership style contributes to a new stereotype that denies women's diversity. This is dangerous because it limits their access to, threatens their permanence in, and justifies their exclusion from positions of power and decision-making spaces, and therefore restricts their political influence. It is prejudicial to consider that this style does not meet what is needed to successfully manage and lead these positions and processes. This is an obstacle that can take a heavy toll on women politicians' career development, leadership and equal political participation. Future work will need to broaden the discussion by incorporating new variables that serve to explore whether there are any differences in leadership styles based on the geographical region involved, the degree of polarisation used in women leaders' communication, or on the strategies they use to express their emotions publicly. (1)