Grand Challenges and Female Leaders: An Exploration of Relational Leadership During the COVID-19 Pandemic

Abstract

Managing grand challenges demands a relational leader who encourages collaboration, coordination, and trust with various stakeholders. Although leaders appear to play a critical role in addressing grand challenges, relatively little research exists about the factors that inform stakeholder perceptions of leaders during a grand challenge. To address this limitation, we integrate implicit leadership theory and gender role theory to consider stakeholders’ gender prescriptive expectations when evaluating leader effectiveness during the COVID-19 pandemic. We theorize that stakeholders advantage female leaders based on mental schemas of what is required in a pandemic—relational leadership—and stakeholders’ prescriptive expectations of female leaders as more relational. Using a laboratory experiment, we find that female leaders are perceived as more relational, and hence, more effective than their male counterparts. Our findings advance scholars’ and practitioners’ understanding of strategic leadership, stakeholder management, and grand challenges.

Management scholars have become increasingly interested in societal grand challenges (George et al., 2016; Howard-Grenville, 2021). A grand challenge constitutes a “specific critical barrier that, if removed, would help solve an important societal problem with a high likelihood of global impact through widespread implementation” (George et al., 2016, p. 1881). Examples of grand challenges include climate change, societal aging, natural resource management, gender inequality, and health and well-being—for example, the United Nations Sustainable Development Goals (2021). Grand challenges are complex and uncertain, with no agreed-upon solution (Ferraro et al., 2015).

The global COVID-19 pandemic fits the definition of a grand challenge facing society. The social, economic, and health consequences of COVID-19 affect numerous stakeholders, and a collaborative effort among them is required to address the pandemic (George et al., 2016; Howard-Grenville, 2021; Pinkse & Kolk, 2012). Given the complexities associated with this effort, and when the path to success is often unknown, prior research has suggested the need for organizational, political, and community leaders (henceforth, “societal leaders” or “leaders”) who can coordinate and collaborate with diverse stakeholders (e.g., business, health, and public policy) while communicating with the public effectively (Howard-Grenville, 2021; Schad & Smith, 2019). Such leaders must often take a relational approach, emphasizing collaboration, open communication, and trust (Carmeli et al., 2009).

Although societal leaders appear to play a critical role in addressing grand challenges, we still know little about the factors that inform stakeholder perceptions of leader effectiveness during a grand challenge. We believe this is a critical omission because prior work across a variety of management disciplines (e.g., leadership, stakeholder management, upper echelons) suggests that stakeholder perceptions of leader effectiveness influence a leader’s ability to lead (Lord et al., 2020; Offermann & Coats, 2018; Shondrick et al., 2010). For instance, examining the role of narcissistic CEOs during the COVID-19 pandemic, Kim and colleagues (2021) explained that stakeholders “pay close attention to CEOs’ personal characteristics and adjust their behaviors accordingly” (p. 1283). Therefore, by gaining a better understanding of stakeholder perceptions of leader effectiveness during a grand challenge, we can deepen our understanding of the factors that impact our ability as a society to address grand challenges. As such, we ask, “Are certain types of leaders perceived as more relational and hence more effective at addressing grand challenges such as COVID-19?”

To answer this question, we recognize that inferences regarding leader effectiveness can be difficult for stakeholders because they are often at arm’s length from leaders (Boivie et al., 2016; Fanelli & Misangyi, 2006). This difficulty is further exacerbated in situations with elevated levels of uncertainty, such as with managing grand challenges. In such contexts, stakeholders rely on heuristics or stereotypes to form judgments (Fiske & Taylor, 1991). Given these circumstances, the strength of unambiguous demographic signals may be especially relevant to perceptions of leader effectiveness. When considering which of a leader’s demographic signals might inform stakeholder perceptions of effectiveness, we focus on gender for two principal reasons. First, psychology research contends that gender provides the most robust basis for stereotyping people (Stangor et al., 1992). Gender stereotypes are widely held (Heilman et al., 1995), automatically activated (Devin, 1989), and influential (Banaji & Hardin, 1996). As such, gender has “a remarkable ability to dominate in impression formation” (Heilman & Parks-Stamm, 2007, p. 49). Thus, unlike other attributes (e.g., education, experience), gender represents a clear and easy-to-observe signal likely to shape stakeholders’ perceptions of leaders’ approaches and—ultimately—effectiveness.

Second, scholarly and popular media accounts present a tension regarding how female leaders may be perceived during a grand challenge. In general, female leaders are often victims of a “think leader—think male” mindset that leads others to perceive female leaders as ineffective in, or unfit for, leadership roles (Koenig et al., 2011; Schein, 1973). This mindset is one reason the media are often accused of being overly critical of female leaders such as Hillary Clinton and Theresa May as they tried to tackle grand challenges (Campus, 2013; Chamorro-Premuzic, 2019). Conversely, female leaders have also been viewed as having a more relational approach than their male counterparts during COVID-19, which may influence perceptions of their effectiveness in addressing grand challenges. For example, Germany’s Angela Merkel has been praised for creating a calm, reasoned message that “hit home” (Rising & Moulson, 2020). Similarly, New Zealand’s Jacinda Ardern has been noted for her “clarity and compassion” (Marks, 2020). The contrast between the “think leader—think male” and “think female—think relational” mindsets that color perceptions of female leaders further highlights the need to establish whether female leaders are genuinely perceived as more effective at addressing grand challenges, such as COVID-19.

With our investigation, we draw on implicit leadership theory (Lord et al., 2020) and integrate it with gender role theory (Eagly & Wood, 2012) to theorize that when facing a grand challenge, stakeholders will seek a relational leader—an open communicator—who encourages collaboration, coordination, and trust (Carmeli et al., 2009; Mazzei et al., 2012). These perceptions align with society’s prescriptive stereotypical view of female leaders as the more communal gender. While not all female leaders will reflect this stereotype, female leaders are often perceived to be—and also socialized to be—more relationship-oriented (Eagly et al., 2000; Heilman, 2012; Prentice & Carranza, 2002). As such, we argue that stakeholders facing a grand challenge, such as COVID-19, will break from the traditional “think leader—think male” mindset and prefer female leaders based on the relational skills stakeholders demand during these periods (a “think female—think relational” mindset; Kahn et al., 2013; Mishra, 1996). In turn, we theorize that stakeholder preferences for a relational leader and stereotypical views of female leaders lead to enhanced perceptions of female leaders’ effectiveness during the COVID-19 pandemic.

We test our arguments with a laboratory experiment. We find that stakeholders’ gender prescriptive expectations of leaders result in female leaders being perceived as 12% more relational in their leadership approach than their male counterparts, all things being equal. We also find that in the context of managing COVID-19, there is a significant and direct link between relational leadership and perceptions of leader effectiveness. Together, these results yield a significant, fully mediated model that links female leaders to increased perceptions of leader effectiveness through relational leadership.

In doing so, we provide several contributions. First, we inform the increasingly important stream of management research focused on societal grand challenges. Yet, this stream remains nascent, consisting mainly of conceptual and editorial articles (Eisenhardt et al., 2016; George et al., 2016) focused on understanding the general effects of grand challenges on businesses (Kulik et al., 2016; Olsen et al., 2016). We build on these efforts and recognize that grand challenges involve intersecting issues related to strategic leadership, stakeholder management, and societal norms. To this end, we integrate key streams of literature on implicit leadership theory and gender role theory to highlight stakeholders’ gender prescriptive expectations of leaders in addressing grand challenges such as COVID-19. Focusing on gender offers evidence that stakeholders naturally advantage certain types of leaders based on mental schemas of what is required in the context and their perceptions of a specific leader. Ultimately, we move the conversation on grand challenges in management research forward by unpacking the complex relational processes between societal leaders and stakeholders involved in addressing these challenges.

Second, as the debate continues around when female leaders benefit or suffer from gender bias (Jeong & Harrison, 2016), we contribute by theoretically and empirically isolating how perceptions of relational leadership are a critical mechanism to explain how stakeholders might favorably evaluate female leaders. Ultimately, contrary to the general “think leader—think male” mindset at the core of gender role theory, we demonstrate that female leaders have a “think female—think relational” advantage in contexts where stakeholders are naturally more predisposed to relational leadership approaches, such as grand challenges (Eagly & Carli, 2003; Post et al., 2019; Rosette & Tost, 2010). In doing so, we answer the recent call “to better understand how women leaders are perceived” (Jeong & Harrison, 2016, p. 37).

Finally, by investigating perceptions of leader effectiveness during COVID-19, we shed light on the crisis management aspect of grand challenges. While prior work demonstrates that leaders can influence stakeholder perceptions (Bundy et al., 2017; Neville, 2020), it has primarily focused on stylistic elements of leaders’ communication strategies and not on perceptions of their attributes (Jun & Wu, 2021). As the relationship between business and society becomes more intertwined (Krause & Miller, 2020), our need to understand who a leader is—as the “face” of an organization or country—seems just as important as what they say.

Theoretical Framework

As defined above, a grand challenge refers to a significant obstacle that, if addressed, would help advance society on social, environmental, and economic dimensions (George et al., 2016). Grand challenges are considered “seemingly intractable” puzzles (Ferraro et al., 2015, p. 367) that do not offer straightforward solutions or consistent assumptions. The 17 Sustainable Development Goals (SDGs) of the United Nations (UN) are commonly viewed as the most comprehensive list of ongoing grand challenges. Some examples include ensuring water and sanitation availability and sustainable management, full and productive employment and decent work, and inclusive, safe, resilient, and sustainable living environments (United Nations, 2021). Recently, an event that fits the definition of a grand challenge—ensuring health and well-being—is the COVID-19 pandemic.

Addressing a grand challenge is complicated; it requires coordinated and collaborative action toward a clearly articulated problem and goal (George et al., 2016; Longhofer et al., 2019; Pinkse & Kolk, 2012). This situation necessitates that societal leaders and stakeholders “coordinate their efforts to secure a common goal that none could obtain without the efforts of others” (Flanagin et al., 2006, p. 30). However, because societal actors’ and other entities’ (e.g., corporations) interests may not always align, consensus related to goals and potential solutions may be hard to achieve, making the type of coordinated and collaborative action needed to address a grand challenge a difficult objective. Given these complications, a grand challenge places a premium on leaders who can articulate a vision and bring diverse stakeholders together.

As stakeholders look toward leaders to guide them under challenging circumstances, their perceptions of leader effectiveness become critical determinants of whether these stakeholders will buy into a leader’s vision and proposed strategies (Hambrick & Lovelace, 2018; Kim et al., 2021). In times of uncertainty, individuals often make quick, heuristic causal attributions to assign responsibility to an individual for the actions of a group or an organization (Bingham & Eisenhardt, 2011; Fiske & Taylor, 1991). This phenomenon is perhaps best known as the romance of leadership (Meindl, 1995; Waldman et al., 2001).

For example, in a corporate context, the CEO is a psychologically easy and attractive option for stakeholders to blame or reward when considering the uncertainty and causal ambiguity associated with corporate events. Leaders that achieve public prominence or generate positive media coverage receive disproportionate praise for firm performance (Hammond et al., 2021). At the same time, research has established that the leader takes the brunt of the blame for adverse events such as financial restatements (Gomulya & Boeker, 2014), performance shortfalls (Sanders, 2001), and fraud (Cowen & Marcel, 2011). The same phenomenon also exists in politics, where voters reward and punish politicians for economic conditions often created before the leaders took office (Nye, 2008).

Given that stakeholders overweight the responsibility of leaders in general and assessing leader quality is hampered by the uncertain and complex nature of grand challenges such as COVID-19, it is essential to consider stakeholder perceptions of leader effectiveness in such situations. Adopting a follower-centric approach, implicit leadership theory asserts that followers’ perceptions of leadership are based on their “schemas of an ideal leader” (Hechanova et al., 2018, p. 917). These schemas are often used to identify the attributes and behaviors that distinguish effective leaders from ineffective ones (Caringal-Go et al., 2021; Holmberg & Åkerblom, 2006). Also, the theory states that the congruence of followers’ mental representations of a leader and the leader’s actual characteristics will influence how they act toward leaders. Surprisingly, however, the literature has not considered the factors that inform stakeholder perceptions of leaders in highly uncertain and complex contexts, such as a grand challenge. Below, we articulate theory explaining why gender may play an outsized role in influencing stakeholders’ reactions to leaders during COVID-19.

Grand Challenges and the Relational Leadership Advantage

When facing grand challenges, scholars advocate for a relational leadership approach based on collaboration, communication, and trust (Gittell, 2006; Maak & Pless, 2006; Schad & Smith, 2019; Williams et al., 2017). As such, relational leadership embodies skills critical to effectively managing grand challenges in several ways. First, grand challenges can significantly alter how individuals, families, and the public relate to one another (Kahn et al., 2013). Relational leaders are better equipped to handle the anxiety, panic, and distress that can arise from these events (König et al., 2020).

Second, crisis management practitioners and scholars advocate for leaders to connect emotionally and psychologically with stakeholders, suggesting that relational leaders’ focus on open communication and trust is effective when facing a grand challenge (Wooten & James, 2008). Medical studies supporting this approach have found that patients receive terrible news better when physicians use an empathetic, other-focused communication style (e.g., more approachable, compassionate) in a non-dominant fashion (e.g., less assertive, less intimidating)—all descriptors of relational leadership (Mast et al., 2005). Even in nonverbal encounters, Griffith and colleagues (2003) showed that patient satisfaction was higher when doctors engaged relationally—smiling, maintaining eye contact, and gesturing more.

Finally, a relational leader can better coordinate the myriad of interdependent tasks and actors (e.g., health providers, government entities, and the public) needed to address a grand challenge that hinges on accurate, frequent, timely, and solution-oriented communications bolstered by “shared goals, shared knowledge, and mutual respect” (Gittell, 2006, p. 74; Hoffer Gittell, 2002). In line with coordination, relational leaders are more likely to seek input from various sources and be more adaptable in evolving situations (Cunliffe & Eriksen, 2011) that are critical to addressing a grand challenge. For instance, relational leadership is linked to generating new ideas and innovation among followers (Peng & Wei, 2018; Schaubroeck et al., 2011); critical inputs in addressing the often-novel issues brought forth by grand challenges.

Overall, the preceding discussion suggests that relational leaders are more successful in addressing grand challenges such as COVID-19 and that stakeholders’ perceptions of how relational these leaders are will help ensure this success. Thus, the question arises: Which leader characteristics influence stakeholder perceptions of relational leadership?

“Think Female—Think Relational”

As we have highlighted, gender has a remarkable ability to influence the perceptions of leaders. Of course, perceptions of gender are also highly prone to stereotyping (Eagly & Mladinic, 1989; Heilman, 2012). Stereotypes are widely held but oversimplified beliefs, expectations, and assumptions of a particular type of person based on their group membership (e.g., males vs. females; Heilman & Parks-Stamm, 2007). Stereotypical judgments are automatic, immediate, and pervasive in daily life. An individual may implicitly hold a stereotype about another person without necessarily endorsing it personally (Macrae & Bodenhausen, 2000). Gender stereotypes, in particular, have been validated across time, cultures, and diverse leadership settings (Heilman, 2012; Williams & Best, 1990). For example, in the context of corporate boards of directors, Zhu and Westphal (2014) demonstrated that even among this supposedly sophisticated group, gender is a more salient attribute than age, education, functional background, or leadership experience when evaluating others.

Furthermore, it has been demonstrated that situational uncertainty and complexity allow for more bias in evaluation (Botelho & Abraham, 2017; Davison & Burke, 2000). The more uncertainty and complexity involved, the less evidence for an accurate assessment, and the more weight is necessarily placed on inference. Token or minority status, as is the case for female leaders, leads to more stereotyped characterizations because observers use gender stereotypes to fill in the missing information in such contexts (Kanter, 1977).

Gender role theory (Eagly & Wood, 2012) suggests that women are stereotyped as having communal qualities associated with their traditional role of homemaker. These descriptions stand in stark contrast to agentic qualities attributed to males—powerful, commanding, and assertive—related to the role of breadwinner. These stereotypes reflect how observers believe the genders are and help set observers’ expectations of how each gender should act. Vital to our investigation is that prescriptive stereotypes establish social rules for how each gender should act and are used to rationalize a social system that has historically been successful (Heilman & Chen, 2005; Heilman & Parks-Stamm, 2007). Prescriptive stereotypes for gender carry more weight in social interactions than other group stereotypes and are often more pervasive than observers realize. For example, Rudman and Glick (2008) argued that it is more socially acceptable to tell a woman she “ought” to be more nurturing or a man he “should” be more aggressive than to tell an Asian- or African American that they are not conforming to racial or ethnic stereotypes.

Prescriptive stereotypes about women specify that women should behave more communally and demonstrate social sensitivity and concern for others’ welfare, such as being kind, sympathetic, and understanding (Heilman & Parks-Stamm, 2007). These normative shoulds also come with reciprocal should-nots. For example, the desirable attributes for males, agentic achievement-oriented behaviors, are deemed inappropriate for women. Conversely, males should not display communal service-oriented behaviors (Eagly & Karau, 2002).

Interestingly, these prescriptive stereotypes are so strong in impression formation that they exist for female leaders, even though studies show that some female leaders are potentially no more communal or relational than their male counterparts. For example, Adams and Funk (2012) found that female directors were more agentic and risk-loving than their male counterparts, suggesting that women in leadership positions do not satisfy many of these traditional gender stereotypes seen in the general population. More recently, Gallus and colleagues (2020) demonstrated that gender differences diminish or become nearly nonexistent as one moves higher up in organizational hierarchies, reinforcing that female leaders may not be different.

Despite this evidence, female leaders are stereotypically perceived as more relational than their male counterparts (e.g., more compassionate, attentive, and sensitive to stakeholder needs; Dennis & Kunkel, 2004; Ibrahim et al., 2009). For example, Ryan et al. (2007, p. 190) highlighted that when participants perceived a match between women’s abilities and the communal skills necessary in managing an underperforming company, they assumed females possessed “special” relational attributes such as “women always want to help the underdog.” Similarly, Oliver and colleagues (2018) demonstrated that even among knowledgeable board chairpersons, CEO gender influenced perceptions of the CEO as a communal steward or an agentic-risk taker.

These findings translate into political settings as well. Lammers and colleagues (2009) found that female candidates are preferred when issues that demand relational and pro-social skills (such as health care) are deemed most important. Female leaders are viewed more favorably when they can negotiate on behalf of the “general welfare” or the “common good” (Anderson, 2020), critical components of battling a grand challenge. Constituents also view female leaders as better equipped to handle social issues, such as education, civil rights, poverty, and homelessness (Falk & Kenski, 2006; Huddy & Terkildsen, 1993), reinforcing that female leaders should be better equipped as relational leaders than males.

Some female leaders even reinforce these stereotypes through their words and actions. The first female CEO of a Big Four accounting firm, Lynne Doughtie of KPMG, stated, “I have found that women are really in their element in a very collaborative approach” (King, 2017). Specific to COVID-19, when Angela Merkel addressed the German public at the onset of the pandemic, she was flanked by multiple public health officials, and she took pains to say that the information she was sharing was collaborative and had come from experts (Bennhold & Eddy, 2020). This image stood in stark contrast to the top-down, autocratic leadership approaches taken by such leaders as the United States’ Donald Trump and Brazil’s Jair Bolsonaro (Dewan, 2020). Finally, a recent qualitative study of U.S. Governors’ remarks following the onset of COVID-19 also suggests that female governors used more empathetic language than male governors in addressing constituents (Sergent & Stajkovic, 2020).

Building on these considerations, we theorize that stakeholders’ evaluations of leader effectiveness during grand challenges will hinge on relational leadership attributes—being open, demonstrating trustworthiness, and collaborating well with others—that align with the prescriptive stereotypes female leaders are assumed to possess. Given the natural preference for a relational leader during a grand challenge, we argue that the “think female—think relational” mindset wins out in impression formation when evaluating female leadership as opposed to the prevailing negative “think leader—think male” mindset that may dominate in more general settings.

Taken together, when assessing a female’s leadership in managing a grand challenge such as COVID-19, stakeholders adopt a “think female—think relational” mindset aligned with prescriptive gender stereotypes. These stereotypes are often reinforced by society and even some current female leaders’ actions, as well as the uncertainty present in tackling a grand challenge and assessing leader effectiveness. Stakeholders expect a female leader to behave more relationally and demonstrate social sensitivity and concern for others’ welfare. As such, these stereotypes lead stakeholders to view female leaders as more relational and hence more effective at managing a grand challenge.

• Hypothesis: By engendering increased perceptions of relational leadership, stakeholders will perceive female leaders as more effective than male leaders at addressing grand challenges such as COVID-19.

Method

This study examines stakeholders’ perceptions of female leadership during a grand challenge, the COVID-19 pandemic. We tested our hypothesis in a way that allowed us to identify and isolate the influence of gender and perceptions of relational leadership on stakeholders’ perceptions of leaders’ effectiveness at managing COVID-19. To this end, an experiment was an ideal study design (Cook et al., 2002).¹ An important advantage of an experimental design is that it allows for higher levels of internal validity than cross-sectional and longitudinal studies by enabling researchers to make clearer inferences about the causal processes involved between the variables (Cook et al., 2002). Furthermore, many confounding factors that could influence stakeholder perceptions of leader effectiveness in the context of COVID-19, such as attribution of blame, severity, or communication style (Bundy et al., 2017), can be held constant to focus on the variables of interest. The principal drawback to this method is that external validity may be limited. However, as we describe below, we believe we have struck the right balance between internal and external validity by (a) carefully selecting an appropriate sample of participants and (b) building our experimental scenario based on realistic instances of leader announcements and communications in the context of a grand challenge.

Research Design and Procedures

Development, Validation, and Pretest of the Experimental Material

To develop the experimental material, we followed recommendations to adhere closely to real-life cases (Aguinis & Bradley, 2014; Highhouse, 2009). Knowing stakeholders turn to media coverage for information during grand challenges and similarly uncertain and complex events (Graf-Vlachy et al., 2020), we worked to establish experimental realism by presenting a scenario based on actual news coverage of COVID-19 (Aguinis & Bradley, 2014). Specifically, the scenario was adapted from a major news article discussing a leader’s response to COVID-19 in April 2020 (Fifield, 2020). The scenario describes the leader’s televised address, presents direct quotes from the leader to the public, and shares accounts of the leader’s conversation with influential stakeholders.

To test the strength of our manipulations and the reliability of our chosen scales, we conducted a pretest on an online sample of adults (U.S.-based participants) recruited through the online platform Lucid, a market research firm (Coppock & McClellan, 2019).² Of the 41 participants, 27 were male, and 14 were female, with a mean age of 46.41 (SD = 18.03) years. Our pretest demonstrated that our manipulation check was successful and that our scales were reliable. During the pretest, we also wanted to understand whether other characteristics of our scenario could be driving perceptions outside of the leader’s gender. Following prior research, participants were also asked to rate grand challenge severity, their state negative affect (Watson & Clark, 1999), and their feelings of hostile sexism to determine whether the scenarios generated any negative feelings that might influence their perceptions of the leader’s gender (Glick & Fiske, 1997).

To assess the severity of the grand challenge, the participants answered the following on a 5-point scale (strongly disagree to strongly agree)—“The crisis is (1) a major problem, (2) significant, (3) severe.” The coefficient alpha was .88. We measured participants’ state negative affect following instructions outlined in the PANAS–X Manual (Watson & Clark, 1999). Example emotions included upset, distressed, irritable, hostile, disgusted, and contempt. The coefficient alpha was .97. We utilized the hostile sexism scale developed by Glick and Fiske (1997) to assess whether any aspect of the scenario generated intense gendered feelings toward the target. The coefficient alpha was .89. No significant differences were found across these three measures (severity, negative affect, hostile sexism) between the male and female scenarios in the pretest. Based on these findings, we made minor changes to the scenarios in the experiment before conducting the main study (see the appendix for scenario).

Main Study Participants

Using the G*Power software, we conducted an a priori power analysis. We specified an effect size of 0.4, a conventional value indicative of a medium effect size (Cohen, 1988; Murphy et al., 2014). Following prior recommendations, we ensured that our chosen effect size was “the smallest effect that would be meaningful in some practical sense” (Fritz et al., 2012, p. 18), given that large effect sizes are not common in behavioral science (Cohen, 1988, p. 284). In addition, we specified a desired α = .05 and power (1 − β) = 0.8. The power analysis revealed a desired sample size of at least 52 participants for the experiment.

We recruited U.S.-based participants via Lucid, an online platform. The sample was restricted via preselection to participants 18 years or older who passed a consciousness check in which they were asked to briefly describe the scenario in their own words briefly. Given that our theory is not stakeholder specific, and our scenario relates to perceptions of country leaders, we deemed individuals that met the U.S. minimum voting age to be an appropriate sample. One hundred six participants accessed the survey instrument through Lucid. A portion of those participants was not included in the final sample because they either did not meet the age requirement, did not sign the informed consent, or did not pass the consciousness check.

The final sample included 82 participants, age (M = 46.41, SD = 16.83), 49.00% female. Regarding education, 35.36% had a high school diploma or equivalent, 31.70% had an associate’s or bachelor’s degree, 31.70% had a master’s degree, and 1.22% had a doctoral or professional degree. In terms of employment, 50.00% were employed full-time, 9.76% were employed part-time, 1.22% were employed seasonally, 24.39% were not employed, and specific to the ramifications of COVID-19, 14.63% were furloughed. In terms of political affiliation, 12.26% identified as independent, 41.46% as Republican, and 46.34% as Democrat.

Data Collection

We collected data to test our Hypothesis using a 2 × 1 factorial between-subjects design with the independent variable being leader gender (female vs. male). We randomly assigned participants to the two conditions, with the manipulated variable being leader gender (female vs. male). In Part I of the study, all participants read media coverage of a leader following the onset of COVID-19. The coverage was partnered with a gendered graphic of the leader behind a podium. We designed the scenario to provide enough background information on the leader, COVID-19, and the leader’s statement for participants to make informed judgments, yet also distract participants from the leader’s gender as the topic of the investigation.

Following the presentation of the leader’s statement, we then asked a series of questions to measure participants’ perceptions of relational leadership and leader effectiveness. These questions represented our mediator and dependent variable. In Part II, participants answered a series of questions related to their demographic and socio-cognitive attributes that were used in supplemental analyses to rule out alternative explanations.

Manipulated Independent Variables

We manipulated the leader’s gender by name and gender-relevant pronouns used in the media coverage. The headline and graphic were used to support the manipulation further. Participants indicated if the leader was male or female. An analysis of variance (ANOVA) revealed a significant main effect based on leader gender, F(1,39) = 26.90, p = .000, M = 0.85 versus −0.71, ensuring the manipulation’s effectiveness.

Dependent Variables

Leader Effectiveness

To assess leader effectiveness, we adapted measures based on perceived leader self-efficacy (Chen et al., 2001; Fast et al., 2014). We presented the items as leader-referent as opposed to self-referent. The participants answered the following on 5-point scales (strongly disagree to strongly agree)— “(1) Compared to others, this leader will do most tasks very well; (2) This leader will be able to achieve most of the goals that have been set; (3) This leader will be able to successfully overcome many challenges; (4) Even when things are tough, this leader will perform quite well.” The coefficient alpha was .90.

Relational Leadership

Adapting measures from Carmeli et al. (2009), we presented the participants with the following on 5-point scales (strongly disagree to strongly agree)—“In my view, this leader . . . (1) encourages collaboration; (2) cultivates a trustful environment; (3) encourages open conversation.’’ The coefficient alpha was .84. We also performed a factor analysis with the combined items from relational leadership and leader effectiveness, which confirmed a two-factor structure with discriminant validity.

Analysis

In our analysis, we conducted univariate ANOVAs and intercell contrasts to test our hypothesis directly. We tested all intercell contrasts (the cell mean differences in each information condition) using Fischer’s least significant test. To further test mediation, we used sgmediation in STATA 16 to perform the Preacher and Hayes (2004) bootstrapped test of mediation (1,000 replications; Edwards & Lambert, 2007; Hayes, 2013).

Results

An analysis of variance (ANOVA) revealed a significant main effect based on gender, F(1,80) = 29.96, p = .000, M = 0.53 versus −0.74, ensuring the manipulation’s effectiveness. We, therefore, did not drop any cases based on manipulation failure. Table 1 presents the relevant means and standard deviations for each dependent variable.

Table 1.

Means, Standard Deviations, and Correlations Among Variables.

Dependent variables	M	SD	1
Relational leadership	4.02	0.97
Leader effectiveness	3.86	0.97	0.75

Dependent Variables

An ANOVA of participants’ ratings on the relational leadership scale revealed a significant main effect for gender, F(1, 80) = 4.45, p = .038. Intercell contrasts provided support that females were viewed as more relational leaders. As displayed in Table 2, female leaders were rated as 12% more relational (4.26 vs. 3.81, 95% confidence interval = [0.025, 0.864]). An ANOVA of participants’ ratings on leader effectiveness did not reveal a significant main effect of gender on leader effectiveness. Given our primary interest lies in understanding the mechanism driving leader effectiveness, a significant test of the direct effect is not a prerequisite for a test of the indirect effect in our hypothesis (Hayes, 2013; MacKinnon et al., 2002; Zhao et al., 2010).

Table 2.

Means and Standard Deviations.

Condition	Relational leadership	Leader effectiveness
Male	3.81a (1.01)	3.76c (1.02)
Female	4.26b (0.88)	3.98c (0.91)

Note. Means within a column with different subscripts differ significantly (p = .038) as indicated by Fisher’s least significant difference procedure.

Mediation

Given the significant direct effect of gender on relational leadership and the significant correlation between relational leadership and leader effectiveness (r = .75, p =.000) in Table 1, we estimated the path estimates of the indirect, direct, and total effects of leader gender on leader effectiveness. In Table 3, the path coefficient from female leader to relational leadership (a = 0.46; 95% confidence interval = [0.042, 0.887]) was significant, as was the path coefficient from relational leadership to leader effectiveness (b = 0.76; 95% confidence interval = [0.608, 0.913]). Likewise, the positive indirect effect of female leadership on leader effectiveness through relational leadership (a × b = 0.35; 95% bias-corrected bootstrap confidence interval = [0.014, 0.692]) estimated with 1,000 replications was also significant. These findings provide support for our hypothesis—leader gender has an indirect and significant effect (fully mediated) on perceptions of leader effectiveness in managing COVID-19, given that participants perceive females as more relational leaders.

Table 3.

Path Estimates of Indirect, Direct, and Total Effects for Gender Predicting Leader Effectiveness.

	Female → Mediator	Mediator → Leader effectiveness	Indirect	Direct	Total
Relational leadership	0.46 (p = .032)	0.76 (p = .000)	0.35 (p = .041)	−0.13	0.22

Note. Significance tests for the indirect and total effects are based on the bias-corrected confidence intervals derived from bootstrapping estimates with 1000 samples.

Supplemental Analyses and Robustness Checks

Given that stakeholders—and their perceptions—are heterogeneous, we were interested to see whether our outcomes differed based on specific stakeholder characteristics. Via a series of ANOVAs, we also tested whether outcomes differed based on participants’ gender, political affiliation (0 = independent, 1 = conservative, 2 = liberal), age, or education level (coded 1 if participants completed at least 4 years of college). We found no significant differences in terms of relational leadership among each of the four variables while controlling for leader gender. Also, none of the four controls had a significant effect on leader effectiveness or weakened the influence of relational leadership on leader effectiveness when included in the model. Furthermore, we ran our mediation analysis with participant gender, political affiliation, age, and education level as controls, and our indirect effect was consistent and significant across each model. The fact that participants’ characteristics did not influence stereotypical perceptions that females are more relational is not surprising, given that gender provides the strongest basis for stereotyping and categorizing people (Haslam & Fiske, 1992; Stangor et al., 1992).

We were also curious about participants’ knowledge of COVID-19 and whether it influenced their leadership perceptions. As such, we created a COVID-19 knowledge measure based on facts listed on the Centers for Disease Control and Prevention’s website. Example items included “(1) The virus is thought to spread between people who are in close contact with one another (within about 6 feet); (2) Symptoms may appear 2–14 days after exposure to the virus; (3) The virus is thought to spread through respiratory droplets produced when an infected person coughs, sneezes, or talks.” Participants responded on 5-point scales (strongly disagree to strongly agree) with a mean answer of 4.51 and coefficient alpha of .80, indicating participants were aware of the pandemic. There were no significant differences in knowledge across the two conditions and no significant relationship between COVID-19 knowledge and leader effectiveness when controlling for leader gender and relational leadership.

We also estimated our main indirect effect using Imai and colleagues’ causal mediation method (Imai et al., 2011) via the medeff command in STATA 16 (Hicks & Tingley, 2011). The estimate of the average causal mediation effect was similar and significant (b = 0.36, 95% confidence interval = [0.021, 0.698]). We also used the medsens command (Cox et al., 2013) to calculate sensitivity analyses for our mediation effect due to the non-random assignment of relational leadership. We found that a confounding variable would have to be correlated with our predictor and outcome variables at (r = .74) to bias our results, providing further confidence in our analyses.

If successful leadership is associated with stereotypically male or agentic qualities, it may not be associated with stereotypically female or relational qualities. This lack-of-fit argument is often referred to as the “think leader—think male” association (Heilman & Haynes, 2005; Koenig et al., 2011; Schein, 1973). A host of studies argue that this mentality results in less favorable assessments of female employees’ and executives’ competence (Eagly & Makhijani, 1992; Lyness & Heilman, 2006). As such, we tested how perceptions of leader agency influenced our relationships. Example items included self-confident, determined, decisive, assertive, and does not get dominated by one’s feelings (Kulich et al., 2018). The coefficient alpha was .91. We also performed a factor analysis to ensure that the leader agency and leadership effectiveness measures loaded onto separate dimensions. Leader gender was not significantly correlated with leader agency. Furthermore, the relationship between relational leadership and leader effectiveness remained significant while controlling for leader agency and leader gender. These findings further support our claims that the benefits females receive via positive stereotypes of their relational skills offset any penalties they may receive via negative stereotypes of their (lack of) agency.

Finally, to ensure that our results were indeed driven by gender and not the stimulus name in our scenario (Michele/Michael), we conducted a posttest to ascertain whether our chosen names conveyed different meanings to respondents (N = 81, 29% female, M_age = 38.40, Amazon mTurk). Participants rated a selection of names (Michele/Michael/Deborah/David) on the extent to which they were wealthy, attractive, or intelligent. We chose these characteristics as prior research suggests certain stimulus names may introduce bias based on these criteria in experiments (Kasof, 1993). Our results yielded no significant differences (p > .05) in ratings for Michele versus Michael, suggesting that both names convey similar meanings for respondents. These posttest findings, as well as our careful use of gendered headlines, graphics, and pronouns, increase our confidence that the results of our main study are not attributable to an effect of a leader’s name but, indeed, are driven by the leader’s gender.

Discussion

We set out to investigate stakeholder perceptions of leaders’ effectiveness at addressing a grand challenge—“formulations of global problems that can be plausibly addressed through coordinated and collaborative effort” (George et al., 2016, p. 1889). Given the strength of gender stereotypes in external impression formation and the role of uncertainty and complexity in biasing judgments, we theorized that female leaders were more likely to signal the desired relational attributes that stakeholders seek during a grand challenge. Specifically, we found that stakeholders do “think female, think relational”—resulting in increased perceptions of female leaders’ effectiveness in the context of managing the COVID-19 pandemic.

Our work makes several contributions to both research and practice. First, practitioners and management scholars recognize the “emerging trend of strategic leaders becoming—in both their own and stakeholders’ perceptions—societal leaders” (Krause & Miller, 2020, p. 1315). At the same time, management scholars are beginning to recognize the necessity of investigating the role of organizations and their leaders in understanding the influence of, and their influence on, the grand challenges facing society. To date, the literature on grand challenges has remained mostly silent on how specific leader characteristics influence stakeholder reactions. Given the importance of understanding stakeholder perceptions in terms of leader effectiveness, and that the cognitive processes through which stakeholders evaluate leaders can be seen as “strongly influenced by human judgment, perceptions, and emotions” rather than hyper-rational (Zadeh, 1975, p. 200), this is a surprising omission. By integrating key insights from implicit leadership theory, gender role theory, and the literature on grand challenges, we advocate for a stakeholder-focused approach to understanding how stakeholders’ gender prescriptive expectations influence perceptions of leader effectiveness during a grand challenge. We hope our work will spur interest in stakeholder-centered theory development when investigating grand challenges and other uncertain and complex contexts. This need is essential as stakeholder perceptions of leader effectiveness can significantly affect a leader’s ability to lead under challenging circumstances. As such, we contribute by considering stakeholder perceptions during a grand challenge to gain a deeper understanding of grand challenges, strategic leadership, and stakeholder responses (Caringal-Go et al., 2021).

Second, we isolated a context and theoretical rationale that illuminates a potential female leadership advantage. In doing so, we contribute to the practitioner and scholarly debate about whether female leaders and female-led organizations benefit or suffer from stereotypical bias (Jeong & Harrison, 2016). By focusing on reactions to female leadership during COVID-19, we also move the conversation past the appointment or election of female leaders (Dixon-Fowler et al., 2013; Lee & James, 2007) and focus on the management of a specific context that provides a female leadership advantage with stakeholders—a grand challenge. Our experimental design also allowed us to isolate the psychological mechanism—relational leadership—driving the female leader advantage during a grand challenge. In doing so, we answer calls to theoretically and empirically “open the black box of women’s leadership” (Hoobler et al., 2018, p. 2488) in a controlled lab setting to understand how prescriptive expectations of gender translate into differential outcomes for female leaders.

Third, we contribute to the crisis management literature. Much of the literature on crisis management is focused on issues in which attributions of responsibility and solutions are mostly clear, including fraud, product recalls, and environmental wrongdoing (Bundy et al., 2017). In contrast, the types of crises associated with grand challenges have garnered little attention from crisis management scholars. We contribute by highlighting the necessity to understand perceptions of leadership in a context in which the root causes and solutions are often uncertain and complex.

Finally, our work has implications for practitioners tasked with navigating a grand challenge. Our findings offer some evidence that leadership style—namely, a relational approach—can critically influence stakeholders’ perceptions during challenging times. Moving forward, leaders must recognize the need to appear relational in contexts requiring collaboration and trust to navigate. More personally for said leaders, stakeholders’ perceptions may have costly career consequences for leaders regardless of whether said perceptions of relational leadership and effectiveness rooted in stakeholders’ mental schemas are accurate.

While we focused on female leaders, we believe our findings may be particularly important for male leaders to understand, given they seem to start at a “relational deficit” in the eyes of stakeholders. We believe this insight may potentially translate across numerous contexts (political, organizational, or civic associations). Organizations, governmental, and public relations agencies may wish to train leaders to demonstrate relational attributes when engaging with diverse and multiple stakeholder groups. Concretely, the literature on crisis communication provides a variety of ways a leader can signal a more relational approach through their impression management and organizational response strategies (Bundy et al., 2017).

While our work focused on how stakeholders generally perceived leaders due to the strength of gender stereotypes, it would be beneficial for leaders to also consider whether their actions and messages relaying relational leadership are received similarly across a heterogeneous set of stakeholders. It may be necessary for leaders to tailor their messaging to build trust and collaboration with disparate stakeholder groups. One can imagine a scenario where the mechanisms for interpreting relational leadership might be quite different based on the worldviews and utility functions of different stakeholders (Bridoux & Stoelhorst, 2014; Harrison et al., 2010; Lange et al., 2022).

Future Research Directions and Limitations

Future research should investigate whether our findings depend on a specific leadership role. For example, did stakeholders respond differently to Dr. Deborah Birx, the White House Coronavirus Task Force coordinator, because her role was more technical and expertise-driven than executive-level leaders, whose expertise is often more empathetic in nature? Do stereotypical views of female leaders as less competent and less likable than their male peers manifest themselves more in lower-level and technical roles where competence is not just assumed (Rudman & Glick, 2008)?

Also, future research should unpack if the desire for relational leadership is isolated to the onset of grand challenges or if these preferences change as the challenge enters various stages (Bundy et al., 2017). Are there residual benefits for being a relational leader who has handled a grand challenge well, or do observers hold a relational leader to a higher standard once the initial challenge is mitigated or underway? Also, while we focused on the stereotypical views of female leaders as relational leaders, future work should investigate the implications of leaders violating stereotypical expectations while managing a grand challenge (e.g., female leaders demonstrating low levels and male leaders displaying high levels of relational leadership). For example, a recent article on BBC.com highlighting CEOs’ responses to COVID-19 suggested that “what many [men] have done is borrow from the female playbook, which involves being incredibly caring of their stakeholders, and upping their communication skills” (Goswami, 2020). Understanding how leaders move between different leadership styles and the ramifications of such moves in terms of stakeholders’ perceptions of both male and female leaders would also be fruitful.

Future research should also investigate whether our theory and findings translate to other grand challenges and national contexts where relational leadership might be less desired. For example, examining what leadership styles are preferred during the Black Lives Matter movement in the United States is ripe for investigation. Keisha Lance Bottoms, the former mayor of Atlanta, received praise for her response to the unrest in her city following the killing of George Floyd in Minneapolis in 2020. She evoked her relational nature in her statement,

I am a mother to four black children in America, one of whom is 18 years old. And when I saw the murder of George Floyd, I hurt like a mother would hurt. . .So, you’re not going to out-concern me and out-care about where we are in America. (Krieg & LeBlanc, 2020)

Like Angela Merkel’s press conference at the onset of COVID-19, Bottoms surrounded herself with a cadre of support that included the police chief, social activists, and local artists to reinforce her message. However, would we expect similar praise from stakeholders in other countries for their leaders under such circumstances?

Given that gender stereotypes are pervasive across many stakeholders, we chose not to theorize that differences could potentially exist across individual participants. Future research might explore how the individual stakeholder attributes (age, political orientation, regulatory focus, and so forth) influence the mental schemas of appropriate leadership deployed by the individual stakeholder. In tandem with these insights, future research could investigate how these individual stakeholder attributes influence their perception of leader characteristics that are not as powerful in impression formation as surface-level characteristics such as gender.

Given the rarity of the COVID-19 context and the limited number of female leaders, we used a scenario-based methodology in our controlled lab experiment to test our arguments. Because of the documented influence of news coverage on public perceptions (Graf-Vlachy et al., 2020), we worked to establish experimental realism by presenting news coverage similar to what the public consumed during the onset of COVID-19 (Aguinis & Bradley, 2014). Future research should work to support our causal link further and measure stakeholders’ reactions to COVID-19 leadership directly or via other nonexperimental means, such as polling data or social media posts.

We are also at a critical point to begin investigating the ramifications of female appointments and elections amid social and economic unrest (two global challenges). Amid the #TimesUp and #MeToo movements, 11 CEOs stepped down in 2017, six of whom were replaced by women, well above the average replacement rate of 18%. In the U.S. midterm elections in 2018, women set records regarding candidates and electoral victories (Cooney, 2018). Motivated by these data, scholars should work to theoretically and empirically unpack the antecedents of increased female representation as organizational and political leaders during unrest, as well as the aftermath of such decisions on stakeholders’ perceptions and the leaders’ careers.

Similarly, organizations and institutions could consider the strategic implications of deploying a female leader during an organizational scandal or crisis. Do stakeholders prefer relational leadership in contexts different from a grand challenge? This logic could be why Sheryl Sandberg, Facebook’s former Chief Operating Officer, often accompanied Mark Zuckerberg as the public face of Facebook’s apology tour following the Cambridge Analytica scandal. “We know that we did not do enough to protect people’s data,” Sandberg said. “I’m really sorry for that. Mark is really sorry for that, and what we’re doing now is taking really firm action” (Sydell, 2018). Similarly, does the National Rifle Association (NRA) strategically deploy a female spokesperson, Dana Loesch, because she “softens” the image of a stigmatized organization (Creitz, 2020)?

Finally, as our research shows, stakeholders do not find a female leader more effective solely because of her gender but rather through perceptions of her relational leadership. For example, initial evidence shows that female-led countries are not experiencing statistically lower death rates from COVID-19 (Elsesser, 2020). Thus, female leaders need to be aware of the power associated with being viewed as female and as relational leaders in managing a grand challenge. Future research should investigate whether relational leaders, male or female, are more effective in their messaging and whether their impact on public perceptions goes beyond a grand challenge context.

Conclusion

To date, scholars have focused on how gender norms (Eagly & Karau, 2002), coupled with the pervasiveness of the stereotypical “think leader—think male” mindset, have limited females’ access to top leadership positions in business and government and often disadvantaged female leaders in terms of evaluations (Dixon-Fowler et al., 2013; Ryan & Haslam, 2007). However, this grand challenge—the COVID-19 pandemic—allowed us to investigate whether there are contexts in which there is a perceived female leadership advantage (Dezsö & Ross, 2012; Eagly & Carli, 2003; Rosette & Tost, 2010). Through our investigation, we disentangled gender differences in perceptions and evaluations of leaders during a grand challenge. Specifically, we provided theoretical grounding and empirical support to substantiate the claim that females’ perceived communality can be an asset during a grand challenge due to the public’s preference for relational leadership. As such, we documented that differences in perceptions of relational leadership based on leader gender can indeed influence evaluations of leaders.

Author Biographies

Abbie Griffith Oliver (PhD, University of Georgia) is an assistant professor at the University of Virginia, McIntire School of Commerce. Her research explores the intersection of corporate governance and social evaluations. She focuses on the socio-cognitive mechanisms that shape strategic decision making and external perceptions of firms. She also explores how diversity influences these relationships in the upper echelons. Her articles have appeared in Academy of Management Journal, Journal of Management, and Strategic Management Journal.

Michael D. Pfarrer (PhD, University of Maryland) is the C. Herman and Mary Virginia Terry Distinguished Chair of Business Administration and Associate Dean for Research and Executive Programs in the Terry College of Business at the University of Georgia. His research focuses on organizational reputation and celebrity, impression and crisis management; media and corporate communications; and the role of business in society.

François Neville (PhD, Georgia State University) is an Associate Professor of Strategic Management at McMaster University, DeGroote School of Business (Hamilton, Ontario, Canada). His research interests focus on strategic leadership and corporate governance with a current focus on the role that strategic leaders play in shaping a firm’s stakeholder strategy. His articles have appeared in Business & Society, Journal of Business Venturing, Journal of Management, Journal of Management Studies, and Strategic Management Journal.

Appendix A.

Experimental Procedure.

Greeting (identical for all participants)
We are interested in your perceptions of leaders. Today, we are going to provide you with various information about a leader, such as information on the leader, the leader’s background, and recent press coverage of the leader. Afterward, you will be asked a series of questions relative to your perceptions and reactions about this leader. Please read the information we provide you as carefully as possible so that you can answer the questions that follow as honestly and truthfully as possible.
Description of the initial scenario (identical for all participants)
The following information provides recent news coverage of the coronavirus disease 2019 (COVID-19) crisis. The names of the country and its leader are disguised for privacy reasons. You will be asked to answer questions about what you read. When you are finished reading, you may continue to the next page.
Manipulated content (each participant was randomly assigned a condition)
Female Leader	Male Leader
News coverage from The Wall Street Journal:	News coverage from The Wall Street Journal:
Country X shut its borders to foreigners March 19. Two days later, Michele Jones delivered a televised address from her office—the first time since 1982 that an Oval Office-style speech had been given—announcing a coronavirus response alert plan involving four stages, with a full lockdown being Level 4.	Country X shut its borders to foreigners March 19. Two days later, Michael Jones delivered a televised address from his office—the first time since 1982 that an Oval Office-style speech had been given—announcing a coronavirus response alert plan involving four stages, with a full lockdown being Level 4.
A group of influential leaders got on the phone with her the following day to urge moving to Level 4. “We were hugely worried about what was happening in Italy and Spain,” said one of them, CEO Tindell, founder of the Warehouse, Country’s X largest retailer. “If we didn’t shut down quickly enough, the pain was going to go on for a very long time,” he said in a phone interview. “It’s inevitable that we will have to shut down anyway, so we would rather it be sharp and short.”	A group of influential leaders got on the phone with him the following day to urge moving to Level 4. “We were hugely worried about what was happening in Italy and Spain,” said one of them, CEO Tindell, founder of the Warehouse, Country’s X largest retailer. “If we didn’t shut down quickly enough, the pain was going to go on for a very long time,” he said in a phone interview. “It’s inevitable that we will have to shut down anyway, so we would rather it be sharp and short.”
On March 23, a Monday, Jones delivered another statement and gave the country 48 hours to prepare for a Level 4 lockdown. “We currently have 102 cases,” she said. “But so did Italy once.” From that Wednesday night, everyone had to stay at home for 4 weeks unless they worked in an essential job, such as health care, or were going to the supermarket or exercising near their home.	On March 23, a Monday, Jones delivered another statement and gave the country 48 hours to prepare for a Level 4 lockdown. “We currently have 102 cases,” he said. “But so did Italy once.” From that Wednesday night, everyone had to stay at home for 4 weeks unless they worked in an essential job, such as health care, or were going to the supermarket or exercising near their home.
A few hours before midnight, my phone sounded a siren as it delivered a text alert: “Act as if you have COVID-19. This will save lives,” it said, referring to the disease caused by the novel coronavirus. “Let’s all do our bit to unite against COVID-19.”	A few hours before midnight, my phone sounded a siren as it delivered a text alert: “Act as if you have COVID-19. This will save lives,” it said, referring to the disease caused by the novel coronavirus. “Let’s all do our bit to unite against COVID-19.”
From the earliest stages, Jones and her team have spoken in simple language: Stay home. Don’t have contact with anyone outside your household “bubble.” Be kind. We’re all in this together. She has usually done this from the podium of news conferences, where she has discussed everything from the price of potatoes to unemployment checks. But she also regularly gives updates and answers questions on Facebook, including one done while sitting at home—possibly on her couch — in a sweatshirt.	From the earliest stages, Jones and his team have spoken in simple language: Stay home. Don’t have contact with anyone outside your household “bubble.” Be kind. We’re all in this together. He has usually done this from the podium of news conferences, where he has discussed everything from the price of potatoes to unemployment checks. But he also regularly gives updates and answers questions on Facebook, including one done while sitting at home — possibly on his couch — in a sweatshirt.