Leadership Development: Born/Made 1 страница
We now explore an area of leadership where more has been written than perhaps in any other area. Unfortunately, volume does not correlate with quality. Numerous popular books profess to have the solution to developing high potential leadership. However, as House and Aditya (1997) concluded, "That management training and development efforts will result in improved management appears to be taken as an article of faith by many organizations, professional management associations and consultants. Yet, despite the immense amount of investment in management training on the part of corporations and government, there is little evidence that such training results in more effective management behavior" (p. 459). Because only 10% of all leadership development programs are evaluated beyond participants' satisfaction with the program, it is premature to say that leadership development can or cannot be developed. For example, Burke and Day (1986) completed a meta-analysis of 70 different management training studies and concluded that there were positive and negative effects for knowledge development. House and Aditya (1997) cited preliminary evidence supporting the positive effects of training that were associated with use of Graen's LMX.
Leadership Development in Context
When evaluating leadership development, it is important to consider how the context promotes or inhibits the transfer of training effects. For example, supervisory training not sun-ported by the management culture of the organization has resulted in higher role conflict and stress and lower job performance (Fleishman, Harris, & Burtt, 1955; House, I960-Sykes, 1962).
Most leadership training fails to recognize that leadership constitutes a complex interaction between leaders, followers and the context (Day, 2000; Fiedler, 1996). Day made a useful distinction between leader development and leadership development. Leader development has the primary goal of enhancing an individual's capacity and potential (H. Gardner, 1983, 1985). Day provided specific examples focusing on areas such as self-awareness, self-regulation, and self-motivation, citing the work of McCauley (2000, 2001). Leadership development focuses on the interaction of the leader within a social-organizational context, an area repeatedly neglected in past leadership development research (Fiedler, 1996; Zaccaro & Klimoski, 2001). Day concluded that "leadership development can be thought of as an integration strategy by helping people understand how to relate to others, coordinate their efforts, build commitments, and develop extended social networks by applying self-understanding to social and organizational imperatives" (p. 10).
Starting with the context, individuals can be developed based on their job assignments and responsibilities. Challenging new assignments can be strategically used to develop the potentials of leaders (McCauley & Brutus, 1998). What leaders learn from both positive and negative experiences on the job represents a fruitful area for future research. Indeed, many companies such as Coca Cola, General Electric, and Citibank strategically utilize work assignments as a way of building individual and collective leadership potential. Research done by McCauley, Ruderman, Ohlott, and Morrow (1994) showed that challenging work assignments are correlated with on-the-job learning, but such learning has not been empirically linked to leader or leadership development. Thus, does leadership develop differentially over time as one moves through various stages of life (Erikson, 1968; Kuhnert & Lewis, 1987)7
Self-Concept andLeadership Development
The manner in which a leader views himself or herself, behaves, and influences followers may stem from an awareness of the leader's self-concept or identity (Hanges, Lord, & Dickson, 2000), self-goals, possible selves (Lord, Brown, & Freiberg, 1999; Sosik, 2000), self-awareness, self-regulation
(W. L. Gardner & Avolio, 1998; Sosik & Dworakivsky, 1998), familial issues (Simonton, 1994), maturation (Erikson, 1968), dependence versus independence (Conger, 1999), defense mechanisms, and repression of a shadow self (Kets de Vries, 1988). To develop fully, leaders must develop self-awareness and acceptance to understand how to interact with followers (Goleman, 1998; Kets de Vries, 1988).
360-Degree Survey Feedback and Leadership Development
One of the central features of most leader development programs is the use of 360-degree feedback systems (Waldman & Atwater, 1998). However, relatively little is known about their impact on creating leader self-awareness and development (Atwater, Waldman, Atwater, & Carrier, 2000). Atwater and her associates offered some positive evidence for the im-
| pact of upward feedback on leader development, as did Hegarty (1974). However, why feedback had a positive impact had not been determined in prior research. As Atwater and her associates argued, improvements could stem from an awareness of self-other rating differences, from highlighting dimensions of leadership that focus one's efforts, from motivational pride to close the gap between self-other ratings, or from some combination of factors. On the negative side, Kluger and DeNisi (1996) reported that in over one third of the cases of providing feedback, performance was actually reduced after feedback.
Atwater et al. (2000) reported that in an organization where cynicism was higher, the level of positive change following feedback was lower: "This finding suggests that cynicism may have contributed to the ratings of leadership that supervisors received after the initial feedback, rather than
^cynicism being related to concurrent ratings of leadership at Time 1" (p. 287).
Impact of Leadership Training
Although only a small number of studies have examined how training can impact neo-charismatic or transformational lead-ership development, several results are worth noting. Avolio and Bass (1998) reported the results of a field study of community leaders that went through a 1-year training interven-tlon, using the full-range leadership model. There were s,gmficant and positive changes in followers' evaluations of ransformational leadership for those participants who cre-ated leadership development plans that were independently ^ed as having clear, specific, and measurable goals, ^rookall (1992) conducted a training study with prison °P supervisors to compare transformational to situational ership training. Group 1 received a transformational
Emerging Areas in Leadership Theory and Research 291
training program, whereas Group 2 received situational training and Group 3 received no training. Both trained groups pre-to posttraining intervention improved on the order of 10% to 50% depending on the criterion measures. Transformational leadership training had a more positive impact on personal growth and performance while improving inmates' respect for supervisors, skills development, and good citizenship behaviors. Barling, Weber, and Kelloway (1996) examined the impact of transformational leadership training with bank managers in a field experiment conducted in a large financial institution. Using a coaching model, they reported that certain aspects of transformational leadership improved as well as managerial performance. Dvir, Eden, Avolio, and Shamir (2002) completed a true field experiment with Israeli platoon commanders randomly assigned to two versions of transformational leadership training. Both programs involved 3 days of training. The newer or experimental transformational leadership program also included a 3-hr booster session approximately 1 month after the close of the first session. The booster session was used to coach individual commanders on leadership development and self-reflection. Results showed significant differences in what was learned about transformational leadership, changes in transformational behavior, and performance effects, which were all in the predicted direction. Transformational leadership ratings increased over time for the group going through the experimental transformational leadership program. Six months following the close of training, groups going through the experimental transformational leadership training had significantly higher performance.
A growing body of evidence suggests that transformational leadership can be developed using in vitro methods. We have also learned that some leadership styles may be learned in vivo. For example, Klonsky (1983) reported that parental warmth, discipline, and achievement demands predicted the type of leadership behaviors observed among high school students. Cox and Cooper (1989) reported that many successful British CEOs experienced an early loss of a parent or were separated from them at an early age. Avolio and Gibbons (1988) used life history interviews of executives and concluded that those who were evaluated as more transformational had parents who set high standards and encouraged them to do their best, came from family circumstances that were challenging, and learned from parents to deal more effectively with disappointment and conflict. Zacharatos, Barling, and Kelloway (2000) examined relationships between parental leadership style and the leadership of children based on self, peer, and supervisor ratings. Results showed that perceptions of one's father's transformational leadership had a significant relationship with self and other transformational leadership ratings.
292 Leadership Models, Methods, and Applications
There is a shift occurring in the field of leadership development where research and practice are focusing more on how people learn within their work context to be more effective leaders (McCauley, 2001; Moxley & O'Connor-Wilson, 1998). Work on coaching and mentoring is a popular emerging area. Coaching has typically been described like mentoring as a longer-term process that focuses on both the context and the individual (Day, 2000; Kilburg, 1996). Like other areas of leader and leadership development, however, there is still very little research to support the effectiveness of coaching (Day, 2000; Kilburg, 1996). There is also a thin literature base demonstrating the effectiveness of mentoring programs (Day, 2000). Research comparing formal to informal mentoring programs concluded that there are more positive benefits for informal mentoring relationships (Ragins & Cotton, 1999). Additional research indicates that gender and racial differences need to be explored in terms of the effectiveness of mentoring and coaching.
Thus, the collective evidence suggests that leaders and leadership can and are developed over time, both in natural settings as well as in workshops. After nearly 60 years of research on leadership development, the field is finally getting around to answering a question that represents one of the core reasons for studying leadership: Can we develop it over time?
Examining Leadership Across Cultures
Culture is a part of the social context in which leadership is embedded, and it is expected to moderate and mediate leader and follower interactions. Culture is a mindset that emerges through social interaction and is transmitted and diffused through the interaction among individuals (Hofstede, 1980, 1983; Triandis, 1994). Over the years, studies have examined a broad range of questions concerning the linkage between leadership and culture: Are certain leader behaviors and styles culturally universal? Do theories of leadership developed in the United States generalize to other cultural settings? House, Hanges, Agar, and Quintanilla (1995) argued that answers to these questions could provide organizations with a strategic advantage for developing a diverse range of future leaders (for additional reviews see House, Wright, & Aditya, 1997; Peterson & Hunt, 1997).
Single-Culture and Single-Country Research to Multiculture and Multicountry Research
Single-culture and single-country research aims to replicate specific leadership theories or models in other cultural settings. For example, Shamir, Zakay, Breinin, and Popper (1998) examined charismatic leadership behavior and its
effects on followers' attitudes and leadership effectiveness in the Israeli military. Multiculture and multicountry research examines arguments concerning whether a particular leader^ ship theory, model, or style generalizes across cultures. The GLOBE project initiated by House and his international research team (Den Hartog, House, Hanges, Ruiz-Quintanilla, & Dorfman, 1999) represents this second category of cross-cultural leadership research. Preliminary evidence cited earlier showed that attributes associated with charismatic leadership were universal (Bass, 1997).
A third category of cross-cultural leadership research has examined the effects of cultural diversity within group or team settings, testing how differences in ethnicity of leaders and followers affected the perceptions and outcomes of leadership within those groups and teams. This research examines how certain styles of leadership affect followers' motivation, effectiveness, and performance when followers are ethnically or culturally different from their leader or when in culturally heterogeneous groups (Hooijberg & DiTomaso, 1996). Such research has compared how different leadership styles displayed within different ethnic groups wimin a single culture affected follower perceptions, motivation, and performance (Jung &Avolio, 1999).
Cultural Values and Leadership
Over the last two decades, cross-cultural research has focused primarily on four cultural dimensions (see Hofstede, 1980, 1993). These dimensions included power distance, uncertainty avoidance, individualism/collectivism, and masculinity/femininity. These four cultural dimensions have been used to identify potential boundary conditions for leadership theories that have been applied across cultures (Dorfman, 1996). For example, a leadership theory that argues for a democratic leader as an ideal style of leadership may not generalize to cultures where an unequal distribution of power is accepted as the norm (Dorfman & Howell, 1988; Jung, Bass, &Sosik, 1995).
Key Results of Research on Cross-Cultural Leadership
Early cross-cultural leadership studies focused on task- and relationship-oriented leadership styles. This research included a wide range of samples from the Philippines and China (Bennett, 1977), Japan (Misumi, 1985; Misumi & Peterson, 1985), India (Sinha, 1984), New Zealand (Anderson, 1983), Mexico (Ayman & Chemers, 1991), and Israel (Fleishman & Simmons, 1970), among many other cultures and nations, and has reached two general conclusions: (a) The most effective leadership styles are a combination of high relationship- and
high task-oriented leadership, and (b) although general patterns of leadership are similar across cultures, specific behaviors and attitudes expressed by leaders appear to differ across cultures (Bass, 1997).
Prior cross-cultural research has also tested social identity and LMX theory. Pelled and Xin (1997) investigated the effect of leader-member demographic similarity on followers1 organizational attachment to Mexican organizations, with leaders coming from a high power distance culture. They reported that Mexican employees, who had a small age gap with their superior, were less likely to be absent and were more attached to their organization. These same followers exhibited lower commitment to their work while expressing higher levels of comfort working for a younger supervisor. Farh, Tsui, Xin, and Cheng (1998) examined the effects of relational demography and personal network ties called Guanxi in China on leader-follower relationships and reported that ^ demographic similarities among leaders and followers had a " positive impact on follower trust in the leader.
Recent cross-cultural leadership research has focused on testing neo-charismatic models of leadership. For instance, Shamir et al. (1998) reported mixed support for the relationship between charismatic leadership and Israeli followers' attitudes toward their respective leader. Only one of three charismatic behaviors (emphasizing units' collective identity) was positively related to followers' identification with and trust in the leader, whereas the other two behaviors (i.e., supportive behavior and ideological emphasis) were either unrelated or negatively related.
Contrary to the Shamir et al. (1998) results, the majority of cross-cultural research on charismatic-transformational leadership has supported the hierarchy of leadership effectiveness styles comprising Bass and Avolio's (1994) model of leadership. Koh, Terborg, and Steers (1991) reported that transformational leadership of Singaporean school principals had significant add-on effects to transactional leadership in predicting organizational commitment, citizenship behavior, and teacher satisfaction. Similar augmentation effects have been reported in a wide variety of samples, including Canadian (Boyd, 1988; J. M. Howell & Avolio, 1993), Mexican (Dorfman & Howell, 1988; Echavarria & Davis, 1994), Italian (Bass & Avolio, 1994), and Danish (Den Hartog, Van Muijen, & Koopman, 1994). Overall, the positive effects that transformational leadership has on follower's motivation and performance have been well documented in other cultural settings, including Austria (Geyer & Steyrer, 1998), Israel (Popper, Mayseless, & Castelnovo, 2000), and Korea (Cho, 1999; Jung, Butler, & Baik, 2000).
Results of multiculture and multicountry research are equally interesting. For example, Dorfman and Howell (3 988)
Emerging Areas in Leadership Theory and Research 293
reported different effects associated with charismatic leadership when comparing Mexican versus American employees. Charismatic leadership had a strong positive relationship with both Mexican and American employee satisfaction levels; however, the relationships were much stronger for American versus Mexican employees. Dorfman et al. (1997) reported that among six leadership behaviors examined across five countries, leader support] veness, contingent reward, and charisma had universally positive relationships with followers' level of satisfaction, whereas participative, directive, and contingent punishment had positive relationships in only two cultures.
Fu and Yukl (2000) compared the perceived effectiveness of influence tactics in the United States and China, reporting that American and Chinese managers favored different influence tactics. American managers rated rational persuasion and exchange as more effective styles compared with their Chinese counterparts. Chinese managers rated coalition building, upward appeals, and gifts as more effective than did American managers. Rao, Hashimoto, and Rao (1997) examined influence tactics employed in Japan and reported that Japanese managers used many of the same influence tactics as did their American counterparts but differed on behaviors.
Valikangas and Okumura (1997) examined differences in follower motivation, comparing leaders in the United States and Japan. They concluded that leadership in the United States was based on followers' utility expectations (i.e., a "right" agency will result in the "right" outcomes). Leadership in Japan was based on identity expectations (i.e., a "right" group or corporate identity will result in "right" behaviors among followers).
How do people in different cultures perceive their ideal leaders? Using an attribute-rating task in which people rated a list of attributes according to how well each attribute fit their prototype of leaders, Gerstner and Day (1994) examined leadership prototypes across eight countries. Gerstner and Day reported that to be determined was a prototypical attribute of leaders in Western countries, whereas intelligence was considered highly prototypical in Asia. Den Hartog et al. (1999) reported that a number of attributes associated with charismatic-transformational leadership were considered "ideal" attributes of leadership across 62 different countries. The following leadership attributes were endorsed across cultures: encouraging, positive, motivational, confidence builder, dynamic, and foresight. On the other hand, being a loner, noncooperative, ruthless, nonexplicit, irritable, and dictatorial were seen as negative facets of leadership (Conger & Hunt, 1999).
Brodbeck et al. (2000) reported differences in leadership prototypes across 22 European countries. Their sample
294 Leadership Models, Methods, and Applications
included middle-level managers (TV = 6,052), who rated 112 traits and behaviors in terms of how well they represented outstanding business leadership. Some leadership concepts were culturally endorsed and grouped according to the values representing a cluster of nations. Interpersonal directness and proximity were more strongly associated with outstanding leadership in Nordic countries versus Near East and Central European countries. Autonomy was associated with outstanding leadership in Germanic countries. Wofford, Lovett, Whittington, and Coalter (1999) argued that differences in leadership prototypes may be a function of the type of organization from which raters are pooled. For example, with raters coming from the United States and Mexico, Wofford et al. (1999) reported that differences in leadership prototypes in their investigation were due to the type of institutions from which participants were sampled, as well as to their national cultures. Xin (1997) reported that different impression-management tactics were used by Asian American and Caucasian American managers. Specifically, Asian American managers used more job-focused (e.g., pointing out past accomplishments to my supervisor) and more supervisor-focused (e.g., offering to do things for my supervisor that are not formally required) impression-management tactics and less self-disclosure (e.g., expressing my feeling to my supervisor). Jung and Avolio (1999) reported that Asian Americans generated more ideas when they worked with a transformational leader in a brainstorming task, whereas Caucasian Americans performed better when working with a transactional leader. Asian Americans also performed better when they worked alone than when they worked as a group generating ideas requiring radical changes.
A common theme running through many cross-cultural studies of leadership concerns how cultural values moderate or mediate relationships between a leader and his or her followers. Support for both cultural-universal and culture-specific aspects of leadership have been provided in prior research. However, several important issues must be addressed to advance the field. First, Hunt and Peterson (1997) pointed out the need to define and measure leadership constructs in a similar way across different studies. Leadership has been measured based on various methods, including the use of influence tactics (e.g., Fu & Yukl, 2000), prototype attributes {e.g., Brodbeck et al., 2000), charismatic and transformational leadership behaviors (Shamir et al., 1998), and motivational differences (Valikangas & Okumura, 1997). Second, there has been insufficient attention to defining the level of analysis at which cultural differences should be examined (Hunt & Peterson, 1997). For example, different types of leadership effectiveness and outcome measures have been compared across different levels of analysis in organizations, making it difficult to discern the impact of
cultural differences reported in this literature. Third, there has been a high degree of variance in the quality of methods used by different researchers. For example, some researchers hired professionals to translate and back-translate their survey instruments to ensure functional and conceptual equivalence whereas others did not use professional translation. Fourth, by and large, culture has generally been included in studies as either a moderator or mediator variable (Hunt & Peterson, 1997), Most prior research has not treated culture as a fundamental variable that drives the relationship between leaders and followers (House, Wright, & Aditya, 1997). Finally, the majority of cross-cultural research has relied on quantitative, survey-based research designs, and more qualitative, unobtrusive research methods are now needed to examine how culture shapes leadership.
Strategic Leadership Research
Beginning with Barnard (1938), research on strategic leadership has concentrated on identifying best practices that contribute to firm success. Recent research has focused on internal firm characteristics (Hoskisson, Hitt, Wan, & Yiu, 1999), either in the form of agency contracts (Jensen & Meckling, 1976) or by focusing on a firm's unique resources. Both approaches take into consideration the role of leaders as representing strategic assets for the firm.
Strategic leadership researchers have argued that organizations become a reflection of their top managers (Hambrick & Mason, 1984; Klimoski & Koles, 2001). Hambrick (1989) emphasized the importance of strategic leadership but also recognized that its impact may be more indirect. Moreover, top managers face ambiguous environments and often experience information overload (March & Simon, 1959). Under these conditions, successful leadership is determined by the frame of reference used by decision makers, which includes their personal background, experiences, education, and other biographical characteristics. Strategic leadership helps to coordinate and maintain organizational systems, while readying it for adaptive change.
Beginning with Kotter (1982) and Hambrick and Mason (1984), strategic leadership research has focused on personal and background characteristics of executives related to firm success. According to Hambrick and Mason, personal characteristics together with environmental constraints and organizational factors constitute the leader's "discretion" (Cannella & Monroe, 1997). The amount of discretion employed by top managers moderates the relationship between their strategic choices and organizational outcomes. Cannella and Monroe criticized the overreliance on using biographical data as predictors of strategic performance. They argued that
strategic leadership theory relies too much on descriptive variables to explain choices that executives make while providing little guidance on how to include the process of leadership in its research designs. Here is where some of the work on the neo-charismatic theories of leadership (e.g., Bass, 1985; House, 1995) may contribute to strategic leadership theory by focusing on how the leader's think, behave, and are affected by the context.
Integrating Transformational and Strategic Leadership
The neo-charismatic theories of leadership focus on interpersonal processes between leaders and followers (House & Aditya, 1997). These theories focus on the process of leadership within organizations, although they could also be applied to leadership across organizations. Transformational or charismatic leaders have followers who emulate them and
^perform beyond expectations (Avolio, 1999; Bass, 1985). Their followers may be more open to shifts in their worldview and to accepting new values and changes in thinking and strategy (Boal & Bryson, 1988). Transformational leaders often convey their ideas using a strategic vision for the organization that includes strategic goals presented in a future-oriented optimistic framework (Berson et al„ 2001). As noted by Zaccaro and Banks (2001), "a fundamental requirement of organizational leadership is setting the direction for collective effort on behalf of organizational progress" (p. 181).
The turbulence that characterizes today's environment dictates constant transformation and even radical change for organizations. The process of radical change begins with a strategic vision that leaders have for their organization. A vision is an outline of a strategic and lofty action plan or a guideline to the "new way of doing things" following the transformation (Nutt
k& Backoff, 1997). A comprehensive vision can help to align the views of multiple stakeholders, which is critical to change and success (Nutt & Backoff, 1997).
Although it seems that strategic leaders would benefit from a charismatic or transformational style, strategic leadership theorists have taken a different position on this issue. They argue that charisma may narrow the executive's information processing orientation, thereby restricting the range of strategic choices (Finkelstein & Hambrick, 1996). Specifically, Pinkelstein and Hambrick suggested that charismatic leaders are more likely to receive filtered and distilled information from their followers and may be less aware of information that contradicts their visions. This occurs when followers are threatened by the charismatic leader's ability to "see the future" and are hesitant to offer ideas that conflict with their fader's vision. Alternatively, Cannella and Monroe (1997) ^gued that charismatic-transformational leadership could