Leadership communication and public trust in government during crisis: do gender traits make a difference?

(1)

Leadership communication and public

trust in government during crisis: do

gender traits make a difference?

Isabel Harmelink (S2682605)

Msc. Thesis Crisis and Security Management

Faculty of Governance and Global Affairs

Leiden University

First reader: dr. Sanneke Kuipers

Second reader: dr. Wouter Jong

Word count: 19509

(2)

Abstract

The research in this thesis will investigate which different leadership communication styles and traits there are, whether they are connected to gender, and to what extent they influence public trust in government during a crisis. More specifically, the thesis answers the following research question: “How do male and female leadership communication styles affect trust in government during the COVID-19 pandemic?”. The relevance for this is the importance of leadership and communication in times of crisis and whether these have explanatory power for influencing public trust. The thesis starts by investigating expectations about leadership communication connected to gender. Then it goes on to exploring female and male styles in leadership communication and additional existing feminine and masculine traits. This is applied to speeches of prime ministers in four different countries in which a comparison can be made between two smaller and two larger countries. The research found that gendered leadership communication does influence trust in government. Even more, when controlling for infection rates in the levels of trust, there still was a difference in the outcome of trust, which is attributed to the actual communication in practice.

(3)

1. Introduction

The infectious disease COVID-19 is caused by a recently discovered coronavirus. The outbreak of this virus began in Wuhan, China, in December 2019. The outbreak quickly unfolded into a global pandemic, affecting many countries around the world. The virus developing and spreading around the world very quickly called for on the spot crisis management of world leaders. They were responsible for preparing for and dealing with health issues during this worldwide pandemic. But how do you respond and communicate about such a great and potentially disastrous situation? And how does this affect citizen’s trust? Crisis management requires leadership and crisis communication, however, the ways in which a leader tries to manage the crisis and communicate crisis messages can influence the public’s trust in your leadership. For leaders it is important to have public trust during a crisis in order to implement policies and measures that the public will follow. As one can see and conclude from the starting point until now, leaders have approached the COVID-19 pandemic in very differently ways, with various forms of communication. As some media pointed out (Ringmar, 2020, p. 13; Wittenberg-Cox, 2020), during the COVID-19 pandemic, many of the honourable leaders in the crisis are women. Therefore, one can ask whether there are differences between women and men when comparing their styles of leadership communication. Hereby, one can examine whether there is a pattern in the style that females or males have in communicating in crises which may affect the trust the public has in them. Accordingly, in this thesis, the research question is as following: ‘How do male and female leadership communication styles affect trust

in government during the COVID-19 pandemic?’

I. Research design and objective

As Yukl (2012, p. 66) notes, the nature of leadership is influencing and facilitating individual as well as collective efforts to accomplish shared objectives. Performance can be improved by altering and determining processes. However, when confronted with a crisis, political leadership has a crucial role in crisis management (Boin, Hart, McConnell, & Preston, 2010, p. 706). Leaders and governments are responsible for protecting and restoring public order, public safety and health (Boin et al., 2010, p. 707). Crisis situations put a challenge to this responsibility because of weakened order. On the other hand, a crisis creates possibilities for changing the status quo, due to changes in the current order.

(6)

6 Related to leadership, crisis communication is a crucial factor during crises. Crisis management is about communicating political decisions to people. The public needs to be informed about the situation, but they should not be addressed by panic and disinformation. Therefore, it is decisive to collect information to diminish the impacts of the crisis, as well as, informing the public and stakeholders about the situation and influencing their perceptions.

Then, leadership and leadership communication influence trust. Leaders have a key role in crisis management, but the extent of trust the public has in them influences their leadership and the outcomes of their management. Leaders can employ different communication styles, which can lead to both more or less trust from the public in their leader’s ideas and ability to manage.

The current ongoing COVID-19 pandemic offers comparable cases for examining leadership and crisis communication in practice and to see the effects on trust in government. The COVID-19 situation rapidly emerged into a global pandemic, making many countries subject to the crisis. Every leader has his or her own leadership style and approaches issues in different ways. In many cases, female leaders seem to be the ones having strategies that work and providing reasonable communication in this crisis. The COVID-19 pandemic is an interesting case study in which to explore different leadership communication styles and to see the effects on trust in government in facing this crisis. All countries are namely facing the same crisis, but the approaches and communication differ and therefore public trust in the government may differ.

The objective of this study is to explore which different leadership crisis communication styles there are, whether one can find patterns when applying these styles to specific cases and what the influence of these styles is on the public’s trust in their government. This research matters because when leadership and crisis communication behaviours have a specific explanatory power, one can see the influence on public trust in government. The reason for doing this research is to see what different gendered leadership communication styles there are and to explore which seem styles to work. By doing this, one can see whether female leaders actually do better in managing this COVID-19 crisis.

II. Scientific and social relevance

In the academic debate, there is an ongoing discussion about whether women are more caring and compassionate compared to men (Ringmar, 2020, p. 13). Traits such as care and love are generally associated with women, (Ringmar, 2020, p. 13) therefore, one can wonder whether these collide with leadership and crisis communication. By including gender in this study, one

(7)

7 can see whether different leadership communication styles exist between men and women and if one of the two has specific traits that are helpful and provide for good leadership and crisis communication. Furthermore, this research goes into the trust of the public in the government which is influenced by the style the leader is portraying. This adds to the academic debate in exploring what leadership traits might work for getting more public trust.

Practically, this study is relevant in exploring which kinds of leadership traits are helpful in assessing and managing a crisis situation. Besides leadership, crisis communication is a crucial focus of the research, which goes into creating a collective understanding, giving a frame, and informing the public about the crisis and its dynamics. In examining leadership communication styles, one can see whether there are differences between the genders and consider having more females at the top because of their approaches. The social value for organisations, but also society itself, is to see which leadership and crisis communication traits are useful in a crisis situation. Besides this, trust is essential for a leader to stay in a leadership position, but also to execute policies that the public will follow. Therefore, the practical value lies in, both, examining leadership styles that work and that increase public trust in government and the leader.

III. Research outline

The following chapter will lay out the body of knowledge on leadership, crisis communication and trust in government. It will address the present gap in knowledge and the further value of the research. After that, the research design and methodology will be discussed in order to set a framework and give justification for doing the research. This includes the research design and the selection of cases, as well as the operationalization and the limitations. In chapter four, the analysis and outcome of the research is examined, which provides for an accurate overview of the findings. Lastly, the conclusion will give a final answer to the research question asked before. Furthermore, the conclusion provides room for discussion of the findings, the limitations of the research and opportunities for future research.

(8)

8

2. Body of knowledge

This chapter starts by exploring the body of knowledge on leadership and crisis management. Then, it goes into crisis communication, as this is a key aspect of leadership, after which different styles of leadership communication are set out. Lastly, the chapter examines theories on the outcome, which is the public trust in government. By doing this, the foundation of the research will be set up and gaps in the body of knowledge, that provide for research opportunities, are explored.

I. Leadership in times of crisis

Managing crises and preventing, preparing, and responding to them is a hard but critical task for leaders. In crisis situations, there is always chaos, huge pressure, stress and incomplete information, making it very difficult to get to proper decision-making (Boin & Hart, 2003, p. 545). Phills (2005) asks the question whether leadership actually matters and how it matters. He proposes that for leadership to matter, one has to believe that leaders can and do influence the performance of organizations and social systems, and that this influence is intentional and rational (Phills, 2005, p. 47). Leaders can vary in their mechanisms of action, the processes to influence performance, which will have different outcomes for the situation. The portrayal of leadership is a crucial factor to the survival of a crisis for a country (James & Wooten, 2005, p. 141). Besides this, the public realm is an important factor, because people believe that the state protects them and their goods in times of crises. Thus, a gap exists between citizens’ expectations of their leaders and the actual leadership efforts in crisis management conducted.

Boin and Hart (2003, pp. 546-548) present six public expectations leaders are often unable to meet, such as preparing for worst-case scenarios and striving to learn from crises. Leadership in times of crisis faces challenges, because leaders are forced to operate in full public view with opportunities for criticism (James & Wooten, 2005, p. 142). Furthermore, decisions are often suboptimal to make the crisis go away (James & Wooten, 2005, p. 142), resulting in undermining leadership.

II. Communication is key

During crises, information and communication are crucial tools for leaders to use. Especially in the current digital age and collided with technological development, it becomes more and

(9)

9 more important that leaders address the demands and opportunities of digital media (Hayashi & Soo, 2012, p. 82). When there are information gaps, or information is misrepresented, it becomes harder for leaders to go against rumours in the public sphere. Further, information gaps or misrepresentation damages people’s trust and integrity in their leader.

Even more, having a valuable reputation is important for leaders, because this is connected to effective leadership and legitimacy. Crisis situations pose a risk to a leader’s or an organization’s reputation. This risk has to do with what Boin et al. (2010) present as the politics of crisis management: inquiry, accountability and blame games. The aftermath of a crisis calls for responsibility questions upon leaders. First of all, leaders have to face public inquiries, because they are subject to investigation upon their crisis approach. Secondly, leaders have to deal with criticism from the public on their performance (Boin et al., 2010, p. 709). Hereby leaders face a dilemma of whether they take responsibility for their actions and faults or whether they deny it and blame others for their mistakes. Both choices will have an influence on the leader’s legitimacy. Thirdly, leaders have to cope with political verdicts as inquiry reports can be very critical of a government’s conduct during crisis.

Boin, Kuipers, and Overdijk (2013) present a framework for assessing leadership in times of crises. In this framework, the authors present ten executive tasks on which one can assess the leadership in times of a crisis. For the sake of this research and subject, the focus will be on the following tasks: sensemaking, making critical decisions, meaning making, and communication. First of all, sensemaking is about making sense of the crisis situation and creating a collective understanding of the environment and its dynamics. This step is important for making informed decisions under uncertainty (Boin et al., 2013, p. 82). Weick (1993, p. 635) describes sensemaking as seeing reality as an ongoing accomplishment emerging from efforts to create order and making sense of what is occurring. Sensemaking requires that people make a situation rationally accountable to themselves and others to create an image and understanding of the situation (Weick, 1993). After sensemaking, leaders need to make critical decisions. Here, deliberation is key as well as a due process of decision-making, instead of only choosing a quick technical solution (Boin et al., 2013, p. 83).

Then, the next important task is meaning making, in which leaders interpret the crisis situation in understandable ways for their people and to create a plan to restore the order. Meaning making is about presenting a frame of the situation to people in which they can interpret events (Boin et al., 2013, p. 85). This is important for the way in which the public views their leader, because the leader presents them a frame of reference (Boin et al., 2013, p. 85). Park (2016, p.

(10)

10 1235) discusses the meaning-making model, focussing on two levels of meaning: global and situational. Global meaning is about people’s fundamental beliefs about themselves and the world, whereas situational meaning explains how global meaning together with the context of a situation influences one’s interpretation and reactions to that situation (Park, 2016, p. 1235). Meaning making is not only visible through words, but also through actions as leaders are to give an understanding of what is going on, reduce uncertainty and give hope with their messages and actions (Jong, 2017, p. 1026). Moreover, meaning making is about creating frames and the beliefs in these frames are decisive for leader’s legitimacy. Connected to the meaning making is communication. In crisis situations, good communication is crucial because of high uncertainty and missing or overloaded information streams. Leaders need to ensure that people are timely informed about decisions and the situation. When communication is effective, this can be a prerequisite for good public relations (Nyenswah, Engineer, & Peters, 2016, p. 201). Standard communication routines will probably not work during a crisis, because of the complex and fast dynamics presenting difficulties in verifying information (Boin et al., 2013, p. 85). Here for, communication techniques need to be revised and altered to fit the crisis situation, this is where crisis communication comes into play.

III. Crisis communication

The politics of crisis management is mostly about communication (Boin & Hart, 2003), therefore, crisis communication is crucial, because in order to manage a crisis, one is in need for information. Coombs and Holladay (2011, p. 20) define crisis communication as the collection, processing, and dissemination of information required to address a crisis situation. In communicating during a crisis, political leaders effectively compete in political framing contests (Salomonsen & Hart, 2020, p. 441). In this framing, the event context, about what is happening, is equally important to the political context, being about everything else happening in the political process (Salomonsen & Hart, 2020, p. 441). Besides these two, with their intentions and skills, the different actors will read the contexts in different ways and will guide their communication accordingly (Salomonsen & Hart, 2020, p. 441).

During the pre-crisis, the situation involving efforts to prevent crises and to prepare for crisis management, crisis communication is about collecting information regarding risks and making decisions about managing potential risks (Coombs & Holladay, 2011, p. 20). In a crisis, communication revolves around collecting and processing information that help crisis teams in

(11)

11 the decision-making process, as well as, creating crisis messages to people. Post-crisis, the crisis communication is about investigating the crisis management effort and coming up with additional messages when necessary (Coombs & Holladay, 2011, p. 20).

Whereas this thesis is about finding out how leadership communication styles connect to public trust in government. The main focus will be on the content of the communication of leaders during a crisis because leaders will respond and communicate in order to reduce and contain harm in the dynamic and unpredictable environment of a crisis.

IV. Leadership communication styles

a. Expectations about leadership

After setting out the main theories on leadership and crisis communication, this part will explore the expectations of the public about leadership. Generally, leadership has predominantly been associated with men (Hoyt & Burnette, 2013, p. 4). These stereotypical beliefs of gender come from descriptive beliefs, which are about the typical characteristics of men and women, and prescriptive beliefs, which are about the ideal characteristics (Vinkenburg, Engen, Eagly, & Johannesen-Schmidt, 2011, p. 10). Eagly and Karau (2002) discuss the notion of gender role in explaining people’s beliefs about women and men. Gender roles are in this instance the beliefs about the specific characteristics of men and women, of which most beliefs are normative, describing desirable qualities or behaviours for each gender (Eagly & Karau, 2002, p. 574). These gender roles do therefore include the descriptive and prescriptive beliefs on characteristics of men and women. Further, the argument is that sex differences and similarities in behaviour reflect certain gender role beliefs which represent people’s perceptions of the social roles of men and women in societies they live in (Eagly & Wood, 2016, p. 459).

These stereotypes and beliefs of leaders being male are connected to the idea of a glass ceiling, which is a barrier of prejudice and discrimination excluding women from higher leadership positions (Eagly & Karau, 2002, p. 573). Gender roles, such as the descriptive and prescriptive beliefs as described by Vinkenburg et al. (2011) before, lead to prejudice. This prejudice can arise from relations or roles perceived by people and the requirements of the social roles that people should occupy (Eagly & Karau, 2002, p. 574). Role congruity theory is about the treatment of the content of gender roles and their importance in promoting sex differences in behaviour (Eagly & Karau, 2002, p. 575). Prejudice about female leadership follow from incongruity between the role of women and the image that people generally have about a leader

(12)

12 and leadership. Men possess traits that match leadership roles and leader prototypes are culturally masculine (Hoyt & Burnette, 2013).

For leaders it is very important that people trust them and their government, as this provides legitimacy and legitimization for their actions and policies. Trust in this instance is about citizens’ expectations of the government, how this government should operate and interact with social and economic institutions and citizenry, and the behaviour of public leaders, civil servants, and citizens (Cheema & Popovski, 2010, p. 4). Trust is dependent upon expectations and to what extent these expectations are met. When citizens set high expectations or when citizens are biased, it is usually more complicated for leaders and governments to meet these expectations, which results in a decline in confidence (Cheema & Popovski, 2010, p. 4). One of the determinants of trust in government is the commitment and inspiration of political leaders (Cheema & Popovski, 2010, p. 6). Trust in government refers to the extent to which the public has confidence in their government doing the ‘right thing’, to act appropriately and honestly on behalf of the public (Beshi & Kaur, 2020, p. 340). Trust is about the quality of the relationship between the public and their government.

An important factor in determining public trust is accountability (Bakker, Bommel, Kerstholt, & Giebels, 2018, p. 278). The extent to which the political leaders are or can be held responsible for the crisis and its consequences is influential in the extent of trust (Meer, Steenvoorden, & Ouattara, 2020, p. 3). Even with a natural disaster, such as Hurricane Katrina in 2005, governments can be held responsible because they are for example badly prepared for such a disaster (Meer et al., 2020, p. 3). When leaders or governments are held responsible, they will not see an increase in support for them. Besides accountability, the information received during a crisis influences people’s behaviour and opinions about the situation (Bakker et al., 2018, p. 278).

This part is about expectations about leadership and bias, for which one can consider the index Hofstede (n.d.) came up with. The Hofstede Index is a model consisting of six basic issues around which a society needs to organize itself. He calls this dimensions of culture, in which culture is ‘the collective programming of the mind that distinguishes the members of one group or category of people form others’ (Hofstede, Hofstede, & Minkov, 2010, p. 6). The dimensions are: Power Distance, Uncertainty Avoidance, Individualism versus Collectivism, Masculinity versus Femininity, Long Term versus Short Term Orientation, and Indulgence versus Restraint (Hofstede, n.d.). For leadership communication, the dimension of Masculinity versus Femininity is interesting, in order to see the division of emotional roles between women and

(13)

13 men (Hofstede, 2011, p. 8). Assertiveness and competitiveness are seen as masculine values, while feminine values are more modest and caring (Hofstede, 2011, p. 12). The masculine pole is focused on earnings, recognition, advancement and challenge, while the feminine pole focuses on managing relationships, cooperation, living area and employment security (Hofstede et al., 2010, p. 139). The Hofstede index on the Masculinity versus Femininity dimension will illustrate whether a society represents a preference for achievement, assertiveness and material reward for masculinity, or a preference for cooperation, modesty and caring as being feminine (Hofstede Insights, 2020).

Harris (2011, p. 11) describes the way that speeches about disasters reinforce intersectional hierarchies of difference for legitimizing gender-based violence. She finds that in all the speeches from leaders in the United States about Hurricane Katrina, masculinity has the upper tone. Thereby, men are found to be the agents, while women are presented as being relying on men’s agency (Harris, 2011, p. 16). In the speeches, men are seen as the ones who can overcome the storm, while women need men to provide for and protect them in order to overcome the storm (Harris, 2011, p. 17). All leaders tried to show strong language and masculinity by focussing on tasks and portraying confidence and dominance.

Taking all together, the following propositions regarding expectations about leadership will be examined in the analysis:

1. A bias exists in leadership expectations which leads to people trusting male leaders more than female leaders.

2. The effect of the expected bias in trust in male leaders is more profound in the, according to the Hofstede Index, masculine societies when compared to feminine societies.

b. Transformational and transactional leadership communication styles

Burke and Collins (2001) discuss the two main leadership communication styles, namely transformational and transactional leadership. Transformational refers to leaders that encourage their employees to focus on the interests of the whole group and that want to strengthen organizational performance (Burke & Collins, 2001, p. 245). Transformational leaders are charismatic, they inspire employees in their motivation, and they may go into individualized consideration upon emotional needs. Transactional leadership, on the other hand, is about setting standards that need to be met and giving out rewards when performance is positive (Burke & Collins, 2001, p. 245). Transactional leadership is about trying to get people to

(14)

14 comply by using rewards and punishments and by keeping things the same instead of trying to change the future, as is the case with transformational leadership (Odumeru & Ogbonna, 2013, p. 358). Transformational leadership is proactive, while transactional leadership styles are primarily passive and are about maintaining the status quo (Odumeru & Ogbonna, 2013, p. 359). Findings present that females are more likely, when compared to males, to have transformational behaviour (Burke & Collins, 2001, p. 250). This means that they are more likely serving as positive role models and encourage their employees to be creative and to go for a collective group interest. Besides this, the skills communication, coaching and developing, and time management were found to be more linked to women, which correlates with the finding that women present more transformational leadership characteristics (Burke & Collins, 2001, p. 252).

Then, trust in government is closely linked to the approval of the current leader in office (Citrin & Green, 1986, p. 444). The public expects their leader to set the national agenda and to respond to the demanding interests in society (Citrin & Green, 1986, p. 444). In their research, Citrin and Green (1986, p. 444) propose that when there is widespread respect and appreciation for the president’s personal qualities, this should boost political trust in the leader and its government. Thus, the persona of the leader influences the degree of confidence in the government. For the public, it makes sense to what extent they view their leader as being moral, knowledgeable, inspiring, strong on the one side, or, contrastingly, power-hungry and dishonest (Citrin & Green, 1986, p. 445). The positive traits of the leader are connected to the view on transformational leadership, as discussed by Burke and Collins (2001) before. Transformational leadership communication is about being proactive, inspiring people to motivate them and to be charismatic and a role model (Burke & Collins, 2001, p. 245). Research found that female leaders were more likely to present transformational behaviours (Burke & Collins, 2001, p. 250) and thus these characteristics are connected to the feminine leadership communication style. Public trust can increase when the leader is seen as a strong, inspiring, and moral leader and therefore, transformational, feminine leadership, can influence the level of public trust in the government in a positive way.

For this part of the research, the following propositions are set up:

3. The proactive transformational leadership communication style dominates in crisis communication by female leaders, while the reactive transactional leadership communication style dominates in crisis communication by male leaders.

(15)

15 c. Female and male leadership communication traits

This part goes further into the traits or characteristics that can be connected to male or female leadership communication. Following a Gallup poll in America, people associate men as being aggressive and courageous, whereas women are associated with the characteristics of being emotional, affectionate, talkative, patient and creative (Newport, 2001). Hence, women are seen as having communal characteristics, highlighting a concern for other people, while men have agentic characteristics that are about confidence and dominance (Hoyt & Burnette, 2013, p. 5). Prejudice towards female leaders thus follows from the incongruity that people perceive between the characteristics of women and the requirements of leader roles (Eagly & Karau, 2002, p. 574).

Eagly and Johnson (1990, p. 236) explore leadership traits and make a distinction between task-oriented, focused on task accomplishment, and interpersonal-task-oriented, focused on relationships, leadership traits. These characteristics are connected to gender because of stereotypes by proposing the argument that men are more self-assertive, independent and dominant, while women are more selfless, helpful, understanding and concerned with others (Eagly & Johnson, 1990, p. 236). Women are therefore characterized with the interpersonal-oriented traits, while men are associated with the task-oriented leadership characteristics.

Men and women have different styles of communicating, which is mainly because of the way that they see the purpose of the conversation (Merchant, 2012, p. 17). Where women use conversations to strengthen social connections and make relationships with people, men use communication to wield dominance and achieve outcomes (Merchant, 2012, p. 17). Both genders come from different cultures of communication, as women are for example more likely to offer sympathy and help to another, whereas men are more likely to avoid problems and resent to receiving another’s help (Basow & Rubenfeld, 2003, pp. 183-184). A study by Basow and Rubenfeld (2003, p. 187) found that strong nurturant/expressive traits, which are associated with feminine traits, are linked to a high possibility of offering sympathy to another and to giving advice, linking with the interpersonal-orientation style. The research further found that traits strongly associated with instrumental and active traits, connected to masculine behaviour, were linked to the lower likelihood of giving sympathy in response to problems of a friend (Basow & Rubenfeld, 2003, p. 187).

Further, the study of Eagly and Johnson (1990) focused on the distinction between democratic versus autocratic leadership, thus, participative or directive leadership. The findings conclude that women have a more democratic style of leading and communicating, while men adopt a

(16)

16 more directive style of leadership (Eagly & Johnson, 1990, p. 247). Hoyt (2010, p. 487) describes the more democratic leadership style of women as well, but states that this style might not always be the most effective style. Women and men differ in communication regarding solving a crisis or coming to a mutual conclusion, namely that men follow their natural tendency of offering a solution, while women seek empathy and are inclined to offer unsolicited advice (Merchant, 2012, p. 20).

There are many variables identified in various research on the difference between male and female speech and communication (Zahn, 1989, p. 60). Females are identified with using clear, gentle and emotional speech, they talk a lot, use good grammar and are polite, while men are described by using more swear words, having dominant speech, being straight to the point and being forceful and aggressive in their speech (Zahn, 1989, p. 60). Female and male communication styles are seemingly different and are therefore labelled as ‘genderlect’ (Maiorescu, 2016, p. 557). The female communication style comes from the willingness to bond and create relationships, therefore, the communication is symmetrical (Maiorescu, 2016, p. 557). This while the male communication style is about gaining status and respect, which creates and asymmetrical communication line.

Maiorescu (2016, p. 556) applied this genderlect theory to analyse two executive crisis communication strategies of one male and one female senior executive and tried to explain this by male vs female-specific communication styles. The study found that in the case of the female senior executive, the themes of exigency and regret were most predominant, which denote a female communication style (Maiorescu, 2016, p. 559). Further, she used a transparent style which is about to maintaining trust, linking to managing the relationship, which is a female attribute. The theme of regret is also a female-specific communication aspect, as women care about others and how decisions affect other people and thus describes the relationship-oriented style (Maiorescu, 2016, p. 559). Looking at the male senior executive, the predominant themes in communication were defence and future action. Using defensive strategies in communication is about facing direct conflict which implies strong communication, describing the male communication style (Maiorescu, 2016, p. 561). Men specifically use defensive communication strategies when it is about protecting status. Besides this, the theme of future action is related to seeing justice as an abstract concept, instead of females which see justice as part of relationships (Maiorescu, 2016, p. 561). Men make decisions based on hard facts and not about relationships or those impacted by a relationship, therefore, the future orientation is about

(17)

17 achieving justice, but in abstract ways. Here for, differences in communication styles can be explained as being specific female or male communication styles.

Mueller (1970) analyses various variables influencing Presidential popularity and discusses the ‘rally round the flag’ variable. This variable seeks to explain that certain extreme international events, such as a great crisis, generate a ‘rally round the flag’ effect that gives a boost to the popularity rating of a president (Mueller, 1970, p. 21). In the light of this theory, people rally and stay behind their executive in power during a crisis (Mueller, 1970, p. 21). The rally effect characterizes a sudden but strong increase in the public’s approval of the president that occurs in response to a kind of dramatic international event (Hetherington & Nelson, 2003, p. 37). During a rally event, the trust in government increases because the public evaluates their government on crisis-induced criteria, while after a crisis, the trust decreases as the public returns to their usual criteria of evaluating their government (Hetherington & Nelson, 2003, p. 40). When a rally exists, the situation is favourable for policy makers making policies to react to the crisis, as the increased trust will lead to higher support for their policies created in response to the crisis (Meer et al., 2020, p. 3). Therefore, in times of crises, public trust in government can increase because people rally behind their leader in uncertain times in order to find unity and sympathy. Here for, the expectation is that leaders that have a more feminine style of leadership communication will have more trust in government.

To sum up, female leaders will display an orientation towards their relationships, focussing on empathy and a concern for other people. Furthermore, they will present more participative and symmetrical traits, relating to the willingness to bond. On the other hand, the male leadership communication traits will be more focused on accomplishing tasks and directing the public. Male leaders will communicate more asymmetrically, while they want to achieve status and gain respect from the people and are therefore more forceful, aggressive, and courageous. Therefore, the following proposition follows from this part:

5. Specifically, for a public health crisis such as COVID-19, female leadership communication qualities will positively influence public trust.

(18)

18

3. Method

This chapter will go into the methodological part of the research. First of all, the design of the study is addressed, which is a comparative case study. Afterwards, the case selection is explained and justified. Then, the data collection and time frame of the research will be discussed. Finally, the chapter will go into the operationalization, in order to create measurements that go with the expectations drawn from theory and that can be used for the analysis in the next chapter. Further, the operationalization part sets out the measurements on trust in government in order to do analysis on the influence of the leadership communication styles.

I. Comparative case study

The design for this research is a comparative case study, as it will examine certain factors in different cases and compare these to each other. A case study allows for the exploration of new processes or behaviours as detailed observations can be made that give a holistic view of the process (Meyer, 2001, p. 330). When doing a case study, it is important to make choices about the design, such as the selection of cases, the timeframe and the data collection procedure (Meyer, 2001, p. 332). One can choose to do either one case or multiple cases, where choosing only one case limits the generalizability and has biases regarding the processing of information (Meyer, 2001, p. 332). Comparative studies allow for research into similarities and differences between the cases.

When doing a comparative case study, one can apply a most similar or a most different method. The most similar method chooses cases that are similar on all the independent variables, except the independent variable of interest (Seawright & Gerring, 2008, p. 304). On the other hand, the most different method is the reverse. The method includes cases that are most different, therefore, cases included will be ones where just one independent variable as well as the dependent variable covary (Seawright & Gerring, 2008, p. 306). All the other variables will show different values, making the cases similar on the variable of interest and the outcome, the dependent variable (Seawright & Gerring, 2008, p. 306). This research will have a most similar design because the cases will be selected upon comparable political, administrative, and cultural structures. The independent variable that will vary and which will be subject for the comparison will be the gender of the leader of the country and the outcome in terms of trust.

(19)

19 By doing a comparative case study, one can allow for a higher generalizability compared to doing a single case study. The internal validity is often high, as the context is understood thoroughly. However, case studies have a low external validity, as, even with comparative case studies, the generalizability is limited. This is because of the specific number of cases that are compared. Further, case studies have a low reliability because they allow for much interpretation in the research. Besides this, there are cognitive limitations and there can be observation bias, as there are limits to the processing of information and researchers often work alone.

II. Case selection

For this research, four cases have been selected to explore leadership communication styles during the COVID-19 pandemic. In order to juxtapose and compare between the styles of the leaders, two smaller and two larger countries are selected for comparison. The two smaller cases are Denmark and the Netherlands, Denmark having a female leader and the Netherlands a male one. The two larger countries in this comparison are Germany and the United Kingdom (UK), in which Germany has the female leader and the United Kingdom a male leader. These countries will thus be put into a comparative case study with a most similar design. The varying variable is the gender of the leader of the country, thus, when comparing Denmark with the Netherlands and Germany with the United Kingdom, the difference will be on the variable of the leader’s gender. The justification for these cases can be found in comparison on the following factors: political, administrative, and cultural structures. These will be explained in the following section.

a. Political structure

Political structure relates to the political system of a country, whether a country is a democracy, republic, monarchy, or an authoritarian state. Countries can have a presidential system, in which the head of the executive, the president, is directly elected. Or, they can have a parliamentary system, in which the head of the executive is not directly elected but stays in office under confidence and tolerance of the parliament, that is directly elected (Heringa & Kiiver, 2012). A semi-presidential system includes both, a directly elected president, but also a prime minister that holds executive powers and is accountable to parliament (Heringa & Kiiver, 2012).

Following the data from Our World in Data (2019), all the cases included in this research can be described as liberal democracies. In a democracy, institutions allow citizens to express their

(20)

20 political preferences (Our World in Data, 2019). Further, there are constraints on executive power and civil liberties are guaranteed. A liberal democracy, the representative democracy, as just described, operates under principles of liberalism, meaning that there are free and fair elections, a separation of powers and a market economy.

Resulting from the CIA (2020) database on government type, Denmark and the Netherlands both are a parliamentary constitutional monarchy. Then Germany is a federal parliamentary republic and the United Kingdom is a parliamentary constitutional monarchy (CIA, 2020). All countries in this comparison have a parliamentary system in which the parliament is elected and makes the decisions in the country.

b. Administrative structure

Then, administrative structure is about the state capacity and power in the country. State capacity is about the extent to which a state is able to implement a range of policies (Besley & Persson, 2009, p. 2). Christensen, Laegreid, and Rykkja (2016, p. 888) define governance capacity as including formal structural and procedural features of the governmental administrative apparatus, but also informal elements about how the features work in practice.

The World Bank Worldwide Governance Indicators (2020) Dataset ranks countries on six criteria of governance, being: voice and accountability, political stability, government effectiveness, regulatory quality, rule of law, and control of corruption. In this dataset, government effectiveness measures the competence of the bureaucracy and the public service delivery (Worldwide Governance Indicators, 2020). The indicator measures for instance, red tape, quality of public schools, government stability, public consistency, basic infrastructure and bureaucrats’ expertise (Walle, 2007, p. 438). The indicator scale ranges from -2,5 till +2,5 in which higher scores refer to better governance. Comparing the smaller countries, Denmark has a score of 1,9 of government effectiveness, where the Netherlands scores 1,8 (Worldwide Governance Indicators, 2020), thus the countries score almost the same on government effectiveness. When comparing Germany to the United Kingdom, Germany scores 1,6 and the UK scores 1,4 on government effectiveness, making the countries comparable. All the scores are quite close in the same range, meaning that all the countries compared have quite similar levels of government effectiveness, which all range largely high.

Then, trust in government before COVID-19 is an important factor that serves as a base to which can be linked back in the analysis part. For measuring trust in government before COVID-19, this research will use results of the Standard Eurobarometer 92 of Autumn 2019.

(21)

21 The Eurobarometer measures public opinion in the European Union and regarding public trust asks whether people tend to trust or not to trust their national government (Eurobarometer, 2019). In the Barometer of November 2019, the results are as following: In Denmark 63% tends to trust their government, while 32% tends not to trust, whereas in the Netherlands 59% tends to trust the Dutch government and 37% tends not to trust (Eurobarometer, 2019). In Germany, 50% tends to trust their government, while 45% tends not to trust, while in the United Kingdom, only 21% tends to trust the UK government and 72% tends not to trust their government (Eurobarometer, 2019).

c. Cultural structure

In order to investigate the cultural dimension for comparison, the Masculinity versus Femininity dimension of the Hofstede Index is used. The Masculinity versus Femininity dimension is about what motivates people, whether they want to be the best (masculine) or whether they like what they do (feminine) (Hofstede Insights, 2020).

Regarding Denmark and the Netherlands, Denmark scores 16 points on the Masculinity versus Femininity dimension, whereas the Netherlands scores 14 points on this dimension (Hofstede Insights, 2020). This means that both societies can be characterized as Feminine. This implies that the societies are focused on inclusion and keeping a work/life balance (Hofstede Insights, 2020). Furthermore, the leader will be supportive of people and will involve them when making decisions. Conflicts are usually solved by compromise instead of competition. Inclusion of people is therefore important, in which leaders are supportive and compromising. Society is less about being the best or wanting to be the best and achieving successes.

Then, by considering Germany and the United Kingdom, both countries score 66 points on the Masculinity versus Femininity dimension (Hofstede Insights, 2020). With this result, both societies can be characterized as Masculine, meaning that the fundamental that motivates people is wanting to be the best (Hofstede Insights, 2020). Furthermore, the German and British society places a great value on performance and status is an important factor. Besides this, managers are usually decisive instead of compromising.

Overall, looking at the comparisons of the cases on political, administrative, and cultural structures, they allow for comparison in a most similar design. For both, the comparison for the smaller and for the larger countries, the data on the structures allow for comparison. Regarding political structures, all cases have a parliamentary structure, most of them in a monarchy, while only Germany has a federal form. Looking at administrative structures, all the countries rank

(22)

22 as having good government effectiveness, making them comparable on administrative features. Culturally, Denmark and the Netherlands can both be considered feminine countries, whereas Germany and the United Kingdom can both be characterized as masculine. Here for, as the countries are comparable on the different structures, this allows for a most similar research design in which the gender of the leader will be the variable that will be different among the countries, leading to a varying outcome in public trust. Table 1 below presents the research design matrix, that provides an oversight of the comparisons in this research.

TABLE 1: RESEARCH DESIGN MATRIX Denmark

• Small country • Feminine society • Female prime minister

The Netherlands • Small country • Feminine society • Male prime minister Germany

• Large country • Masculine society • Female prime minister

The United Kingdom • Large country • Masculine society • Male prime minister

III. Infection rates

Besides the factors mentioned in the previous part, infection rates are an important factor in determining public trust. Infection rates influence the public trust in government and therefore, there is a need to control for the change in infection rates. In this way it is possible to see whether changes in trust are influenced by communication or whether changes are due to infection rate change. The analysis part will therefore also go into the changes in infection rates for the cases at different points in time, in order to see whether they can account for a possible change in trust. For data on infection rates, the country websites of the World Health Organization (WHO) are used because they contain the current numbers on infections and also show the infection rates over time (World Health Organization, 2020).

IV. Time frame and data collection

For the time frame, for all the countries the response phase will be taken. This starts with the first COVID19 -infected person in the country, until the first peak of the virus is over. For every country, this time frame will be somewhat different, so the specifics will be as following:

(23)

23 TABLE 2: TIME FRAME OF COVID-19 PER CASE

Country First COVID-19 case End of first peak Denmark 27 February 2020 End of April 2020 The Netherlands 27 February 2020 End of May 2020 Germany 28 January 2020 Start of May 2020 The United Kingdom 31 January 2020 End of May 2020

The data collection is done by collecting online speeches of the leaders regarding the COVID-19 virus in their country. The following table gives an overview of the number of speeches per country, gives the length of the speech transcripts and numbers them for usage in the analysis.

TABLE 3: PRIME MINISTER’S SPEECH TRANSCRIPS ON COVID-19 PER COUNTRY

Country Date Speech code and pages

Denmark 11 March 2020 30 March 2020 6 April 2020 DK1 – 5 pages DK2 – 6 pages DK3 – 6 pages The Netherlands 9 March 2020

12 March 2020 19 March 2020 23 March 2020 31 March 2020 7 April 2020 15 April 2020 21 April 2020 6 May 2020 19 May 2020 27 May 2020 NL1 – 2 pages NL2 – 3 pages NL3 – 3 pages NL4 – 5 pages NL5 – 3 pages NL6 – 2 pages NL7 – 3 pages NL8 – 5 pages NL9 – 6 pages NL10 – 5 pages NL11 – 2 pages Germany 19 March 2020 9 April 2020 20 April 2020 23 April 2020 DE1 – 3 pages DE2 – 3 pages DE3 – 3 pages DE4 – 8 pages

(24)

24 30 April 2020

6 May 2020

DE5 – 3 pages DE6 – 3 pages The United Kingdom 3 March 2020

9 March 2020 12 March 2020 16 March 2020 17 March 2020 18 March 2020 19 March 2020 20 March 2020 22 March 2020 23 March 2020 25 March 2020 30 April 2020 10 May 2020 11 May 2020 24 May 2020 25 May 2020 28 May 2020 UK1 – 1 page UK2 – 1 page UK3 – 2 pages UK4 – 2 pages UK5 – 2 pages UK6 – 2 pages UK7 – 2 pages UK8 – 2 pages UK9 – 2 pages UK10 – 3 pages UK11 – 1 page UK12 – 3 pages UK13 – 4 pages UK14 – 2 pages UK15 – 3 pages UK16 – 2 pages UK17 – 4 pages V. Operationalization

This part will firstly go into the type of research performed in this thesis. After that it will present the operationalization table of the themes and indicators in order to analyse the speeches. Then, this part will discuss the measurement on the trust in government variable and will look into a check variable to test the results after the analysis.

The research performed in this thesis will focus on thematic analysis, which is about describing and interpreting content in a qualifying way. In this research, deductive thematic analysis is performed, using the propositions derived from theory and operationalizing these into measurable indicators. This to examine leadership crisis communication styles in the speeches of the various leaders. By using thematic analysis, one can identify the general themes and styles and connect these to indicators. A concern with doing this type of research is the reliability, because thematic analysis allows for a variety of interpretation by the researcher. Besides this, themes and codes can get mixed when finding and verifying them. However, the

(25)

25 approach allows for a variety of theory application by which the interpretation can be backed up. For all leaders, all themes and indicators are examined, otherwise this research would be a self-fulfilling prophecy by only looking at the male characteristics for male leaders and vice versa. Table 4 is set up to present the themes, connected indicators, describes the measurements, and gives examples in order to examine the speeches on these themes.

TABLE 4: OPERATIONALIZATION OF THEMES AND INDICATORS

Theme Indicator Measurement Example

Female leadership communication style

Concern for other people

• Express a concern for (specific groups of) people (e.g. the sick, weak, elderly) • Offer help or

show sympathy

“I would like to emphasize again that we have a huge responsibility to help especially the weak people in our society, the most vulnerable people with chronic diseases, cancer patients, elderly.” (DK1, p.1)

Empathy • Show emotions

in speech

“We sympathize with family and relatives” (NL5, p.1)

Participative • Asking people to

participate and to follow rules • Two-way style of communicating and leading

“You are doing your bit in following this advice to slow the spread of the disease” (UK9, p.2)

Proactive • Being a role

model for people • Giving people

motivation to follow the rules • Motivate people

to ideas to

change the future • Being

charismatic

“We did not choose for the corona ourselves. But we choose our own response. That what it is all about” (DK3, p.6)

Encouraging • Offering advice,

help and expressing sympathy to others “Everyone who avoids unnecessary encounters helps all those who are in hospitals providing care to more and more people each day” (DE1, p.3)

(26)

26 Symmetrical power relationship • Focus on the other/the people instead of on him/herself • Focus on the ‘we’ instead of ‘you’, ‘I’, ‘the people’ • Symmetrical

power relation to the people spoken to

“For this we need everyone, 17 million people, we need to fight the battle with all of us as a country and also see to win this together.” (NL2, p.3)

Male leadership communication style

Confidence • Show confidence

in the speech and the proposed rules/solutions • Giving opinion

in a powerful way

“We can truly slow the spread of the virus” (DE3, p.3)

Courage • Express courage

in the future and solutions

“But we will get through this. We will get through it

together, and we will beat this virus” (UK8, p.2) Dominance • Engaging in direct conflict to show dominant and defensive strategies • Direct speech in order to present dominance so people will follow

“I’ve got to be clear, we’ve all got to be clear, that this is the worst public health crisis for a generation.” (UK3, p.1) Task accomplishment • Focused on accomplishing a task • Speech is focused on achieving an outcome • Setting up standards or rules (e.g. lowering the infection rate)

“We must work to bring down the number of infected people still further, … Above all, the aim is also for us to be in a position to trace the chains of infection”

(27)

27

Directive • Stating rules that

people need to follow • One-way style of communicating and leading “Stop shaking hands” (NL1, p.2) Reactive • Focus on maintaining the status quo • Primarily passive style • Try to get compliance through reward and punishment

“Making sure the private sector can go on as long as

possible in the best way possible” (DK1, p.2) Asymmetrical power structure • Focus on him/herself instead of on the other/the people • Asymmetrical power relation, leader vs people, they need to listen and follow

“To ensure

compliance with the Government’s instruction to stay at home.” (UK10, p.2)

a. Public trust in government measure

Measuring public trust in government will be done in two ways. First of all, results of the Eurofound e-survey on ‘Living, working and COVID-19’ will be used. The data of this survey will be compared with the data on pre-COVID-19 trust in order to say something about changes in trust. The Eurofound survey was created to investigate the impact of the pandemic on the lives of the people in Europe (Eurofound, 2020). The survey consisted of two rounds, the first one in April/May 2020 and the second one in June/July 2020. The question asked regarding trust in government is ‘Please tell me how much you personally trust each of the following institutions – Your country’s government’ (Eurofound, 2020). Trust is measured on a scale of 1-10, in which 1 means no trust at al and 10 means complete trust (Eurofound, 2020). This scale is transformed into percentages by multiplying the numbers with 10. This survey gives the results for Denmark, the Netherlands and Germany, but not for the United Kingdom as they are not a member of the European Union anymore. Therefore, a different measure for the UK is needed.

Here for, the research uses the Governmental approval tracker by YouGov, which asks ‘Do you approve or disapprove of the Government’s record to date?’ (YouGov, 2020). This tracker

(28)

28 presents the approval, disapproval and ‘don’t know’ rates over time (YouGov, 2020). The choice for this measure needs further explanation because it does not measure trust directly. Extensive research was done to compare the YouGov measure to newspaper articles and other data sources that examined trust in government in the UK. The Reuters Institute (2020) did research on how the British public navigates information and disinformation about the coronavirus and on how the government is responding to the pandemic. In their fourth factsheet on comparing findings of mid-April to end-May, they find that trust in the UK government as a source of information has substantially declined in this period (Fletcher, Kalogeropoulos, & Nielsen, 2020). In mid-April, 67% did trust the government on news and information, while at the end of May, only 48% did (Fletcher et al., 2020). Comparing this to the data of YouGov, at mid-April, 50% did approve of the UK government, while at the end of May, only 37% did (YouGov, 2020). The numbers differ, but the trend is similar in comparison. Another research by Fancourt, Steptoe, and Wright (2020) on politics, trust and behaviours during the pandemic found that public trust in government increased around mid-March when the lockdown was initiated. Further, after 10 May, when easing measures of the lockdown were announced, they found that trust decreased (Fancourt et al., 2020). The YouGov (2020) governmental approval rate shows that between 23 March and 10 May the approval rate is high, with a peak at the start at 52% approving and being at 44% approval at 10 May. After this date, the approval rate of the UK government only decreases, and more people disapprove of government than they approve. The trends in the approval rate are similar to studies that did research on aspects of trust; therefore, this measure is in this case suitable to use for this research.

Secondly, the trust in government is connected to the rally round the flag theory, in which the public rallies behind their leader in times of crisis. In order to see whether the public rallies behind their leader in this research, the opinion polls from POLITICO are used. POLITICO tracks polling data regarding every European election and country (POLITICO Europe, 2020). Here for, the national parliament voting intentions for all the four cases will be analysed. In the database from POLITICO, all the parties of a country are represented and the model presents the voting intention for each party, measured in different periods by opinion polls (POLITICO Europe, 2020). By analysing these polls and the voting intention, one can analyse the changes in the voting and preferences over time and can therefore see whether rallies exist. This measure is only for in-country comparison of the voting intentions for the leader. These voting intentions cannot be compared between countries because the countries have different parliamentary systems. The UK has a Westminster system which is essentially consisting of two parties, while

(29)

29 the other countries have a multi-party system. Here for, a certain percentage of trust in a party will have a different meaning in a multi-party or in a two-party system.

(30)

30

4. Analysis

This chapter will present the results of the thematic analysis. It will therefore go into the resulting codes, related to the set-up themes, that are found in the speeches. Firstly, the results of Denmark and the Netherlands will be presented, after which these countries can be compared on their leadership communication styles and the outcome of public trust in government. Then, the results and comparison of Germany and the United Kingdom will be presented. In order to relate to the outcome, the research design matrix of Table 1 is presented here once again. Additionally, the table now includes the outcomes of trust for the countries. Please note that, as discussed in the methods chapter, another measure had to be selected for the COVID-19 period of trust in government in the United Kingdom because previous data tools were no longer available due to Brexit.

TABLE 5: RESEARCH DESIGN MATRIX Denmark

• Small country • Feminine society • Female prime minister

• Trust pre-COVID-19: 63% tends to trust and 32% tends not to trust • Trust April/May: 73% tends to

trust

• Trust May/June: 64% tends to trust

The Netherlands • Small country • Feminine society • Male prime minister

• Trust pre-COVID-19: 59% tends to trust and 37% tends not to trust • Trust April/May: 60% tends to trust • Trust May/June: 50% tends to trust Germany

• Large country • Masculine society • Female prime minister

• Trust pre-COVID-19: 50% tends to trust and 45% tends not to trust • Trust April/May: 59% tends to

trust

• Trust May/June: 57% tends to trust

The United Kingdom • Large country • Masculine society • Male prime minister

• Trust pre-COVID-19: 21% tends to trust and 72% tends not to trust • Approval April/May: 49% approving

and 30% disapproving

• Approval May/June: 35% approving and 44% disapproving

After each country comparison, a table is presented which shows how often the elements of Table 3 of the operationalisation were found in the speeches of the prime ministers. The

(31)

31 presence of these elements is shown with an index that ranges from -- to ++. In this index, -- refers to an element, for example empathy, being present 0-2 times in the speeches. – refers to an element being present 3-5 times and +/- means that the element is present 6-8 times in the speeches. Then, + implies that the element is present 9-12 times and lastly, ++ means that the element is present 13 times or more in the speeches. This index applies to all cases. The following table, Table 6, presents the empty table that will be filled in at the end of this analysis. After each part of the comparison, the table is filled for the specific countries in the comparison.

TABLE 6: COUNTRY COMPARISON (FINDINGS AVAILABLE ON PAGE 49) Theme Indicator Denmark The

Netherlands Germany The United Kingdom Female leadership communication style Concern for other people Empathy Participative Proactive Encouraging Symmetrical power relationship Male leadership communication style Confidence Courage Dominance Task accomplishment Directive Reactive Asymmetrical power structure I. Denmark

a. Concern for people, empathy and encouraging

The prime minister of Denmark, Mette Frederiksen, expresses much concern for people in her speeches. She does this by expressing a concern for people’s lives and the losses of life, but also about specific groups of people. For example, she states that “every loss of life is a tragedy” (DK2, p.1) and that the Danish strategy is focused on “first of all to save lives. If less people get infected, less will die. We therefore need to care for our elderly and vulnerable” (DK2, p.1). Furthermore, she speaks with empathy and emotions, in trying to get understanding by for example saying that the rules “mean that we as Danish people will see each other less” (DK1,

Leadership communication and public trust in government during crisis: do gender traits make a difference?