Abstract
Despite the proliferation of leadership research in the past 75 years, investigating the ways in which women and men leaders enact and experience leadership continues to surface unanswered questions. Through the framework of selection, development, leadership style, and performance, we report gender-related findings from a broad survey of existing literature from the past three decades. Findings include differential rates of selection for women and men leaders; leader development considerations that vary by gender; evidence in favor of general similarities in leadership style (with noted exceptions) between women and men leaders; and similar performance outcomes between women and men leaders. The importance of context, be it job type, group composition, organizational culture, or industry/sector, was also revealed. Implications for practitioners and academics alike are offered throughout this report.
Introduction
Leadership in all its forms—organizational, political, military, and even sports—continues to command the general public’s attention. In academia, leadership research has also flourished, with a recent Web of Science search for “leadership” revealing more than 165,000 articles. Women in leadership have also come under study, with approximately 5% of current Web of Science articles addressing both leadership and gender related issues. With the nomination of the first female presidential candidate and the rapidly changing demographics of the United States, it is becoming even more important to understand the ways in which social identity variables, such as gender interacts with leadership. More specifically, determining whether or not qualitative differences exist between male and female leaders. The purpose of this report is to provide the beginnings of an answer to this question, through a systematic summary of the state of leadership research as it specifically pertains to women leaders, providing utility for both practitioners in the field, as well as researchers in academia.
Using a framework of selection, development, leadership style, and performance, we explore the experiences and perceptions of female leaders throughout all stages of the leadership process. This framework is a point of departure from previous work in this domain that often explains gender difference when it comes to leadership by situating the findings in the traditional diversity literature (Northouse, 2015). Prior research on women leaders focuses heavily on stereotypes, prejudice, and discrimination as prominent explanations for gender differences in the leadership experience (e.g., Eagly & Karau, 2002; Heilman, 1983; Rudman & Glick, 2001). These explanations provide an important departure point for our research especially with regard to understanding how individual-level factors affect women leaders. Our framework allows for a different understanding of gender differences by distilling gender and leadership research into four discrete domains: selection, development, style, and performance, which are most commonly assessed in the leadership field and in organizational life (Hogan, Curphy, & Hogan, 1994).
Using the framework outlined above, we examine how, and under what conditions, male and female leaders are selected, the ways in which they are developed (or would ideally be developed, given the appropriate resources), the leadership styles they are likely to enact, and their effectiveness. Viewing the existing gender-related leadership research through these four domains allows us to explore whether men and women are selected for leadership positions at different rates and whether they are given the same developmental opportunities. Furthermore, we can begin to address whether men and women demonstrate different leadership styles, and whether men or women generally perform better as leaders. The answers to these questions are complex, varied, and have been the focus of countless studies to date.
We provide a recent view of the field as it relates to women and leadership and expand on Eagly and Johnson’s (1990) meta-analysis on leadership behavior by offering additional dimensions of analysis (i.e., selection, development, and performance). Due to the depth and comprehensiveness of Eagly and Johnson’s (1990) review, studies published before 1990 were not included in our report. Thus, this survey of the literature has two aims: to document the recent findings on leadership as it relates to women and to provide a new framework in which to look at and understand the experiences of women leaders.
Selection
Our exploration into women and leadership begins with a review of the public and scientific discourse devoted to leader selection, the first tenet of our leadership frame. As selection is the first phase of the leadership process, it is an appropriate departure point for our report. To that end, this section will review the recent literature on gender and leader selection in order to determine how often women are selected for positions of leadership, the conditions surrounding their appointments, and future areas of exploration within this domain.
Is There a Gender Difference in Selection Rates for Positions of Senior Leadership?
Despite the fact that women make up nearly half of the U. S. workforce (47%; U.S. Bureau of Labor Statistics, 2011), men hold the vast majority of leadership positions in the corporate and political arenas (Catalyst, 2016; Center for American Women and Politics, 2016; Eagly & Carli, 2007). While women have had success obtaining supervisory and middle management positions in rates proportional to their representation in the workforce, access to the upper echelons of leadership still remains relatively illusive (U.S. Bureau of Labor Statistics, 2011). In S&P 500 companies, women occupy less than a third of executive- and senior-level positions (25.1%), less than a fifth of all board seats (19.9%), and less than 5% of all chief executive officers (4.2%; Catalyst, 2016). When it comes to political leadership, women are also underrepresented in elected office. To date, the United States has never elected a woman to the presidential office. Of the 535 seats in the U.S. Congress, women occupy only 104 seats (19.4%; Center for American Women and Politics, 2016). Furthermore, only 24.6% of state legislators, 12% of state governors, and 18.4% of mayors are women (Center for American Women and Politics, 2016).
The underrepresentation of women in senior leadership positions is not limited to the United States alone. Gender inequality when it comes to leader selection is well documented worldwide (e.g., Grant Thornton, 2016; Inter-Parliamentary Union, 2016). Women hold less than a quarter of senior leadership positions in corporations across the globe, with a third of businesses having no women at all in senior roles (Grant Thornton, 2016). Internationally, women also are underrepresented in political office. Currently, only 19 countries have a female head of state or government (United Nations, 2016). Worldwide, less than a quarter of parliamentarians are female (22.6%; Inter-Parliamentary Union, 2016).
Taken together, these statistics demonstrate that selection rates for senior leadership positions do differ for men and women, both in the United States, and across the globe. When it comes to being selected for high-level leadership positions, men have a distinct advantage.
Why Are Women Underrepresented in Elite Leadership Roles?
In recent decades, a diverse range of explanations have been offered to explain the leadership gap between men and women. In particular, prior research has focused on stereotyping, gender bias, and discrimination against women as main contributors to the gender gap at the top levels (e.g., Eagly & Karau, 2002; Garcia-Retamero & López-Zafra, 2006; Heilman, 2001). Research has also explored the possibility of a pipeline problem that has resulted in a scarcity of qualified women to fill senior leadership positions (e.g., Eagly & Carli, 2007; Ibarra, Ely, & Kolb, 2013). We will discuss both sets of explanations in turn.
Stereotyping/Gender Bias/Discrimination
Gender stereotypes are culturally shared beliefs that dictate expectations about how women and men are and how they ought to behave (Burgess & Borgida, 1999; Eagly & Karau, 2002). Thus, stereotypes can be both descriptive and prescriptive in nature. When it comes to gender norms, descriptive stereotypes dictate that women are communal and warm (Fiske, Cuddy, & Glick, 2007; Heilman, 2001). Whereas, prescriptive stereotypes prescribe that women should be communal (Eagly & Karau, 2002). On the other hand, men are often stereotyped with agentic characteristics such as being confident and assertive (Huddy & Terkildsen, 1993). Incidentally, agentic characteristics are often seen as requisite traits for leadership (Dodge, Gilroy, & Fenzel, 1995; Koenig, Eagly, Mitchell, & Ristikari, 2011; Rudman, Moss-Racusin, Phelan, & Nauts, 2012). However, research has shown that women who behave agentically can be subjected to denigration and backlash for violating the prescriptive stereotype of being communal (Phelan, Moss-Racusin, & Rudman, 2008; Rudman & Glick, 2001). Therefore, women leaders must consider how to exhibit the agentic characteristics deemed necessary for leadership without violating gender stereotypes.
Descriptive and prescriptive stereotypes are pervasive and can often lead to biased judgments (Fiske, 1998; Heilman, 2001). For this reason, stereotypes are at the heart of several theories and frameworks seeking to explain the gender gap in leader selection rates. Most notably, role congruity theory (RCT; Eagly & Karau, 2002), which builds on several prior theories including Schein’s (1973) think-manager, think-male paradigm, Heilman’s (1983) lack-of-fit model, and Eagly’s (1987) social role theory. Together, these frameworks have greatly influenced the work that has been done so far with respect to understanding how gender bias and stereotypes can significantly alter the perception and evaluation of female candidates aspiring to leadership positions.
Schein’s (1973) think manager–think male paradigm, posits that successful leaders are perceived as possessing characteristics, attitudes, and temperaments that are closely aligned with the global masculine stereotype (e.g., competent, verbal skills, determination, and industriousness; Sczesny, 2003). In a replication study of Schein’s original work, Heilman, Block, and Martell (1995) demonstrated that women were depicted as being less similar to successful managers than men. Furthermore, Koenig and colleagues (2011) conducted a meta-analysis that examined the extent to which stereotypes of leaders are culturally masculine. They found that across studies, prototypical leader traits had a stronger correlation with masculine traits than with feminine traits. Other studies in this domain have also shown that both genders show a preference for a masculine ideal of leadership (e.g., Schein, 2001). This effect is even more prevalent in organizations with traditions of male leadership (Bruckmüller & Branscombe, 2010; Phillips, 2005). Taken together, the research suggests that on the whole, the role of manager is often associated with masculine traits and behaviors.
While the aforementioned studies focused primarily on managers, recent literature has demonstrated that when we think—leader we also think—male (e.g., Jackson, Engstrom, & Emmers-Sommer, 2007). This poses a challenge for female candidates who do not fit the masculine construal of leadership. The incongruity between the female gender role and the leadership role can result in prejudice against women. Because the traditional perception of what leadership looks like is based on masculine-oriented concepts, women are less likely to measure up to this ideal. This mismatch creates a “lack-of-fit” (Heilman, 1983, 2001). Similarly, RCT (Eagly & Karau, 2002) also promotes this idea that stereotypical attributes associated with women tend to be inconsistent with the attributes required for senior leadership positions. This incompatibility can fuel the perception that women are less qualified for top leadership positions and lead to discrimination against women seeking senior leadership positions.
Pipeline Problem
Most of the aforementioned literature has focused on the demand side of the equation, implying that gender bias and prejudice against women candidates lowers the demand for female leadership. However, other researchers have pointed to the scarcity of qualified women as a contributing factor to the gender gap in leader selection. This has been coined in the literature as the pipeline problem where there are not enough women with the appropriate level of education and work experience to select from for senior leadership positions (Eagly & Carli, 2007). On the whole, women have been earning more degrees than men since 2000, with 57% of all bachelor’s degrees, 59.9% of all master’s degrees, and 51.8% of all doctorate degrees being awarded to women (National Center for Education Statistics, 2016). Women are, however, underrepresented in MBA programs, with only about one third of all MBAs being granted to women (Catalyst, 2009).However, even this level of representation does not continue when it comes to senior leadership positions. Given these statistics, there is minimal support for the notion that women are less educated than men.
There is, however, some support for the notion that women have less experience than their male counterparts. Research has shown that women tend to have more career interruptions than men, due largely to having greater domestic responsibilities (Bianchi, Milkie, Sayer, & Robinson, 2000; Bowles & McGinn, 2005; Eagly & Carli, 2007). However, given that approximately half of all middle managers are women (U.S. Bureau of Labor Statistics, 2015), this may provide evidence that women are being given the necessary opportunities to get relevant management experience before being considered for senior leadership positions. In any case, despite the prevalence of women in middle management positions, there may still be corporate policies and practices, or unconscious stereotypes and biases that prevent women from accessing the top leadership ranks. Taken together, these findings would suggest that women are in the pipeline, but that there are systematic barriers in place that prevent women from reaching senior leadership positions.
Under What Conditions Are Women Selected for Positions of leadership?
The proportion of women in senior leadership positions varies considerably according to industry, which would indicate that leader selection is highly context specific. Women tend to be overrepresented in services industries like education, healthcare, government, and hospitality (Bowles & McGinn, 2005; Eagly, Karau, & Makhijani, 1995), while leadership teams in more traditional industries (e.g., manufacturing, construction, and financial services) are predominantly men (Oakley, 2000). In addition, when women are selected for leadership positions, many of these roles are concentrated in the support function of the organization (as opposed to the core operations of a business), in less visible positions, and in roles that have less responsibility (Bowles & McGinn, 2005). This body of research would suggest that both industry and role moderates the relationship between gender and leader selection and that the overarching context has an influence on selection decisions.
Research has also shown that women are more likely to be selected for senior leadership roles when the position is associated with a state of crisis or a high risk of failure. This phenomenon has been coined the glass cliff (Ryan & Haslam, 2005). Using archival records from 100 companies on the London Stock Exchange, Ryan and Haslam (2005) found that women were more likely to be appointed to the board of failing companies, whereas men were more likely to be selected for board positions in succeeding companies. Thus, an interaction exists between company’s performance and candidate’s gender such that in successful companies, a think-leader think-male bias emerges, whereas in an unsuccessful company, a think-crisis think-female bias occurs (Ryan, Haslam, Hersby, & Bongiorno, 2011). Other studies have demonstrated this phenomenon as well (e.g., Haslam & Ryan, 2008; Ryan, Haslam, & Kulich, 2010). Bruckmüller and Branscombe (2010) also found that women are more likely to rise to positions of organizational leadership in times of crisis than in times of success. Furthermore, these researchers found that in successful companies, agentic characteristics mattered more for leader selection, whereas in times of crisis, interpersonal attributes were deemed more important.
Researchers have offered several hypotheses as to why women are preferentially selected for leadership positions in problematic organizational circumstances. The first explanation positions women as a symbolic antidote to the current situation (Ryan & Haslam, 2007). Companies who are experiencing economic challenges often recognize the necessity of change. Since leadership is a transformational factor affecting all aspects of the organization (Burke, 2014), a change in leadership could provide the necessary perturbation to a system. If a man was previously in charge, having a women successor may be an easy way to signal to stakeholders that the company is headed in a new direction. Therefore, women are put in positions of power as a remedy to the current failed situation. Unfortunately, these positions are associated with greater risk, more criticism, and a higher likelihood of failure (Ryan et al., 2011).
Researchers have also explored the possibility of women being perceived as more suitable for these types of positions that require advance interpersonal skills necessary to manage an organization in crisis (Bruckmüller & Branscombe, 2010; Sczesny, Bosak, Neff, & Schyns, 2004). Indeed in leaderless groups, women are more likely to emerge as leaders when the task is interpersonal in nature versus when there is more of a task-focused due to the perception that they are more communal in nature (Eagly & Karau, 1991). Thus, women may be offered these challenging positions to (1) give a company a fresh direction and (2) because they are perceived as being more skilled than men at handling an organization in crisis.
Selection Summary
In terms of selection, findings indicate that men are disproportionately represented in organizational and political leadership positions across the globe. While there is scant evidence to support that there is a pipeline problem when it comes to having qualified women in the hiring pool, there is far more support for the notion that there is a lower demand for female leadership and that the glass ceiling is in full effect. The most prominent explanations for the leadership gap centers on gender stereotypes and biases that result in prejudice and discrimination against women aspiring to be senior leaders.
Current selection methods may disadvantage women because they often lack accountability, which allows the opportunity for gender bias to influence the decision-making process (Powell & Graves, 2003). Therefore, future research on gender and leader selection could explore the ways in which organizations can make leadership decisions that are less biased with respect to gender. One way to do so may be to concentrate on traits that have been demonstrated to correlate with effective leadership such as emotional intelligence, self-awareness, and learning agility (de Meuse, Dai, & Hallenbeck, 2010; Palmer, Walls, Burgess, & Stough, 2001; van Velsor, Taylor, & Leslie, 1993). In this way, decision makers can help ensure that women have equal opportunities to attain significant and important leadership positions.
Development
The second tenet of our framework is development. The terms leadership development and leader development are often confused or used interchangeably. However, the difference between the two is quite significant (Day, 2001). Leadership development occurs at the system level and is often a combination of multiple methods used to develop leaders. On the other hand, leader development is an individual-level process that often parallels an individual’s development into being an adult (Day, Fleenor, Atwater, Sturm, & McKee, 2014; Day, Harrison, & Halpin, 2009). While the development of women from a system standpoint is important to the current exploration of gender and leadership, this section will focus primarily on the unique experiences women have when they take up the role of leader. Therefore, when we discuss development, we will emphasize the individual level of analysis and focus on leader development.
Why Is Leader Development Important?
In general, the purpose of leader development is to develop an individual’s skills and capacities such as interpersonal awareness and job-related skills. This is often accomplished through a careful assessment of strengths and weakness, self-reflection, and coaching (Day, 2001). High-potential employees are often targeted for leader development because they are critical to an organization’s current and future success given that they may soon advance into positions of leadership. For high potential women, leader development is especially important because it provides the tools necessary to succeed and ensures that individual performance is maintained at a high level. In addition to the benefits on an individual level, leader development can also help an individual become more effective in leading others, which benefits the organization as well (Day, 2001). By increasing an individual’s self-awareness, self-regulation, and self-motivation, leader development can be beneficial to the entire organization. Therefore, leader development is an important component of organizational success.
What Do We Know About the Framework and Methodology Used for Women’s Leader Development?
Leader development is a highly individualized process that is tailored specifically to the individual. Accordingly, individual differences such as gender will likely influence all aspects of the leader development experience. Each gender has a unique set of realities and ways of knowing and understanding the world (Vinnicomb & Singh, 2002). Therefore, the experiences of women in the workplace are likely to be different from the experiences of men. This is especially true in industries and work environments that have traditionally been imbalanced when it comes to gender representation (e.g., law firms, financial industry, manufacturing). The experiences of women are shaped and colored by their subjective reality about what it means to operate in a historically male-dominated space (Ely, Ibarra, & Kolb, 2011; Vinnicomb & Singh, 2002). In this way, the biases against women and the perceptions that women have of themselves greatly influence the leader development process.
To date, research has been fairly limited with respect to developing overarching leader development frameworks or models geared toward women. There are two key issues contributing to the lack of research that exists today on how to effectively develop women leaders. First, most publications to date have focused on recommendations that organizations should create and implement in order to develop their women leaders (Ely et al., 2011; Hopkins, O’Neil, Passarelli, & Bilimoria, 2008) as opposed to actually implementing and evaluating said recommendations in an organizational setting. Therefore, little is known about the true effectiveness of frameworks and methods for developing women leaders (Ely et al., 201l; Hopkins et al., 2008).
Despite this, Ely and colleagues (2011) created a framework outlining how the leader development process for women is unique and why leader development methods should be tailored to them. Their framework proposed two related concepts that they identified as being critical to understanding gender differences regarding leadership: (1) second-generation bias and (2) identity work.
Second-generation bias stifles the advancement of women due to pervasive beliefs about gender as well as patterns of interaction, informal norms, and workplace structures that inadvertently favor men (Ely et al., 2011; Sturm, 2001). Due to this set of beliefs and related phenomena, second generation bias can result in subtler and more complex forms of workplace inequity and greatly affect the experiences of women. This is especially true in male-dominated domains and organizations with a strong tradition of male leadership. With more concerted research efforts in the near future, the effect of second-generation bias may lessen over time and reduce the necessity of having separate leader development programs for men and women employees (Ely et al., 2011). The second concept in Ely and colleagues’ (2011) framework, identity work, can be thought of as a set of processes that serve to construct a sense of identity through active exploration. For women, the development of a leader identity can be challenging because they must display the characteristics necessary for leadership without violating prescriptive norms about their gender.
Ely and her colleagues (2011) postulate that the experiences of women in the workplace are riddled with systematic gender biases that disadvantage women. They suggest that leader development efforts should recognize and integrate the unique experiences that women have due to their gender identity and the systematic bias that occurs in workplaces that have traditionally favored men. Failing to do so could reduce the impact and effectiveness of leader development for women (Ely et al., 2011). While this framework provides guidance about what a leader development program should look like, it has not been empirically tested like much of the research that precedes it. Therefore, testing Ely et al.’s (2011) theoretical framework in an organizational setting would be a valuable starting point for understanding what is necessary to develop women leaders to reach their maximum potential.
How Are Common Leader Development Methods Currently Tailored to Women?
As stated previously, the majority of publications to date have focused on adaptations to common leader development methods that organizations should implement in order to successfully develop their women leaders (Ely et al., 2011; Hopkins et al., 2008). The purpose of this section is to showcase common leader development methods and recommendations tailored to women’s leader development, specifically multirater feedback, executive coaching, mentorship, and networking.
Multirater Feedback
Multirater feedback has been used with leaders as a mechanism to enhance self-awareness by gathering feedback from superiors, peers, direct reports, and key stakeholders. In particular, multirater feedback provides a unique opportunity for women to gain honest feedback that they may not have otherwise received. Even more, multirater feedback can highlight “blindspots” or misalignment in self and rater perceptions that can serve to help tailor leader development plans. Being aware of “blindspots” is arguably more important for women than men because ratings of women tend to be biased by raters’ varying perceptions of what constitutes a “good female leader.” However, these “blindspots” could actually be the result of contradictory feedback about their performance. Ely et al. (2011) refer to these messages as a double bind, in which women are subjected to a double standard and are expected to balance the attributes of being both female and a leader.
Given that multirater feedback and other leader development assessments may be biased due to gender stereotypes, the literature suggests that organizations can best support their women leaders by creating a culture that not only welcomes and values developmental feedback but also educates all raters (including women themselves) on double-bind biases (Ely et al., 2011; Hopkins et al., 2008). Raising awareness of the double bind can give women the freedom to disregard contradictory messages and focus their energy on their work instead. For other raters, raising awareness could be the first step toward addressing the issue.
It should be noted that the research on multirater feedback shows that providing the feedback by itself does not affect performance. In fact, research suggests that coaching paired with conscious reflection is the best combination to provide the self-awareness needed to improve performance (Luthans & Peterson, 2003; Seifert, Yukl, & McDonald, 2003; Smither, London, Flautt, Vargas, & Kucine, 2003).
Executive Coaching
In addition to using multirater feedback as a development tool, leaders may also meet with an executive coach. An executive coach works to collaborate with the leader to design a development plan that leverages the individual’s strengths and corrects any identified weaknesses. One challenge that women often have with being coached is a lack of consideration of the need to balance work and their personal lives. While men also face this challenge, women are seen consistently as the primary caretakers of the home and thus the burden of familial responsibilities falls disproportionately on women (Gordon & Whelan, 1998). Therefore, women may need to sacrifice the speed or trajectory of their career in order to balance both their professional and personal lives.
Organizations that use executive coaching as a leader development method for their women leaders should encourage executive coaches to design developmental plans that are tailored to women’s unique experiences in the workplace (Hopkins et al., 2008). In other words, they need to be mindful of the intersection of work and life that may affect the direction of a woman’s career and substance of coaching.
Mentorship
Though mentorship takes on many different forms and thus has many different definitions, for the purposes of this article we define mentorship as a relationship between two people in which the mentor is typically experienced and can provide “technical and psychological support” to a less experienced and high potential mentee (R. J. Burke & McKeen, 1990; Kram & Isabella, 1985, p. 111). With respect to developing women leaders through mentorship, women are often paired with male mentors since men hold the majority of leadership positions. This can be problematic for high potential women because gender and power dynamics stemming from second-generation bias could influence the effectiveness of these relationships.
To combat some of these challenges, Hopkins and her colleagues in a nonempirical study (2008) suggest encouraging women leaders to have both male and female mentors. In order for this to be successful, organizations are advised to create formal mentoring programs and strategically pair women with senior leaders of both genders. Furthermore, because there are still so few women in leadership roles, women must actively engage as both mentor and mentee (Hopkins et al., 2008). Ultimately, mentorship offers little to no tangible reward to the mentor, which may affect sustainability. As a result, sponsorship as opposed to mentorship may be a more viable option for developing women leaders. Sponsorship is similar to mentorship with one critical difference—there is “skin in the game” such that sponsors are often putting their professional reputation, time, and other resources toward developing the sponsee and thus is more likely to be invested in his or her mentee’s success (Berhane, 2015)
Networking
Finally, networks are instrumental in leadership and leader development because membership in certain networks often provides opportunities in terms of relationships, work assignments, and/or promotions, as well as creating a social space to earn professional recognition and credibility (Ely et al., 2011). These activities are important components of career development and advancement. However, not all networks in the workplace are created equally (Ibarra, 1997). In fact, an empirical study looking at the qualitative differences between women and men’s networks suggest that each serve different purposes (Ibarra, 1992). Specifically, men’s networks tend to be used for practical reasons related to their job and or career, whereas women’s networks tend to be used for relationship building and emotional support. While both types of networks could be valuable for leader development in organizations, it appears that women are generally isolated from or have limited access to the male-dominated networks that might result in advancement (Ibarra, 1992).
For organizations that have a desire to develop their women leaders, it is important to educate men and women about the different types of networks and provide recommendations on how to access seemingly exclusive networks. Hopkins et al. (2008) suggest that illustrating the potential benefits of membership in both male and female networks could help women as they grow into positions of leadership.
Development Summary
Overall, the recommendations for successful implementation of leader development programs for women encourage a development process rooted in awareness and integration of women’s subjective experience (Vinnicomb & Singh, 2002) as well as an understanding of the strong impact of second-generation bias. Having a theoretical understanding of gender roles, stereotypes, and biases can help illuminate some of the challenges that many women face as they develop as leaders. Given that the majority of literature on leader development for women is not grounded in empirical research, it is important to establish the applicability and relevance of the recommendations put forth by researchers in this domain. To this end, the authors encourage practitioners and scholars alike to implement and empirically test the aforementioned recommendations with women leaders in real organizations.
Style
In addition to selection and development, leadership style comprises another broad category in which research has burgeoned and thus is the third tenet of our framework. With regard to gender issues, this body of work has sought to examine whether the ways in which women leaders take up their role reliably differ from those of men leaders, across a range of theoretical models.
What Do We Know About Leader Style?
Since the early days of leadership research, scholars have been interested in investigating leadership “style,” which was perhaps first coined by Blake and Mouton (1964), and defined as the “relatively stable” set of behaviors exhibited by a leader (Eagly & Johannesen-Schmidt, 2001, p. 781). Often, style is described and depicted as falling somewhere between two intersecting axes, one of which is related to task, and the other to interactions with people. Blake and Mouton (1964) labeled these axes “concern for production” and “concern for people,” while earlier, Fleishman (1953) described them as “initiation of structure” and “consideration,” and even contingency theorists such as Hersey and Blanchard (1969) and Fiedler (1967) examined these two aspects of leader style. House (1971) and Vroom (1964), among others, offered a different take on the behavioral manifestations of leaders, coding leaders as acting in more or less participative or directive ways as they lead followers. Thus, the degree to which leaders focus on people versus task and the ways in which leaders engage their followers have been threads that run throughout past leadership work.
Examining leadership behavior through the lens of transformational, transactional, and laissez-faire styles has also become popular in the past few decades, where follower interactions are key to a leader’s style (Bass, 1990). Transformational leaders are characterized as inspiring, motivating, being attentive to and intellectually challenging their followers, as well as engendering passion for the group or organizational mission, whereas transactional leaders are described as contractual, corrective, and critical in their interactions with employees (Bass, 1990). Other recent bodies of work, authentic leadership (see, e.g., Avolio, Gardner, Walumbwa, Luthans, & May, 2004) and servant leadership (Greenleaf, 1977; Liden, Wayne, Liao, & Meuser, 2014), have also documented the relationship between leaders and their followers.
All these frameworks have undergirded the work that has been done with regard to gender differences in leadership style. Beyond laying the theoretical foundations for the field and shaping the work that continues to be done, the content of these theories can be seen to map onto gender stereotypes, roles, and expectations in potentially different ways for women versus men leaders.
How Do Female Leaders Lead, and Does It Differ From the Way Male Leaders Lead?
The research reviewed above summarizes formal leadership theories that have been offered as one lens through which leader style can be understood. Before delving into those theories as they relate to gender, it is important to note that there may be broad differences in expectations for the types of chronic and daily behaviors that men and women display in the workplace, due to descriptive and prescriptive stereotypes (Heilman, 1983; Eagly & Karau, 2002). It is possible that day-to-day behaviors appear to differ, or actually do differ, for men and women given these expectations (ingrained in both the leader and his or her followers), and also that perceptions of a particular leader style may vary according to the gender of the leader enacting those behaviors (Ayman, Korabik, & Morris, 2009). There may also be other behavioral differences that manifest between men and women, such as conflict style (e.g., Brewer, Mitchell, & Weber, 2002; Korabik, Baril, & Watson, 1993), organizational citizenship behaviors (Kidder, 2002), or basic interpersonal interaction styles, that do not neatly load into these broader theories of formal leadership style, but do influence perceptions of women leaders.
For example, women may be more likely to engage in a relational approach to work than men (Matthew, Buontempo, & Block, 2013) and also might be more apt to balance dominant and affiliative interaction styles in a selection context, as opposed to men (who favor a dominant approach; Luxen, 2005). Women are also less likely to advocate for themselves, less likely to ask for what they want, and less likely to initiate negotiations (Bowles, Babcock, & Lai, 2007; Kray & Thompson, 2004). Findings have also indicated that women are expected to present themselves in a more modest way (Rudman & Glick, 1999) and are more effective in male-dominated groups when they demonstrate hedging and self-doubt in their speech (Carli, 1990) and are self-effacing in their behavior (Rudman, 1998). Furthermore, women are less likely to self-promote (Bowles & McGinn, 2005), which has been shown to be a critical component of professional success, contributing indirectly to hiring and promotion decisions (Kacmar, Delery, & Ferris, 1992; Rudman & Glick, 2001; Stevens & Kristof, 1995). Women are also more likely to shy away from formal leadership roles and instead opt for informal roles like facilitator or organizer (Fletcher, 2001).
Women leaders are also rated as better at developing others (Cavallo & Brienza, 2006), as well as inspiring, and motivating others, building relationships, and collaboration and teamwork than their male peers (Zenger & Folkman, 2012). Furthermore, Kidder (2002) found evidence for various facets of gender identity and job type relating to differential rates and types of organizational citizenship behaviors, and Cavallo and Brienza (2006) reported that women were rated higher than men by both their peers and direct reports in service orientation. Conflict style also appears to vary by gender, with findings indicative of associations between feminine gender identity and an avoidant conflict style, masculine identity and a dominant style, and androgynous identity with an integrative approach to conflict (Brewer et al., 2002). Finally, women on Boards of Directors appear to approach decisions through “complex moral reasoning” (Bart & McQueen, 2013, p. 97) and by considering multiple viewpoints, accounting for varying interests of different stakeholder groups.
In each of these cases, women leaders’ behaviors may be influenced by innate sex-related differences, gender role identity, and/or societal and organizational expectations. As Eagly and Johannesen-Schmidt (2001) state,
. . . female leaders’ choices are constrained by threats from two directions: Conforming to their gender role can produce a failure to meet the requirements of their leader role, and conforming to their leader role can produce a failure to meet the requirements of their gender role” (p. 786)
Therefore, one overarching caveat in the examination of gender as it relates to leadership style is the ubiquitous influence of gender stereotypes on expectations for, perceptions of, and manifestations of, leader behavior, from more micro and informal behaviors like organizational citizenship behaviors, to more macro behaviors like leader style. With that said, delving more deeply into possible gender differences across formal leadership theories also paints an interesting picture.
Interpersonal Style and Task Style
Overall, findings have indicated that men and women do not consistently and reliably differ in their enactment of interpersonal versus task style in leadership roles (van Engen & Willemsen, 2004). In an early meta-analysis, Eagly and Johnson (1990) concluded that there was a small tendency for women leaders to be more interpersonally oriented than men. However, this finding was moderated by the setting of the study, wherein this finding held up in lab settings, but not in the field. Conversely, a later meta-analysis indicated that women trended toward interpersonal versus task leadership in the field, but not in the lab (van Engen & Willemsen, 2004). Thus, no robust pattern of women enacting an interpersonal style in the workplace has emerged from these findings, and changes in broader conceptualizations of leadership (e.g., incorporating more feminine or androgynous behaviors) may also contribute to equivocal findings (van Engen & Willemsen, 2004).
Potential moderators of interpersonal versus task style have been explored, with Gardiner and Tiggemann (1999) reporting that there were no differences in interpersonal leadership style in male-dominated industries, but in female-dominated industries, women displayed higher rates of interpersonal leadership style than men. Van Engen and Willemsen (2004) and Eagly, Karau, and Johnson (1992) also found evidence of interpersonal and task style differences in men and women across industries. Women leaders in academia and business settings were found to be more task-oriented than male leaders, whereas women leaders in government and educational institutions were less task-oriented than male leaders (van Engen & Willemsen, 2004). However, Eagly et al. (1992) reported that women leaders in an educational setting (i.e., school principals) were more likely to manifest task-oriented leadership than male leaders. In line with the finding of context moderating behavior, Eagly and Johnson (1990) reported that when women are completing female-typed jobs, they are seen as more task-oriented than males, whereas when men are completing male-typed jobs, they are seen as more task-oriented than women. Finally, one other potential moderator of behavior may be task type, in that there is evidence that social tasks tend to correlate with the emergence of female leaders, whereas work-related tasks tend to correlate with the emergence of male leaders (Eagly & Karau, 1991). Arguably, different types of tasks may also elicit different styles of leader behavior. Thus, the overarching context, as well as the type of task, seems to impinge on the impressions men and women make in the workplace with regard to leadership style.
In sum, this particular theoretical frame has produced varied and complex findings, and there does not seem to be a clear prescription for how a woman will act in a position of leadership, in relation to task and interpersonal style.
Autocratic/Directive and Democratic/Participative
In contrast to the aforementioned findings, a more robust, though small, difference between men and women leaders has been found in relation to the level of participation afforded to followers with men leaders operating in a more autocratic way than women leaders, and conversely, women leaders enacting a more democratic style than men leaders (Eagly et al., 1992; Eagly & Johnson, 1990; van Engen & Willemsen, 2004; for an exception, see Cuadrado, Navas, Molero, Ferrer, & Morales, 2012). In general, this finding appears to hold up across both lab and field studies. In a related thread of work, Adams and Funk (2012) did find that women directors are more benevolent and universally concerned than male directors, in line with democratic or participative tendencies.
Research on gender differences in democratic versus autocratic style has not proliferated in the same way as research on other styles, perhaps due to construct ambiguity (Gastil, 1994), the more limited narrow conceptualization of style offered by this approach (Eagly & Johannesen-Schmidt, 2001; Field, 1979), or the complex relationship between democratic versus autocratic style and effectiveness (Foels, Driskell, Mullen, & Salas, 2000; Vecchio, 2002).
Transactional, Transformational, Laissez-Faire
Conversely, research on transactional and transformational styles has grown and expanded in the past few decades, with relevant implications for women leaders. In a meta-analysis of organizational leaders across sectors and industries, Eagly, Johannesen-Schmidt, and van Engen (2003) found that women display higher rates of transformational leadership than men (with the exception of the “idealized influence: behavior” subscale). Furthermore, women leaders display higher scores on the “contingent reward” subscale, a component of transactional leadership, than men, which is seen as being one of the more effective components of transactional leadership. In comparison with women, men leaders were more likely to display two other components of transactional leadership (management by exception, active and passive) and laissez-faire leadership, though it is notable that these effect sizes were small, yet significant. Further bolstering these findings, Groves (2005) reported that women leaders were rated more highly than men leaders on charisma, a key component of transformational leadership.
Interestingly, while others have reported similar findings (e.g., Carless, 1998; Druskat, 1994; van Engen & Willemsen, 2004), Ayman et al. (2009) found that there was an interplay between the gender of followers and the rated effectiveness of the transformational style. Similar to other findings, women were more likely to manifest aspects of transformational leadership than men. However, the higher female leaders scored on the individualized consideration and intellectual stimulation subscales of transformational leadership, the less effective their male (but not female) subordinates rated these women leaders (Ayman et al., 2009). Therefore, even when manifesting the same behaviors as men, or when manifesting behaviors hypothesized to contribute to effectiveness, women leaders may be perceived differently (Cuadrado et al., 2012; Eagly, 2013). Context, whether defined as the gender composition of a workgroup, or the gender type of industry, is clearly a key element to be considered in assessing transformational and transactional leadership styles (Antonakis, Avolio, & Sivasubramaniam, 2003; van Engen & Willemsen, 2004).
Authenticity
Authentic leadership has become more prominent in recent years, as organizations continue to recover from economic crises, corrupt leaders, and broader societal challenges (Walumbwa, Avolio, Gardner, Wernsing, & Peterson, 2007). Authentic leadership is defined as “authentic self-regulation processes, including internalized regulation, balanced processing of information, relational transparency, and authentic behavior” coupled with high ethical standards (Walumbwa et al., 2007, p. 92).
Notably, studies specifically examining gender differences in antecedents to, enactments of, and reactions to, authentic leadership are limited (Liu, Cutcher, & Grant, 2015). Furthermore, authentic leadership has been critiqued as being gendered, due to stereotypical expectations for women’s behavior (Fudman, 2015; Hopkins & O’Neil, 2015). As women are often caught in a “double-bind,” wherein they must manifest competence but soften those competent behaviors with warmth (Cuddy, Glick, & Beninger, 2011), enacting truly authentic leadership, without self-monitoring to manage stereotypes, may be challenging. Furthermore, practicing authentic leadership may prove problematic for groups that have not traditionally had access to leadership, as followers’ reduced trust in nontraditional leaders may negatively affect identification with the leader (Eagly, 2005). Finally, as leaders negotiate their authority with followers, authentic leadership depends on followers’ reactions and perceptions of a given leader. These are often strongly influenced by deeply ingrained stereotypes, which may actually in turn affect leader behavior, likely in line with traditional gender roles (Liu et al., 2015).
In sum, while an interesting and important line of scientific inquiry, the exploration of gender differences in authentic leadership is nascent, with a dearth of empirical data specifically oriented toward examining gender effects, as well as strident critiques of the possibly gendered construction of the theory.
Is There an Ideal Leadership Style?
Though research on style continues to proliferate and contributes to our understanding of leadership, there have been critiques of this body of work. Researchers have asserted that there is no “one style fits all” solution to leadership issues and that the efficacy of various styles is contextual (Eagly, 2013; Gastil, 1994). Furthermore, some leadership styles might be challenging to define and enact (e.g., Ibarra, 2015). Also, the historical trajectory of leadership research has shown us that in seeking the “holy grail” of the perfect leader, whether it is trait based, behaviorally based, or through documenting behavioral proscriptions given a particular situation or context, we still have not managed to capture why and how the best leaders are successful.
It is also important to note that the style enacted by women leaders may be confounded with gender normative expectations on the part of a woman leader, or those around her (Eagly & Johannesen-Schmidt, 2001). It is challenging to determine whether women and men innately differ in leadership style, or if differential behaviors in the workplace stem from ingrained stereotypes and gender role definitions.
Style Summary
Taken together, findings indicate that women leaders tend to be more likely to manifest a democratic style and elements of transformational leadership than their male peers. Findings relating to other styles have been equivocal (i.e., task vs. interpersonal style) or are not yet robust enough for an empirical comparison by gender (i.e., authentic leadership). Despite the findings that have reached significance, as Eagly and Carli (2007) note “differences in men’s and women’s styles generally appear as mild shading, with considerable overlap” (p. 127) and some researchers offer the interpretation that gender differences are overemphasized and overstated (e.g., Hyde, 2005). Furthermore, while gender differences have been found in some underlying attitudes, values, and abilities (see Eagly, 2013) a clear mediating path between these individual differences and leadership style has yet to be established, thus, it is not clear what may be driving the differences that are found.
In moving to the examination of performance, we also see, on the whole, highly overlapping distributions between female and male leaders.
Performance
Thus far, this article has discussed gender differences in leadership with regard to selection, development, and leadership style. The final tenet of our framework is performance. One critical question remains, do we observe differences in effectiveness between male and female leaders? This question is relevant to researchers and practitioners alike who wish to determine whether increasing the number of women in positions of leadership will result in value-added for organizations. If it becomes clear that women have a “leadership advantage,” as some researchers argue (e.g., Eagly & Carli, 2003), there may be profound ramifications for the study and practice of leadership selection and development. For instance, if the leadership advantage perspective is valid, then why are men preferred over women for promotion to leadership positions? Furthermore, why are there so few development opportunities for female leaders? Even if research determines that men and women are equally competent with respect to leadership ability, these questions still hold, as restricting leadership opportunities for women ultimately results in a constrained pool of leadership talent from which organizations can choose.
For these reasons, it is clear that it is important to measure and document differences in leadership performance between genders. Additionally, if there is a difference, it is incumbent on researchers to investigate the process by which these differences operate and the conditions under which the differences are most apparent (e.g., given certain organizational cultures). To this end, the goal of this section is to review the most recent literature on gender and leadership performance. The key question is whether men or women are more effective leaders. If so, when and why?
In this section, we divide the gender and leadership performance literature into two main parts based on how performance is measured. The first research stream addresses effectiveness ratings as determined by the perceptions of subordinates, supervisors, peers, or the leaders themselves. The second stream pertains to measures of leader performance that are based on organizational outcomes, such as financial performance or other key performance indicators.
Is There a Gender Difference in Perceptions of Leader Effectiveness?
One of the most important tools among researchers and practitioners for measuring leader effectiveness is performance evaluations. These allow the leader to be evaluated by those who have detailed knowledge of both the leader and the business, and are therefore able to assess his or her performance from an insider perspective. From a research standpoint, this is valuable because it is often difficult to determine a priori the standards to which leaders from different types of organizations should be held. At the same time, performance evaluations can be highly subjective and prone to a number of environmental circumstances (Eagly et al., 1995), which raises the question about whether they should be used as an instrument to determine whether there are gender differences in leadership performance. However, leadership effectiveness is the result of a number of factors, including followers’ expectations and prejudices (Eagly & Chin, 2010). Furthermore, part of a leader’s task is to establish commitment from followers, regardless of their preconceived notions and biases. We would argue that a leader must be perceived as effective to be effective in practice; therefore, while performance perceptions may not be entirely reflective of a leader’s abilities, per se, perceptions and ability are inextricably linked, and must therefore be included in any discussion about gender differences in leadership ability.
In an early meta-analysis, Eagly et al. (1995) synthesized the research to date on the relative effectiveness of male and female leaders. These leaders usually were assessed in the context of real organizations, but some were college students analyzed in a lab setting. Though leader performance in the selected studies was operationalized as either “objective” (e.g., subordinate performance) or “subjective” (e.g., effectiveness ratings) measures, the number of studies using subjective measures far outweighed those using objective measures. In the studies the authors included, the majority of leaders were first-level (i.e., line) leaders but some were middle level and some were unspecified. Across all studies, the authors did not detect a main effect for gender on general leader effectiveness, suggesting that there was no obvious difference between men and women on leadership performance.
Eagly and colleagues (1995) also examined the potential influence of several moderating variables, including study setting (i.e., organization or lab), level of leadership (i.e., first level or middle level), percentage of men among leadership roles, and “congeniality of leadership roles” (i.e., the perceived masculinity or femininity of the roles). The authors determined that men were evaluated more favorably than women in a military setting and when occupying a first-level leadership role, while women were evaluated more favorably in the context of government, and social service, and when occupying a middle-level leadership role. Additionally, the researchers found that male and female leaders were evaluated more positively when their roles were perceived to be masculine or feminine, respectively. Male leaders also appeared to be more effective when their role was typically male-dominated and associated with male subordinates.
Overall, this meta-analysis was the first to analyze the literature systematically on differences in leader effectiveness. Although there were no main effects for gender, several moderating variables pertaining to leadership context emerged (e.g., sex-type of role). However, it is not entirely clear whether these results measured leader performance or simply gender bias, or a combination of the two. Yet one fact became clear: The fear that organizations would put themselves at risk by selecting women for leadership positions was likely unfounded. Finally, these findings set the stage for RCT (Eagly & Karau, 2002). As mentioned previously, RCT posits that perceived incongruity between the characteristics of women and requirements of leader roles can result in prejudice against female leaders, which may manifest in performance evaluations.
Nearly two decades later, Paustian-Underdahl, Walker, and Woehr (2014) conducted another meta-analysis focusing only on perceptions of leadership effectiveness (as opposed to objective measures of effectiveness). The researchers sought to update previous meta-analytic findings by considering the plethora of research that has been carried out in recent years. Additionally, the authors sought to apply RCT to men and women to determine whether incongruence can result in a disadvantage for men, as well as women. The researchers addressed the issue of potential differences in perceived leader performance when self-rated or rated by another individual. Finally, they also attended to other potential moderators in their meta-analysis, including study publication date, type of organization, hierarchical level of the leader, study setting, and percent of male raters.
Similar to Eagly et al. (1995), Paustian-Underdahl et al. (2014) did not find a main effect for gender on perceptions of leadership effectiveness across studies, implying a lack of difference in perceived effectiveness between men and women overall. Publication date was considered as a moderator because the increasing number of women in leadership positions over time may serve as disconfirming information of role stereotypes and thereby weakening the perceived incongruity between women and leadership. Therefore, the authors expected female leaders to be evaluated more negatively in earlier years compared with later ones. However, although male leaders were seen as more effective in older studies and female leaders were seen as more effective in newer studies, the interaction term was not significant. Regarding organization type as a moderating factor, data across studies supported RCT predictions that men would be perceived as more effective in organizational settings that are dominated by men. However, differences between men and women for female-dominated organizations were nonsignificant. For hierarchical level, women were perceived as more effective than men in middle management positions, which mirrors the findings of Eagly et al. (1995). Self- versus other-rating was found to significantly moderate the relationship between leader gender and performance evaluation, such that men rate themselves more highly than women rate themselves, but others rate women more highly than they rate men.
Finally, the analysis showed that rating source interacted with hierarchical level, such that women serving as senior leaders were perceived by others as more effective than men at senior levels. This finding is particularly germane to this article because (1) CEO gender and top management team (TMT) gender diversity are especially important when considering the impact of leaders on whole-organization outcomes and (2) other-ratings, as opposed to self-ratings, are arguably less prone to bias and are therefore more reliable (Conway & Huffcutt, 1997). In all, there is tentative evidence to suggest that women leaders are evaluated more favorably when occupying higher leadership positions, though this effect is small, context dependent, and influenced by the rating source.
Does Leader Gender Have an Impact on Organizational and Group Outcomes?
Although perceptions can be a valuable measure to determine gender differences in leadership performance, it is also important to consider how male and female leaders may differentially have an impact on organizational outcomes. Theoretically, we must analyze organization- and group-level outcomes to capture the whole construct of leader effectiveness. Practically, many organizations will only be spurred to employ more women as leaders once they are confident that this will have a beneficial (or at least not detrimental) impact on the organization. Demonstrating the value-added of female leaders through organizational, especially financial, metrics would help establish a “business case” for gender diversity among leadership (Lückerath-Rovers, 2013; Mahadeo, Soobaroyen, & Hanuman, 2012; Stephenson, 2004).
Kulik and Metz (2015) reviewed the literature on the link between women in leadership and wide range of organizational outcomes. The most common outcome in this literature is firm financial performance, likely because it is relatively easy to operationalize and typically thought of as a key indicator of organization success. Some studies have also included shareholder/investor reactions (e.g., Kolev, 2012; Lee & James, 2007; Martin, Nishikawa, & Williams, 2009; Wolfers, 2006), corporate social responsibility (CSR; e.g., Bear, Rahman, & Post, 2010; Boulouta, 2013; Post, Rahman, & Rubow, 2011), group processes (e.g., Abbott, Parker, & Presley, 2012; Adams & Ferreira, 2009; del Carmen Triana, Miller, & Trzebiatowski, 2014), organizational practices (e.g., Melero, 2011; Post, 2015), and organizational demography (e.g., Gould, Kulik, & Sardeshmukh, 2014; Kurtulus & Tomaskovic-Devey, 2012; Matsa & Miller, 2011). In this review, we focus on financial performance, CSR, group processes, and organizational practices because these outcomes are more likely the result of actions of a firm’s leaders, whereas other measures, such as investor reactions, are probably due to signaling effects that have less to do with leader performance, per se.
Kulik and Metz (2015) also partition their review by level of analysis with respect to the leader’s position (i.e., CEO, corporate board, TMT, and managers), partially because different theories are brought to bear when predicting and explaining results at each level. For instance, we may use leadership style to explain performance differences between male and female CEOs (e.g., women may be more transformational, leading to higher organizational performance). However, when analyzing groups of top decision makers, such as corporate boards, we may focus on the effect of varying demographic composition on group decision-making processes. In this section, we will review the literature on gender differences according to each major operationalization of organizational performance while keeping in mind the different levels of analysis.
Financial Outcomes
Financial outcomes are important when assessing the effectiveness of male versus female leadership because these measures are often considered a proximal measure of a firm’s overall success (Venkatraman & Ramanujam, 1986). Most researchers studying the effect of leader gender on financial performance focus on measures such as return on assets (ROA), return on investment (ROI), and return on equity (ROE; Kulik & Metz, 2015).
The majority of research examining the impact of the gender of CEOs specifically has found that firms run by female CEOs often report better ROA, ROE, and sales performance (Jalbert, Jalbert, & Furumo, 2013; Khan & Vieito, 2013; Peni, 2014; Smith, Smith, & Verner, 2006). However, it should be noted that most of these studies examine large firms in the United States, and research including smaller firms from a wider variety of countries is still needed. Indeed, research looking at the CEOs of small- and medium-sized firms has not found strong evidence of a female leadership advantage (e.g., Davis, Babakus, Englism, & Pett, 2010; Du Rietz & Henrekson, 2000), nor has research in China (Lam, McGuinness, & Vieito, 2013) or Sweden (Du Rietz & Henrekson, 2000).
Another stream of research focuses on the effect of corporate board gender composition on firm financial outcomes (i.e., relative representation of men and women on corporate boards). Studies in this area are much more diverse with respect to geography than the CEO literature (Kulik & Metz, 2015); however, findings have been very mixed with positive effects found in Australia (Ali, Ng, & Kulik, 2014) and the United States (Carter, Simkins, & Simpson, 2003), negative effects in Indonesia (Darmadi, 2013) and the United States (Dobbin & Jung, 2011) and nonsignificant effects in Norway (Bøhren & Strøm, 2010) and the United States (Miller & del Carmen Triana, 2009). Overall, there is no clear evidence that more women on corporate boards improve firm financial performance. Kulik and Metz (2015) suggest that this may be due to the overall low representation of women on boards, the culture variations that affect how women are selected to boards, or variability in financial performance metrics used across studies.
Very little research has focused on the relationship between the number of women on the TMT and firm financial performance. Shrader, Blackburn, and Iles (1997) found no relationship between TMT gender diversity and firm ROA, ROI, ROE, or return on sales (ROS). However, Krishnan and Park (2005) found a positive relationship between TMT gender diversity and firm ROA and ROS. Finally, Dezsö and Ross (2012) found that TMT gender diversity positively correlated with Tobin’s Q, ROA, and ROE, but only for firms whose strategy was focused on innovation.
Last, there are a small number of studies that have examined the relationship between the percentage of female managers throughout the organization and firm financial performance. Dwyer, Richard, and Chadwick (2003) found that more women in the management ranks has a positive impact on firm financial performance in companies that demonstrate growth-orientation and have cultures that emphasize innovation, flexibility, and interaction. Richard, Kirby, and Chadwick (2013) found that management gender diversity had a positive impact on firm outcomes only when participative strategy making (a measure of inclusiveness) was high. Finally, Ali, Metz, and Kulik (2015) determined that management gender diversity had a negative impact on firm outcomes in organizations with few work–family programs; however, in organizations with many work–family programs, this relationship is positive, though not significant.
Corporate Social Responsibility
CSR can be broadly defined as the extent to which an organization incorporates social and environmental concerns into their business operations on a voluntary basis (Dahlsrud, 2008). These concerns go beyond what is required for profit maximization (McWilliams & Siegel, 2000) and are therefore distinct from the financial metrics discussed above. It has become an increasingly important construct due to public corporate scandals (e.g., Enron) and financial crises in recent years (Boulouta, 2013). Some authors argue that female leaders are more likely than male leaders to raise issues that benefit society (e.g., Bear et al., 2010; Boulouta, 2013; Larkin, Bernardi, & Bosco, 2013; Post et al., 2011). Overall, research on corporate board gender composition has indicated that an increased proportion of female board members is associated with more CSR strengths (Bear et al., 2010; Post et al., 2011) and fewer CSR concerns (Boulouta, 2013), broadly implying that female leadership in the boardroom may positively affect nonfinancial organizational outcomes.
Group/Team Processes and Organizational Practices
Several studies have indicated that board and management gender composition have an impact on internal outcomes (i.e., those that affect individuals inside the organization rather than outside). Nielsen and Huse (2010) found that corporate boards with more female directors are more likely to engage in activities that help internal processes, such as board evaluations and board development programs. More gender-diverse boards allocate more effort to internal monitoring, female directors have better attendance records than male directors, and male directors display fewer attendance problems when the board is diverse (Adams & Ferreira, 2009). Additionally, boards with at least one female are less likely to issue financial restatements, implying that gender-diverse boards are more attentive to detail and less prone to groupthink than all-male boards (Abbott et al., 2012; Kulik & Metz, 2015). However, del Carmen Triana et al. (2014) found that board gender diversity is negatively associated with strategic change. The authors suggest that this may be the case because heterogeneous groups are less cohesive and have more diverse cognitive styles, and may therefore have trouble agreeing on strategic change.
At the management level of analysis, Melero (2011, p. 385) found that “workplace management teams with a higher proportion of women monitor employee feedback and development more intensely.” In addition, one study found that female leaders are better able to foster cohesion in larger and functionally diverse teams, and may induce more cooperative learning and participative communication in larger and geographically dispersed teams (Post, 2015). These results provide some preliminary evidence that work teams benefit in terms of group process when there are more women in management.
Performance Summary
Our review of the research on gender differences in leadership performance indicates that there is little evidence of a clear advantage for men or women across contexts. This applies when performance is operationalized using leadership evaluations (e.g., Paustian-Underdahl et al., 2014) or organizational and group outcomes (e.g., Kulik & Metz, 2015). In light of this finding, Kulik and Metz (2015) suggest that future research on gender and leadership performance should proceed by attending to the various levels of analysis when analyzing the relationship between leader gender and performance. This is because different theoretical perspectives can explain the relationship between leader gender and performance at difference levels. For instance, female and male CEOs may perform differently due to different leadership styles or mind-sets, while mixed-gender groups of managers may be more or less effective due to the group dynamics that emerge from group diversity. Meanwhile, meta-analyses show that evaluation of male and female leaders may be highly susceptible to context and rater bias (Paustian-Underdahl et al., 2014), so theories of gender role congruity (Eagly & Karau, 2002) must be incorporated into any research involving leadership evaluation.
Kulik and Metz (2015) also suggest incorporating moderators and mediators into analyses of gender and leadership performance. In particular, national context, organizational strategy, and organizational culture should be investigated as key factors that influence whether a difference in leadership performance is observed. Additionally, if a difference is observed, it is important to determine why, rather than simply documenting the discrepancy. To this end, leadership behavior (i.e., style, values, attitudes, etc.) may be one source of difference between men and women leadership performance. However, these gender differences in leader performance may also be attributed to differences in group decision-making processes or even the stakeholder reactions. Therefore, we recommend that future work on gender and leadership performance pay special attention to intervening processes (i.e., moderators and mediators) in order to explicitly test the theory behind gender and leadership.
Discussion
Many researchers have looked at the leadership gap from a gendered perspective, relying heavily on the stereotyping and discrimination literature to explain gender differences when it comes to leadership (e.g., Eagly & Karau, 2002; Heilman, 2001). We take a slightly different approach and instead explore the experiences of women leaders using a leadership framework, that is, looking at how gender influences selection, development, style, and performance of women leaders.
When it comes to selection, the first tenet of our framework, the preference for male leadership has been demonstrated time and time again. It is a well-documented phenomenon that although women compose approximately half of management positions (51.5%; U.S. Bureau of Labor Statistics, 2015), representation sharply decreases as organizational level increases, the so-called glass ceiling effect. While the percentage of women occupying elite leadership positions is on the rise, most senior executives are men (Eagly & Carli, 2007). Gender differences also occur when it comes to development, with women having far less developmental opportunities than men (Ely et al., 2011). Findings in leadership style, our third area of interest, indicate that while there are some differences between men and women when it comes to style, these differences do not lead to a clear advantage of either gender across contexts. Finally, evidence suggesting gender differences when it comes to performance, our final tenet, is even less concrete and conclusive.
One limitation of many of the studies cited is a lack of attention paid to the role race, ethnicity, sexual orientation, and other types of diversity play in the experiences of leadership. Research on diverse samples in this domain remains relatively sparse (Hoyt & Chemers, 2008). Future explorations on the leadership process for women should consider intersectionality literature as a way to further understand the ways in which women with multiple social identities experience leadership (e.g., Richardson & Loubier, 2008; Sanchez-Hucles & Davis, 2010). Furthermore, because the majority of these findings were based in the United States, future research should explore gender and leadership in alternative national contexts to see whether these findings can be generalized beyond Western samples. Given that the leadership gap has been well documented worldwide, it is important for future research to examine gender differences when it comes to development, leadership style, and performance from a cross-cultural perspective.
It has become increasingly apparent that women face significant challenges when it comes to being selected for top leadership positions and that these challenges prevent women from advancing into the upper echelons of leadership at the same rate as their male counterparts. Recently, Eagly and Carli (2007) have abandoned the glass ceiling metaphor describing the obstacles that women encounter as a leadership labyrinth instead. This labyrinth metaphor better speaks to the ongoing and unpredictable challenges that women encounter at every stage of the leadership process. Understanding the many aspects of the labyrinth is crucial to combating gender inequality when it comes to elite leadership positions. Tackling this issue from multiple perspectives will help ensure that women have equal opportunity at obtaining senior executive positions. Last, recognizing that there are significant challenges that prevent women from ascending the leadership hierarchy is the first of many steps toward instituting better practices when it comes to selecting, developing, and evaluating women leaders.
Implications for Practice
Understanding gender differences when it comes to leadership is also of particular interest to practitioners, especially given the increasing amount of attention paid to women in leadership positions. The current exploration into this domain found that despite a lack of significant difference in style and performance between men and women, biases against women in selection and access to developmental opportunities is abundantly clear and remains a problem. What, then, might we recommend to senior executives, human resource professionals, and other key players in organizations who make decisions about leadership?
First, let us consider selection: Serious mistakes are made at this initial stage of decision-making regarding leadership. It may be that the most egregious error in selection is that we select for the wrong traits and behaviors. A recent Harvard Business Review article makes the case that since we often mistake confidence for competence, we are fooled into believing men are better leaders than women (Chamorro-Premuzic, 2013). Furthermore, we tend to choose candidates who fit the corporate culture, which may produce biases against women (e.g., Giberson, Resick, & Dickson, 2005; Jackson et al., 2007). Another likely error in selection is to assume that because the organization has been through a “rough patch” and some healing is needed a woman should be selected as the leader and successor (Ryan & Haslam, 2007). This, of course, creates the a double-bind for women, among other issues. Criteria for selection should be based on evidence, that is, known qualities of effective leadership such as self-awareness, learning agility, that is, being able to learn and adapt quickly to new situations, accepting of coaching and mentoring, and a good fit between the candidate’s personality and the organization’s culture (e.g., de Meuse et al., 2010; Goleman, 2004).
Second, in regard to the development of leaders, we need to examine the purpose of programs that are specifically designed and conducted for women managers and executives. Typically these programs are seen as providing support and encouragement for women who are currently in, or preparing for, positions of major responsibility. This is all well and good but the substance of such programs is important. The focus needs to emphasize a deeper understanding of women’s issues in leadership positions—the double bind, role congruence, and so on—but not exclusively. Women will not be well served unless they are also given opportunities to increase their self-awareness, emotional intelligence, ability to learn under never-before-encountered tasks and situations, mechanisms for self-regulation and control, which are all key ingredients of effective leadership regardless of gender.
Finally, context matters. Time and again in this report on women and leadership the importance of context has arisen. In all four areas of our framework—selection, development, style, and performance—context plays a significant role, particularly in terms of moderating and mediating variables. Leaders are embedded in groups, organizations, and the broader societal context, such that they cannot be separated from the environment in which they are operating (Lewin, 1939). Therefore, with regard to making decisions about leadership, it is important to take into account the organization’s mission, strategy, culture, history, as well as its goals regarding social responsibility, diversity in general, and degree of globalization. Furthermore, practitioners should consider how the aforementioned factors interact with gender especially with regard to selection and development. In this way, we can better understand how the context affects the experiences of women leaders and those aspiring to be in senior leadership positions.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
