Journal of Social Issues, Vol. 57, No. 4, 2001, pp. 781-797 The Leadership Styles of Women and Men Alice H. Eagly* and Mary C. Johannesen-Schmidt Northwestern University As women increasingly enter leadership roles that traditionally have been occupied mainly by men, the possibility that the leadership styles of women and men differ continues to attract attention. The focus of these debates on sameness versus difference can obscure the array of causal factors that can produce differences or similarities. Adopting the perspective of social role theory, we offer a framework that encompasses many of the complexities of the empirical literature on the leadership styles of women and men. Supplementing Eagly and Johnson's (1990) review of the interpersonally oriented, task-oriented, autocratic, and democratic styles of women and men, we present new data concerning the transformational, transactional, and laissez-faire leadership styles. Whether men and women behave differently in leadership roles is a much-debated question. Although there is general agreement that women face more barriers to becoming leaders than men do, especially for leader roles that are male-dominated (see Eagly & Karau, in press), there is much less agreement about the behavior of women and men once they attain such roles. This issue is usually discussed in terms of leadership styles, when style is understood as relatively stable patterns of behavior that are manifested by leaders. Differences in styles can be consequential, because they are one factor that may affect people's views about whether women should become leaders and advance to higher positions in organizational hierarchies. To approach this issue, we first analyze traditional thinking about the leadership styles of women and men. Then we present our own * Correspondence concerning this article should be addressed to Alice H. Eagly, Department of Psychology, Northwestern University, 2029 Sheridan Road, Evanston, IL 60208 [e-mail: eagly@ northwestern.edu]. The writing of this article was supported by a research grant to the first author from the National Science Foundation, SBR-9729449. The authors thank Marloes Van Engen and Claartje Vinkenburg for comments on a draft of this article and Bruce Avolio for making available analyses that compared the responses of male and female managers on the Multifactor Leadership Questionnaire in a large norming sample (Center for Leadership Studies, 2000). 781 © 2001 The Society for the Psychological Study of Social Issues 782 Eagly and Johannesen-Schmidt theoretical framework for understanding these issues and examine and interpret relevant research findings. It is not surprising that women are the usual focus of discussions of the impact of gender on leadership. Because social perceivers generally concentrate on the nonprototypical members of categories (Miller, Taylor, & Buck, 1991), people direct their attention to the adequacy of women's leadership styles. For example, Elaine La Roche commented in reference to her experience as an executive at Morgan Stanley "that issues of style with respect to women can unfortunately often be more important than issues of substance" (Thrall, 1996, p. C4). Female politicians thus worry about "projecting gravitas," as former U.S. Congressional Representative Patricia Schroeder noted (Schroeder, 1999, p. A17). In contrast, because men have long held these roles, they have defined the styles to which people have become accustomed. Despite this focus on women's leadership, there is little agreement about how women actually lead. These debates reflect the common cultural debate about difference and similarity, which has been especially important in feminist writings (see Kimball, 1995). Some feminists thus fear that the perception of sex differences in leadership style or other attributes can provide a rationale for excluding women from opportunities and especially from male-dominated leadership roles. Other feminists believe that the perception of sameness would fail to acknowledge the relational qualities that are a traditional source of female pride and that may contribute to superior performance by women leaders. In this article, we escape the dichotomy between difference and similarity by explaining why sex differences in leadership behaviors are sometimes present, appearing and disappearing with shifts in social contexts. Contrary to our view that sex differences and similarities vary with social contexts, experts who have written about this topic have generally maintained that either differences or similarities prevail. The advocates of difference include several writers of trade books who have drawn on their personal experience in organizations as well as informal surveys and interviews of managers. These writers have claimed that the leadership styles of women and men are different, mainly along the lines of women being less hierarchical, more cooperative and collaborative, and more oriented to enhancing others' self-worth (e.g., Book, 2000; Helgesen, 1990; Rosener, 1995). In contrast, social scientists have typically either claimed that female and male organizational leaders do not differ or minimized the importance of those differences that have been observed (e.g., Powell, 1990). Careful examination of relevant research, however, has revealed more complex findings than acknowledged by the advocates of difference or the advocates of similarity. To consider these issues, we discuss some theoretical principles that underlie male and female leadership styles and evaluate relevant empirical research. Leadership Styles 783 Theoretical Rationale for Sex Differences and Similarities in Leadership Style Analysis of the situation that women and men face as leaders provides a rationale for expecting differences and similarities. From the perspective of social role theory of sex differences and similarities (Eagly, Wood, & Diekman, 2000), this analysis begins with the principle that leadership roles, like other organizational roles, are but one influence on leaders' behavior. In addition, leaders elicit expectancies based on people's categorization of them as male and female. These expectancies constitute gender roles, which are the shared beliefs that apply to individuals on the basis of their socially identified sex. These roles are assumed to follow from perceivers' observations of men and women as concentrated in different social roles in the family and paid employment. Aspects of gender roles that are especially relevant to understanding leadership pertain to agentic and communal attributes (see Eagly et al., 2000). Agentic characteristics, which are ascribed more strongly to men than women, describe primarily an assertive, controlling, and confident tendency—for example, aggressive, ambitious, dominant, forceful, independent, daring, self-confident, and competitive. In employment settings, agentic behaviors might include speaking assertively, competing for attention, influencing others, initiating activity directed to assigned tasks, and making problem-focused suggestions. Communal characteristics, which are ascribed more strongly to women than men, describe primarily a concern with the welfare of other people—for example, affectionate, helpful, kind, sympathetic, interpersonally sensitive, nurturant, and gentle. In employment settings, communal behaviors might include speaking tentatively, not drawing attention to oneself, accepting others' direction, supporting and soothing others, and contributing to the solution of relational and interpersonal problems. Simultaneous Occupancy of Gender Role and Leader Role Managers and other leaders occupy roles defined by their specific position in a hierarchy but also simultaneously function under the constraints of their gender roles. Although it would be consistent with a structural interpretation of organizational behavior (e.g., Kanter, 1977) to predict that men and women who occupy the same leadership role would behave very similarly, gender roles ordinarily continue to exert some influence, with the result that female and male occupants and potential occupants of the same organizational role may behave somewhat differently. Consistent with this reasoning, Gutek and Morasch (1982) argued that gender roles spill over to organizations, and Ridgeway (1997, p. 231) maintained that gender provides an "implicit, background identity" in the workplace. 784 Eagly and Johannesen-Schmidt Despite the likely influence of gender roles on leaders' behavior, formal leadership (or managerial) roles should be of primary importance in organizational settings, because these roles lend their occupants legitimate authority and are regulated by relatively clear rules about appropriate behavior. This idea that the influence of gender roles can be diminished or even eliminated by other roles was foreshadowed by experimental demonstrations of the lessening or disappearance of many gender-stereotypic sex differences in laboratory settings when participants received information that competed with gender-based expectations (see Eagly etal., 2000; Wagner & Berger, 1997). In contrast, research in natural settings suggests that, although some gender-stereotypic differences erode under the influence of organizational roles, other differences do not. Particularly informative is a field study by Moskowitz, Suh, and Desaulniers (1994) that examined the simultaneous influence of gender roles and organizational roles. This study used an experience-sampling method in which participants monitored their interpersonal behavior in a variety of work settings for 20 days. In general, agentic behavior was controlled by the relative status of the interaction partners, with participants behaving most agentically with a supervisee and least agentically with aboss. Communal behaviors, however, were influenced by the sex of participants, regardless of participants' status, with women behaving more communally than men, especially in interactions with other women. Although research that considers the joint impact of gender roles and organizational roles is sparse (see Eagly etal., 2000, for other examples), it suggests some tentative generalizations about the increased similarity of women and men who are in the same organizational role. It is thus likely that leadership roles, like other organizational roles, provide norms that regulate the performance of many tasks, which would therefore be similarly accomplished by male and female role occupants. For example, a manager is obligated to carry out a range of activities, such as monitoring subordinates' performance and gathering and disseminating information. Despite pressures to conform to such norms, managers generally have some leeway to vary the manner in which they carry out these required activities. Managers may thus be friendly or more remote, consult few or many colleagues about decisions, and so forth. Organizational behaviors include in addition a wide range of more informal actions that are not narrowly regulated by organizational roles (e.g., chatting about sports, commemorating coworkers' birthdays). It is these elective and discretionary aspects of organizational behavior that may be most likely to vary according to gender. As Eagly et al. (2000) argued, this influence of gender roles on organizational behavior occurs not only because people react to leaders in terms of gendered expectancies and leaders respond in turn, but also because most people have internalized gender roles to some extent (Wood, Christensen, Hebl, & Rothgerber, 1997). As a consequence of these differing social identities, women and men have somewhat different expectations for their own behavior in organizational settings Leadership Styles 785 (Ely, 1995). Self-definitions of managers may reflect a blending of their managerial role and gender role, and, through self-regulatory processes, these composite self-definitions influence behavior. Such a blending was suggested by a metaanalysis of findings obtained on a measure of "motivation to manage," which assesses the desire to satisfy the requirements of the managerial role that has traditionally existed in hierarchic organizational contexts, particularly within business firms (Miner, 1993). Across 51 data sets (Eagly, Karau, Miner, & Johnson, 1994), men scored slightly higher than women on this measure, especially on subscales that assessed the desire to manifest competitive and assertive qualities in managing. Such qualities are strongly masculine in connotation and, as we explain in the next subsection, may especially elicit negative evaluations when enacted by women. Congruity of Leader Roles and Gender Roles Female leaders' efforts to accommodate their behavior to the sometimes conflicting demands of the female gender role and their leader role can foster leadership styles that differ from those of men. Gender roles thus have different implications for the behavior of female and male leaders, not only because the female and male roles have different content, but also because there is often inconsistency between the predominantly communal qualities that perceivers associate with women and the predominantly agentic qualities that they believe are required to succeed as a leader. People thus tend to have similar beliefs about leaders and men but dissimilar beliefs about leaders and women, as Schein (this issue) has demonstrated. Nonetheless, the degree of perceived incongruity between a leader role and the female gender role would depend on many factors, including the exact definition of the leader role, the activation of the female gender role in a particular situation, and individuals' personal approval of traditional definitions of gender roles (see Heilman, this issue). As Eagly and Karau (in press) argued, perceived incongruity between the female gender role and typical leader roles tends to create prejudice toward female leaders and potential leaders that takes two forms: (1) less favorable evaluation of women's (than men's) potential for leadership because leadership ability is more stereotypic of men than women and (2) less favorable evaluation of the actual leadership behavior of women than men because agentic behavior is perceived as less desirable in women than men. The first type of prejudice stems from the descriptive norms of gender roles, that is, the activation of descriptive beliefs about women's characteristics and the consequent ascription to them of female-stereotypic qualities, which are unlike the qualities expected and desired in leaders. The second type of prejudice stems from the injunctive (or prescriptive) norms of gender roles, that is, the activation of beliefs about how women ought to behave. If female leaders violate these prescriptive beliefs by fulfilling the agentic 786 Eagly and Johannesen-Schmidt requirements of leader roles and failing to exhibit the communal, supportive behaviors that are preferred in women, they can be negatively evaluated for these violations, even as they may also receive some positive evaluation for their fulfillment of the leader role. The role congruity analysis thus suggests that 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. Particularly consequential for leadership style would be the second form of prejudice, that is, the negative reactions that women may experience when they behave in a clearly agentic style, especially if that style entails exerting control and dominance over others. In summary, the social role argument that leadership roles constrain behavior so that sex differences are minimal among occupants of the same leadership role must be tempered by several more complex considerations. Not only may gender roles spill over to organizational settings, but leaders' gender identities may also constrain their behaviors in a direction consistent with their own gender role. Also, the female gender role is more likely to be incongruent with leader roles than the male gender role is, producing a greater potential for prejudice against female leaders. Such prejudice could result in negative sanctions that affect leaders' behavior. Types of Leadership Style The impact of gender on leadership style should emerge especially clearly on measures of style that reflect the agentic norms associated with the male gender role and the communal norms associated with the female gender role. Using measures that reflect this design, the classic work on leadership defined styles that are primarily agentic or primarily communal (see Bass, 1990; Cann & Siegfried, 1990). Most common was a distinction between two approaches to leadership: task-oriented style, defined as a concern with accomplishing assigned tasks by organizing task-relevant activities, and interpersonally oriented style, defined as a concern with maintaining interpersonal relationships by tending to others' morale and welfare. This distinction was introduced by Bales (1950) and developed further in the Ohio State studies onleadership (e.g., Hemphill & Coons, 1957). In this research, task-oriented style, labeled initiation of structure, included behavior such as encouraging subordinates to follow rules and procedures, maintaining high standards for performance, and making leader and subordinate roles explicit. Inter-personally oriented style, labeled consideration, included behavior such as helping and doing favors for subordinates, looking out for their welfare, explaining procedures, and being friendly and available. Another aspect of leadership style that has been popular in research is the extent to which leaders (1) behave democratically and allow subordinates to Leadership Styles 787 participate in decision making or (2) behave autocratically and discourage subordinates from participating in decision making. This dimension of democratic versus autocratic leadership (or the similar dimension of participative versus directive leadership) follows from early experimental studies of leadership style (e.g., Lewin & Lippitt, 1938) and has been developed since that time by a number of researchers (e.g., Vroom & Yetton, 1973). Although democratic versus autocratic style is a narrower aspect of leader behavior than task-oriented versus interperson-ally oriented style (see Bass, 1990), the democratic-autocratic dimension also relates to gender roles, because one component of the agentic norms associated with these roles is that men are relatively more dominant and controlling—in other words, more autocratic and directive—than women are. In the 1980s and 1990s, many researchers turned their attention to other types of leadership styles by distinguishing between leaders who are transformational and those who are transactional (Bass, 1998). This effort was initially inspired by Burns's (1978) argument that existing analyses of leadership style left out some of the most important aspects of effective leadership. To capture these neglected aspects, he proposed that researchers study a type of leadership that he labeled transformational. Such leaders set especially high standards for behavior and establish themselves as role models by gaining the trust and confidence of their followers. They state future goals and develop plans to achieve them. Skeptical of the status quo, transformational leaders innovate, even when the organization that they lead is generally successful. By mentoring and empowering followers, such leaders encourage them to develop their full potential and thereby contribute more capably to their organization. Burns contrasted leaders with these characteristics to transactional leaders, who establish exchange relationships with their subordinates. Such leaders manage by clarifying subordinate responsibilities, monitoring their work, and rewarding them for meeting objectives and correcting them for failing to meet objectives. Researchers also distinguished a laissez-faire leadership style that is marked by a general failure to take responsibility for managing. Although transformational and transactional styles are not as obviously related to gender roles as the leadership styles investigated by earlier researchers, transformational leadership has communal aspects, especially the theme of individualized consideration, whereby leaders focus on the mentoring and development of their subordinates and pay attention to their individual needs. Consistent with the possibility that transformational leadership may be somewhat more aligned with the female than the male gender role are studies showing that subordinates perceive greater correspondence between leaders' feminine personality attributes and their transformational style than their transactional style (Hackman, Furniss, Hills, & Paterson, 1992; Ross & Offermann, 1997). In summary, to the extent that gender roles spill over to influence leadership behavior in organizational settings, the behavior of female leaders, compared with that of male leaders, may be more interpersonally oriented, democratic, and 788 Eagly and Johannesen-Schmidt transformational. In contrast, the behavior of male leaders, compared with that of female leaders, may be more task-oriented and autocratic. In addition, the greater incongruence of the female than male gender role with typical leader roles may make it more difficult for women than men to manifest the more agentic leadership styles. Because of the constraining impact of leadership roles, however, any differences between women and men who occupy the same role are unlikely to be large in size. Empirical Research Comparing Male and Female Leadership Styles A large number of studies have compared the leadership styles of women and men. Most of these studies have focused on task and interpersonal styles, and smaller numbers have examined autocratic versus democratic style or transformational, transactional, and laissez-faire styles. Although researchers have used a diversity of methods to assess style, the most common measures have each leader's colleagues—often his or her subordinates—rate their leader on items that describe the critical features of the styles that researchers desire to assess. Task-Oriented, Interpersonally Oriented, Democratic, and Autocratic Styles Eagly and Johnson (1990) reviewed studies that compared men and women on task and interpersonal styles and democratic and autocratic styles. In their metaanalysis the comparison between male and female behavior for each relevant study was represented in terms of its effect size (or d), which expresses the sex difference in units of the study's standard deviation. With each finding represented by an effect size, multiple studies were collectively represented by the average of their effect sizes. Means of the effect sizes averaged across the studies for three types of leadership style appear in the first row of Table 1. Although men and women did not differ on task-oriented style, the very small tendency for women to be more inter-personally oriented than men was significant. On measures that assessed tendencies to be democratic versus autocratic (or participative versus directive), men were more autocratic or directive than women, and women were more democratic or participative than men. To clarify these overall findings, Eagly and Johnson (1990) divided the studies into three types according their social context: (1) laboratory experiments, which compared the leadership styles of male and female leaders of laboratory groups; (2) assessment studies, which compared the leadership styles of people not selected for occupancy of leadership roles (e.g., nonmanagerial employees or business students); and (3) organizational studies, which compared the leadership styles of male and female managers who occupied the same organizational role (e.g., elementary school principal). Leadership Styles 789 Table 1. Meta-Analytic Comparisons of the Leadership Styles of Men and Women in All Studies and Organizational, Assessment, and Laboratory Studies Type of í ityle Democratic vs. Type of study Interpersonal Task autocratic n d+ n d+ n d+ All 136 0.04* 139 0.00 23 0.22* Organizational 120 -0.01a 120 -0.02a 13 0.21* Assessment 12 0.25b* 12 0.08b 6 0.29* Laboratory 4 0.37b* 7 0.19b 4 0.20* Note. From "Gender and Leadership Style: A Meta-Analysis," by A. H. Eagly and B. T. Johnson, 1990, Psychological Bulletin, 108, pp. 233-256. Copyright 1990 by American Psychological Association. Adapted with permission. Mean effect sizes (d+) are positive for differences that are stereotypic (women more inter-personally-oriented or democratic than men, and men more task-oriented or autocratic than women) and negative for differences that are counterstereotypic. n = number of effect sizes that are averaged; d+ = weighted mean of effect sizes. The categorical model comparing organizational, assessment, and laboratory studies is significant for the interpersonal style effect sizes, p < .001, and the task style effect sizes, p < .01. Effect sizes within columns that do not share a subscript differ atp < .05. *Effect size (d) differed significantly (p < .05) from 0.00 (exactly no difference). By examining the mean effect sizes within each of these types of studies (see second, third, and fourth rows of Table 1), Eagly and Johnson (1990) found a significant relation between the social context of the research and the extent to which leadership styles were gender-stereotypic. Specifically, in the laboratory and assessment settings, the tendency for participants to have gender-stereotypic styles—women interpersonally oriented and men task-oriented—was stronger than it was in the organizational settings. Because constraining managerial roles were not present in the laboratory experiments or the assessment studies, men and women were more likely to approach leadership with the gender-congruent shading by which men behaved agentically and women communally. These findings thus resembled the stereotypic sex differences observed in most of the research literature on small-group interaction (see Ridgeway, this issue; Wagner & Berger, 1997). In the organizational studies, however, these gender-stereotypic tendencies in task and interpersonal style were eliminated, presumably because gender became merely a background influence as the managerial role took precedence. The tendency for women to be more participative and democratic than men, however, was intact in all three classes of studies, including organizational studies. Although the findings on task and interpersonal styles thus provided some support for the social role principle that the constraints of leadership roles cause sex differences to decrease in magnitude, the absence of this pattern on measures of democratic versus autocratic style invites interpretation. To the extent that female managers favor more democratic and participative styles than male managers, this tendency may reflect the attitudinal bias against female leaders that arises from the incongruity of the female gender role and many leader roles (Eagly & Karau, in press). The resulting lack of legitimacy for female leaders can make the clear-cut 790 Eagly and Johannesen-Schmidt exercise of power and dominance difficult for women because they encounter resistance to their authority (Ridgeway, this issue). Women may thus encounter negative reactions when they take charge in the especially authoritative manner of autocratic and directive leaders (see Carli, this issue; Carli & Eagly, 1999; Rudman & Glick, this issue). This interpretation is also in line with Eagly, Makhijani, and Klonsky's (1992) meta-analysis of studies examining evaluations of male and female leaders whose behavior had been experimentally equated. Eagly, Makhijani, and Klonsky's findings showed that participants evaluated autocratic behavior by female leaders more negatively than they evaluated the equivalent behaviorby male leaders. Because men are not so constrained by others' attitudinal biases, they are freer to lead in a more autocratic and nonparticipative manner, should they so desire. Furthermore, as research on motivation to manage suggests (Eagly et al., 1994), men are somewhat more interested than women in taking charge in a clear-cut manner in hierarchic relationships. Placating subordinates so that they accept a woman's leadership may to some extent require that she allow them some degree of control over these decisions. This sort of collaborative decision making no doubt introduces interpersonal complexity not encountered by leaders who proceed in a more directive manner. Because women's communal repertoire encompasses social skills (e.g., Hall, 1998), it may be easier for women than men to behave in this participative manner. Moreover, to the extent that female leaders have internalized gender-stereotypic reservations about their capability for leadership, they may gain confidence by making collaborative decisions that they can determine are in line with their associ-ates' expectations. Thus, proceeding in a participative mode may enable many female leaders to overcome others' resistance, win their acceptance, gain self-confidence, and thereby be effective. The implications of women's more democratic and participative styles for their effectiveness are not clear-cut, in view of arguments that the effectiveness of these styles is likely contingent on features of the group or organizational environment (e.g., Vroom & Yetton, 1973). Meta-analyses reviewing the effects of democratic and autocratic leadership on group productivity and membership satisfaction have confirmed the importance of moderating conditions (Foels, Driskell, Mullen, & Salas, 2000; Gastil, 1994). These reviews have not, however, evaluated the hypothesis that democratic and participative styles may be especially effective for female leaders, because of the ambivalence that many people have about ceding power to women. Prejudice against female leaders should especially emerge in leadership roles that are male-dominated or regarded as requiring masculine qualities (Eagly & Karau, 2001). To examine the possible disadvantages of gender-incongruent leader roles, Eagly and Johnson (1990) developed measures of the congruity between gender roles and the leadership roles investigated in the studies included in their meta-analysis on leadership style. Their measures derived from students' Leadership Styles 791 ratings of each of these roles (e.g., ratings of how interested the average man and woman would be in occupying each role). Congruity effects emerged in this meta-analysis as well as in a subsequent meta-analysis of studies of the effectiveness of male and female leaders (Eagly, Karau,& Makhijani, 1995). Specifically, to the extent that a leader role was more congruent with the male than female gender role, men were more task-oriented than women and more effective in the role; to the extent that a leader role was more congruent with the female than male gender role, women were more task-oriented than men and more effective in the role. Evidently occupancy of a gender-incongruent leadership role is associated with leaders' lacking (or being perceived to lack) the skills necessary to organize effectively the task-relevant aspects of their environment. Gender-incongruent leaders, such as female military officers and male elementary school principals, may tend to lack the authority required to organize people and resources to accomplish the task-relevant goals that are inherent in their role. Transformational, Transactional, and Laissez-Faire Styles We investigated transformational, transactional, and laissez-faire styles of male and female leaders in a large sample of managers who had been assembled to provide norms for the most widely used measure of these styles, the Multifactor Leadership Questionnaire (MLQ; Center for Leadership Studies, 2000). These managers were predominantly from the United States, but the group included managers from eight other nations. Ratings of the managers (by managers' subordinates, peers, or superiors orby the managers themselves) indicated how frequently a manager engaged in the behaviors that are prototypical of the five subscales of transformational leadership, the three subscales of transactional leadership, and the one laissez-faire scale. In our current research we are also meta-analyzing a group of 47 studies that compared women and men on the MLQ and similar measures. As shown in Table 2, most of these measures of leadership style yielded small but significant sex differences in the norming sample. Women exceeded men on three transformational scales: the attributes version of idealized influence, inspirational motivation, and individualized consideration. These findings suggest that the female managers, more than the male managers, (1) manifested attributes that motivated their followers to feel respect and pride because of their association with them, (2) showed optimism and excitement about future goals, and (3) attempted to develop and mentor followers and attend to their individual needs. Women also exceeded men on the transactional scale of contingent reward. This finding suggests that the female managers, more than the male managers, gave their followers rewards for good performance. The largest of these differences in the female direction was on the individualized consideration scale, which has the most obviously communal content of these subscales. Table 2. Definitions of Transformational, Transactional, and Laissez-Faire Leadership Styles and Comparison of Female and Male Managers Type of MLQ scale and Description of leadership style Female leaders Male leaders subscale M SD M SD t d Transformational Idealized influence Demonstrates attributes that motivate respect and pride by 3.00 0.74 2.89 0.77 -6.70** -0.14 (attributes) association with him or her Idealized influence Communicates values, purpose, and importance of mission 2.77 0.78 2.75 0.74 -1.33 -0.03 (behavior) Inspirational motivation Exhibits optimism and excitement about goals and future states 2.93 0.76 2.90 0.77 -1.96* -0.04 Intellectual stimulation Examines new perspectives on problem solving and task 2.79 0.72 2.77 0.71 -1.28 -0.03 completion Individualized Focuses on development and mentoring of followers and attends to 2.94 0.79 2.76 0.80 -10.19** -0.23 consideration individual needs Transactional Contingent reward Exchanges rewards for satisfactory performance by followers 2.94 0.75 2.83 0.75 -6.70** -0.15 Active management- Attends to followers'mistakes and failures to meet standards 1.60 0.99 1.74 0.92 6.31** 0.15 by-exception Passive management-by- Waits until problems become severe before attending and 0.88 0.73 1.08 0.78 11.61** 0.26 exception intervening Laissez-faire Exhibits widespread absence and lack of involvement during 0.57 0.61 0.69 0.68 8.50** 0.18 critical junctures Note. Data are from a sample of managers (women = 2,874, men = 6,126) assembled to produce norms for the Multifactor Leadership Questionnaire (Center for Leadership Studies, 2000). Ratings are on a 5-point scale, ranging from 0 (not at all) to 4 (frequently, if not always). Positive fs and ds indicate male leaders higher than female leaders, and negative fs and ds indicate female leaders higher than male leaders. *p<.05. **/><.001. Leadership Styles 793 In contrast, men exceeded women on the transactional scales of active management-by-exception and passive management-by-exception and on laissez-faire leadership. These findings suggest that male managers, more than female managers, (1) paid attention to their followers' problems and mistakes, (2) waited until problems became severe before attempting to solve them, and (3) were absent and uninvolved at critical times. The largest of these differences in the male direction was on the passive management-by-exception scale. The relatively negative behaviors associated with the scales on which men exceeded women cannot, however, be regarded as typical of male managers, because raters perceived relatively low frequencies of these behaviors for both sexes, albeit higher frequencies for male than female managers. These findings have implications for the effectiveness of male and female leaders. In the norming sample, correlations between managers' rated effectiveness and these styles were positive and relatively large for contingent reward and all of the transformational subscales, rs (1,570) > .54. In contrast, these correlations were negative for passive management-by-exception, r (1,570) = -.28, and laissez-faire leadership, r (1,570) = -.36 (Center for Leadership Studies, 2000). Moreover, a meta-analysis of 39 studies confirmed these positive relationships of transformational leadership and the contingent reward aspect of transactional leadership to managers' effectiveness (Lowe, Kroeck, & Sivasubramaniam, 1996). Therefore, both women's higher scores on the transformational subscales and contingent reward and men's higher scores on passive management-by-exception and laissez-faire leadership suggest that the female managers in this norming sample were more effective than the male managers. In support of this conclusion, the women in the norming sample scored significantly higher than the men on a measure of perceived effectiveness. Why did women fare better than men on the measures of styles and effectiveness? One possible interpretation is that women have to meet a higher standard than men to attain leadership roles and have to maintain better performance to retain these roles. Substantiating this interpretation is research demonstrating the operation of a double standard in perceiving women as highly competent (see Biernat & Fuegen, this issue). In addition, men's greater likelihood of manifesting ineffective styles—namely, passive management-by-exception and laissez-faire leadership—suggests that men may have greater leeway to remain in leadership roles, despite poor performance. Another reason that women fared better than men on these measures may be the tendency for the female gender role to foster more feminine styles. Thus, individualized consideration and, to some extent, contingent reward may involve being attentive, considerate, and nurturing to one' s subordinates, tendencies that are consistent with the female gender role. Being encouraging and supportive of subordinates may foster showing optimism and excitement about the future, the tendencies assessed by the inspirational motivation subscale. Perhaps these qualities then 794 Eagly and Johannesen-Schmidt foster the respect and pride that are assessed by the idealized influence (attributes) subscale. Yet another possibility is that female managers may encounter resistance if they proceed in the more traditional command-and-control leadership styles, and they opportunistically discover the advantages of the more interpersonally sensitive but inspirational type of leadership that is captured by measures of transformational leadership (see Yoder, this issue). Conclusion Empirical research comparing the leadership styles of women and men yields a pattern of findings that is more complex than that generally acknowledged by social scientists or writers of popular books on management. Consistent with research comparing women and men on numerous social behaviors (Eagly et al, 2000), we have established that leadership style findings from experimental settings tend to be gender-stereotypic. In such settings, people interact as strangers without the constraints of long-term role relationships. Gender roles are moderately important influences on behavior in such contexts and tend to produce gender-stereotypic behavior. In addition, somewhat smaller, stereotypic sex differences appeared in assessment studies, in which people not selected for leadership responded to instruments assessing their leadership styles. Because respondents who were not under the constraints of managerial roles completed measures in these studies, some tendency for leadership styles to appear stereotypic was expected from the perspective of social role theory. When social behavior is regulated by leadership roles in organizational settings, it should primarily reflect the influence of these other roles and therefore lose much of its gender-stereotypic character. Indeed, Eagly and Johnson's (1990) findings for interpersonal and task styles supported this logic. However, gender-incongruent leader roles appeared to compromise leaders' task-oriented styles and their effectiveness. Also, women's leadership styles were more democratic than men's even in organizational settings, possibly reflecting the special legitimacy problems that female leaders face if they attempt to take charge in a clear-cut, traditionally hierarchical manner. On measures of transformational, transactional, and laissez-faire leadership styles, which were designed to predict effectiveness, yet another pattern appeared. Female leaders exceeded male leaders especially on the female-stereotypic transformational dimension of individualized consideration and scored higher than men on two additional subscales of transformational leadership as well as on the contingent reward scale of transactional leadership. In contrast, men exceeded women on the active and passive management-by-exception and laissez-faire subscales. It is likely that the greater effectiveness of female than male leaders in this sample of managers reflected the negative relationships of the passive management-by-exception and the laissez-faire styles to effectiveness and the positive relationships of the transformational and contingent reward styles to effectiveness. Leadership Styles 795 One consideration in interpreting our findings is that even the largest of these sex differences would be described by most social scientists as small. As Martell. Lane, and Emrich (1996) demonstrated, however, small differences, when repeated over individuals and occasions, can produce large consequences. Moreover, because investigators face many barriers to achieving well-controlled studies of leadership style, especially in organizational settings, uncontrolled variability would decrease the magnitude of any systematic effects, including those representing sex differences. Additional primary research is needed to clarify the mechanisms underlying these findings. Based on existing evidence, we suggested that two underlying processes may be especially influential: (1) the spillover of the female and male gender roles onto leadership behavior and (2) the prejudice women may encounter in leadership roles, especially if these roles are male-dominated or if women behave in an especially masculine style. One manifestation of this prejudice is the operation of a double standard by which women have to meet a higher standard of effectiveness to attain leadership roles and to retain them over time. Finally, the aspects of these findings that have the clearest implications for the effectiveness of female and male leaders pertain to transformational, transactional, and laissez-faire styles. Women's more transformational style and greater use of contingent reward as well as their lesser use of passive management-by-exception and laissez-faire style should enhance organizational effectiveness (see also Yoder, this issue). These findings thus resonate with the attention that journalists have given to the possibility that women are better managers than men. For example, an article in Business Week asserted that "after years of analyzing what makes leaders most effective and figuring out who's got the Right Stuff, management gurus now know how to boost the odds of getting a great executive: Hire a female" (Sharpe, 2000). Women's advantages in leadership style may sometimes be countered, however, by a reluctance, especially on the part of men, to give women power over others in work settings. Moreover, social and organizational changes place women, more often than men, in the position of being newer entrants into higher level managerial roles. As newcomers, women may reflect contemporary trends in management (see Fondas, 1997), including an emphasis on transformational leadership, that may threaten older, more established managers. A reluctance to allow women to ascend in organizational hierarchies may thus reflect resistance to changing managerial styles as well as a prejudicial tendency to evaluate women's leadership behavior less positively than the equivalent behavior of men (Eagly & Karau, 2001). Nonetheless, on the whole, research on leadership style has very favorable implications for women's increasing representation in the ranks of leaders. 796 Eagly and Johannesen-Schmidt References Bales, R. F.(1950). Interaction process analysis: A method for the study of small groups. Reading, MA: Addison-Wesley. Bass, B. M. (1990). Bass & Stodgill 's handbook ofleadership: Theory, research, and managerial applications (3rd ed.). New York: Free. Bass, B. M. (1998). Transformational leadership: Industrial, military, and educational impact. Mahwah, NJ: Erlbaum. Book, E. W. (2000). Why the best man for the job is a woman. New York: HarperCollins. Burns, J. M. (1978). Leadership. New York: Harper & Row. Cann, A., & Siegfried, W. D. (1990). Gender stereotypes and dimensions of effective leadership behavior. SexRoles, 23, 413-419. Carli, L. L., & Eagly, A. H. (1999). Gender effects on influence and emergent leadership. In G. N. Powell (Ed.), Handbook of gender & work (pp. 203-222). Thousand Oaks, CA: Sage. Center for Leadership Studies. (2000). MultifactorLeadership Questionnaire: Norms [On-line]. Available : http://cls.binghamton.edu/mlq.htm Eagly, A. H, & Johnson, B. T. (1990). Gender and leadership style: A meta-analysis. Psychological Bulletin, 108, 233-256. Eagly, A. H., & Karau, S. J. (In press). Role congruity theory of prejudice toward female leaders. Psychological Review. Eagly, A. H, Karau, S. J., & Makhijani, M. G. (1995). Gender and the effectiveness of leaders: A meta-analysis. Psychological Bulletin, 117, 125-145. Eagly, A. H, Karau, S. J., Miner, J. B., & Johnson, B. T. (1994). Gender and motivation to manage in hierarchic organizations: A meta-analysis. Leadership Quarterly, 5, 135-159. Eagly, A. H, Makhijani, M. G., & Klonsky, B. G. (1992). Gender and the evaluation of leaders: A meta-analysis. Psychological Bulletin, 111, 3-22. Eagly, A. H, Wood, W., & Diekman, A. B. (2000). Social role theory of sex differences and similarities: A current appraisal. In T. Eckes & H. M. Trautner (Eds.), The developmental social psychology of gender (pp. 123-174). Mahwah, NJ: Erlbaum. Ely, R. J. (1995). The power in demography: Women's social constructions of gender identity at work. Academy of Management Journal, 38, 589-634. Foels, R., Driskell, J. E., Mullen, B., & Salaš, E. (2000). The effects of democratic leadership on group member satisfaction: An integration. Small Group Research, 31, 676-701. Fondas, N. (1997). Feminization unveiled: Management qualities in contemporary writings. Academy of Management Review, 22, 257-282. Gastil, J. (1994). A meta-analytic review of the productivity and satisfaction of democratic and autocratic leadership. Small Group Research, 25, 384—410. Gutek, B. A., & Morasch, B. (1982). Sex-ratios, sex-role spillover, and sexual harassment of women at work. Journal of Social Issues, 38(4), 55-74. Hackman, M. Z., Furniss, A. H, Hills, M. J., & Paterson, T. J. (1992). Perceptions of gender-role characteristics and transformational and transactional leadership behaviors. Perceptual and Motor Skills, 75,311-319. Hall, J. A. (1998). How big are nonverbal sex differences? The case of smiling and sensitivity to nonverbal cues. In D. J. Canary & K. Dindia (Eds.), Sex differences and similarities in communication: Critical essays and empirical investigations of sex and gender in interaction (pp. 155-177). Mahwah, NJ: Erlbaum. Helgesen, S. (1990). The female advantage: Women's ways ofleadership. New York: Doubleday/ Currency. Hemphill, J. K, & Coons, A. E. (1957). Development of the Leader Behavior Description Questionnaire. In R. M. Stogdill & A. E. Coons (Eds.), Leader behavior: Its description and measurement (pp. 6-38). Columbus, OH: Ohio State University, Bureau of Business Research. Kanter, R. M. (1977). Men and women of the corporation. New York: Basic. Kimball, M. M. (1995). Feminist visions of gender similarities and differences. Binghamton, NY: Haworth. Leadership Styles 797 Lewin, K., &Lippitt, R. (1938). An experimental approach to the study of autocracy and democracy: A preliminary note. Psychometry, 1, 292-300. Lowe, K. B., Kroeck, K. G., & Sivasubramaniam, N. (1996). Effectiveness correlates of transformational and transactional leadership: A meta-analytic review of the MLQ literature. Leadership Quarterly, 7, 385-425. Martell, R. F., Lane, D. M., & Emrich, C. G. (1996). Male-female differences: A computer simulation. American Psychologist, 51, 157-158. Miller, D. T., Taylor, B., & Buck, M. L. (1991). Gender gaps: Who needs to be explained? Journal of Personality and Social Psychology, 61, 5-12. Miner, J. B. (1993). Role motivation theories. New York: Routledge. Moskowitz, D. W., Suh, E. J., & Desaulniers, J. (1994). Situational influences on gender differences in agency and communion. Journal of Personality and Social Psychology, 66, 753-761. Powell, G. N. (1990). One more time: Do male and female managers differ? Academy of Management Executive, 72,731-743. Ridgeway, C. L. (1997). Interaction and the conservation of gender inequality: Considering employment. American Sociological Review, 62, 218-235. Rosener, J. B. (1995). America's competitive secret: Utilizing women as a management strategy. New York: Oxford University Press. Ross, S. M., & Offermann, L. R. (1997). Transformational leaders: Measurement of personality attributes and work group performance. Personality and Social Psychology Bulletin, 23, 1078-1086. Schroeder, P. (1999, February 22). And please call me Ms. President. New York Times, p. A17. Sharpe, R. (2000, November 20). As leaders, women rule. Business Week, p. 74. Thrall, P. (1996, July 2). Success and sharp elbows: One woman's path to lofty heights on Wall Street. New York Times, p. C4. Vroom, V. H., & Yetton, P. W. (1973). Leadership and decision-making. Pittsburgh, PA: University of Pittsburgh Press. Wagner, D. G., & Berger, J. (1997). Gender and interpersonal task behaviors: Status expectation accounts. Sociological Perspectives, 40, 1-32. Wood, W., Christensen, P. N, Hebl, M. R., & Rothgerber, H. (1997). Conformity to sex-typed norms, affect, and the self-concept. Journal of Personality and Social Psychology, 73, 523-535. ALICE H. EAGLY is Professor of Psychology at Northwestern University. She has published widely on the psychology of attitudes, especially attitude change and attitude structure. She is equally devoted to the study of gender. In both of these areas, she has carried out primary research and meta-analyses of research literature. She is the author of Sex Differences in Social Behavior: A Social Role Interpretation and (with Shelly Chaiken) The Psychology of Attitudes and the coeditor of two volumes. Eagly is also the author of numerous journal articles, chapters, notes, and reviews in her research specialities. MARY C. JOHANNESEN-SCHMIDT is a graduate student in the social psychology program at Northwestern University. She has studied the content of income stereotypes and their differential effects upon perceptions of men and women, moderators of sex differences in preferred mate characteristics, and stereotypes of men's and women's leadership styles.