International Journal of Business and Management August, 2006 A Critical Review of Leadership Research Development Jun Liu Business School, Renmin University of China, Beijing 100872, China Xiaoyu Liu School of Labor Relations & Human Resources, Renmin University of China, Beijing 100872, China Abstract Leadership research has gone through several phases of development in the past 80 years or so. The paper identifies the major theories in each phase, and investigates the strengths and weaknesses of the research.
Among those theories, transformational/charismatic leadership and leader-member exchange (LMX) are heavily discussed. The paper also discusses the future trend of research in leadership areas. Keywords: Review, Transformational Leadership, LMX Leadership can be defined as a process that an individual influences a group of individuals to implement strategies and achieve the collective goals (Yukl, 2002). Research in the area seeks to find out what types of leaders are likely to be successful and what factors determine leadership effectiveness.
Generally, leadership research has gone though four paradigms: trait approach, behavioral approach, situational approach, and the contemporary theories of leadership. The paradigms of leadership theories shift with the progress from static to dynamic view of leadership with the trait and behavioral theories reflecting a personal, the situational theories an interpersonal, and the contemporary theories a relational approach to conceptualizing leadership. That is to say leadership has been examined as intrapersonal competencies, interpersonal processes, and relational dynamics.
More specifically, I review those approaches, their strengths and weaknesses as followed: The trait approach The trait theories assumed that leaders were born, not made. In the literature, Stogdill (1948, 1974) completed two comprehensive reviews by synthesizing more than 200 studies of the trait approach. His two surveys identified a group of traits that were positively associated with leadership such as intelligence, self-confidence, initiative, and persistence. However, Stogdill concluded that no combination of traits would guarantee leadership effectiveness.
An individual does not become a leader solely because he or she possessed certain traits. Rather, the traits have to be relevant to situations in which the leader is functioning, thus situation should also be a part of leadership. Moreover, his studies showed that leadership was not a passive and static state but resulted from a working relationship between the leader and other group members. In essence, Stogdill’s research invoked the development of the later leadership approaches. The behavioral approach Behavioral theories focus on leader behaviors and assume that effective leaders are common in their behavior modes.
Most of behavioral theories relied on the Ohio and Michigan’s studies (Stogdill, 1948), which described leadership styles based on the two dimensions of initiating structure (concern jobs and tasks, often called task-oriented behavior) and considerations (concern people and interpersonal relationships, often called relation-oriented behavior). The simple two-factor model provided a good starting point and basis for later researchers to conceptualize leadership. For example, Blake and Mouton (1964) joined the two dimensions in a model called “Leadership Grid” to describe various leadership styles.
To enrich the two-factor model, later theorists also identified some supplementary leader 3 International Journal of Business and Management August, 2006 behaviors, for example, the participative behavior involving power sharing, delegating, and empowering (Lewin, Lippitt, and White, 1939; Miller and Monge, 1986). A key research issue in the behavioral approach is the influence of the two behavioral dimensions on organizational outcomes. Task-oriented behavior is found positively associated with subordinate performance, whereas relation-oriented behavior is related to subordinate satisfaction.
Relation-oriented behavior has a moderating effect on the relationship between task-oriented behavior and performance. For example, Schriesheim and Murphy (1976) reported that task-oriented leadership without personal attention to group members might have negative effects on satisfaction and even on performance. Theorists generally agree that the two behavioral dimensions have additive effects on outcomes and “high-high” leaders who concern for both tasks and relations were expected to be more effective, though the statement has received support from only a few studies (e. . Misumi, 1985). Critique to the trait and behavioral approach The trait and behavioral approach are drawn from a personal construct perspective, which assumes that leadership occurs when a leader “express leadership” towards followers. The trait approach believes that leadership is mainly a personal attribute, better leadership results from developing the personal competencies of leaders (Day, 2000). The behavioral approach extends the perspective by focusing on what leaders do rather than simply telling who the leaders are.
The significance of the two approaches should not be underestimated because they are intuitively appealing and, with a great deal of research validating the bases of these perspectives, the approaches provide an in-depth understanding of the leader components in the leadership process (Bass, 1985). There are obvious limits with the two approaches. For example, trait approach is criticized to not provide very useful implications for training and developing leadership due to the nature of traits (Bass, 1985). Furthermore, as stated above, situation influences leadership.
It is thus difficult to identify a universal set of leader traits for various contexts. Even some traits can help differentiate leaders from non-leaders, it is theoretically difficult to link traits to leadership outcomes such as follower motivation and group productivity (Yukl and Van Fleet, 1992). For the weaknesses of the behavioral approach, one criticism is that the two types of behaviors are too abstract to describe complex leadership styles; another refers to the inconsistent link between task- and relation-oriented behavior and outcomes such as morale, job satisfaction, and productivity.
It is not clear how leaders’ styles are associated with performance outcomes (Bryman, 1992). The third criticism, being shared with the trait approach, is that the approaches pay too much attention to the leader perspective without considering followers and situations in which leaders exhibit their qualities and behaviors. It seems that leadership effectiveness can not be well understood unless interactions among leaders, followers and situations are studied.
The situational approach Incorporating findings obtained from the previous studies, the situational approach assumes that there is no one best way that is transcendent across all situations, and further, the approach tries to discover the situational moderating variables that influence the effects of leader behaviors. The situational leadership model (Hersey and Blanchard, 1982), the contingency model (Fiedler, 1967), and the path-goal theory (House and Mitchell, 1974) are the representative situational theories.
The theories have somewhat different emphases regarding their basic arguments. Specifically, situational theory emphasizes leadership flexibility – leaders should find out about their subordinates’ maturity (job and psychological) and task characteristic and then adapt their styles accordingly (Hersey and Blanchard, 1982). In the contingency model, however, leader behaviors are much more consistent and less flexible. Leadership effectiveness is 4 International Journal of Business and Management August, 2006 rimarily determined by selecting the right leader for a certain situation (characterized by assessing leader-member relations, task structure, and position power; Fiedler, 1967) or by changing the situation to fit the particular leader’s style. House’s path-goal model, being more complex than the other two, assumes that leadership styles vary not only with different subordinates but also with the same subordinates in different situations. On the other hand, the theories clearly share some common points. First, all the theories contain situational moderating variables.
Second, they implicitly assume that leaders can properly assess pivotal follower and situational factors. Third, leaders make their behaviors contingent on the followers and the situation. Compared to the former approaches, situational models conceptualize leadership as an interpersonal process, concern with follower, task, and situational variables rather than focus only on the leader perspective. By addressing all the factors involving the leadership process, the approach provides a more comprehensive picture of the nature of effective leadership.
The models share a number of weaknesses that limit their implications. First, the theories contain situational moderator variables, but the variables are often defined too ambiguously to operationalize. For example, in the situational leadership model, it is even unclear how “subordinate commitment” is combined with “competence” to form four distinct levels of “subordinate development” (Hughes, Ginnett, and Curphy, 1996). Similarly, task structure and task complexity have been ambiguously defined and measured in different ways. As a consequence, researchers could hardly generate specific and testable hypotheses.
Second, the theories assume that leaders can properly identify the characteristics of followers and situation. However, the assumption is not realistic and it is very likely that different leaders in the same situation may conclude distinctively in regard to followers’ level of knowledge, maturity, leader-follower relationships, the degree of task structure, or the level of role ambiguity being experienced by followers (Vroom and Jago, 1995). These differences in leader perception could, in turn, cause the leaders to take very different actions in response to the same situation.
Third, the approach tends to view leadership as an interpersonal process, but it lacks specific analyses on the dynamics of the influence process. Rather, they tend to view leadership as a “passive” process, leadership styles have to “fit” the followers and situations to obtain effectiveness. Leadership substitutes theory (Kerr and Jermier, 1978), being unique in its conceptualization, severely questions the conventional leadership wisdom by describing two types of variables (substitutes and neutralizers) that reduce the importance of formal leaders in organizations.
Substitutes make formal leader unnecessary and redundant, and neutralizers prevent a leader from acting in a specified way or nullify the effects of the leader’s actions. Theorists tend to classify the theory into situational approach due to its focus on the factors (subordinate, task, and organizational characteristics) concerned by situational theories. On the other hand, Jermier and Kerr (1997) contend that the framework of leadership substitutes should be less viewed as a situational model, which calls for treating the traditional context as the independent variable to explain individual and group effectiveness.
It can be believe that such efforts will complement our understanding of leadership functioning and its effectiveness. Contemporary leadership theories Unlike traditional approaches that view followers as unchanged or part of the situation, contemporary leadership theories treat followers as the counterpart of the leader and seek for positive transformation, dynamic relationships, and relational association with followers. The models try to provide insights into leadership effectiveness and implications to leadership development as well.
Of the many contemporary theories, transformational/charismatic leadership and leader-member exchange (LMX) have been most heavily studied. 5 International Journal of Business and Management August, 2006 The transformational/charismatic theory Rather than concerns with all levels of supervisors in the organization, the transformational/charismatic theory shifts its concern to the upper leaders by separating transformational leadership (leaders) from transactional leadership (managers) (Burns, 1978).
In the past twenty years, the stream of research generated large numbers of both conceptual and empirically studies with several major research issues having been addressed: First, specific leader behaviors reported by followers are used in this approach to define the transformational/charismatic leadership. In other words, the theories seem to rely heavily on the assumption of “one best way” by seeking transcendent leadership styles. For example, the four I’s (idealized influence, individualized consideration, inspirational motivation) have been most frequently examined in this stream of research.
Furthermore, being different from traditional view, charisma is also conceptualized as one dimension of leader behavior, which is based on followers’ perceptions, although some dissenting opinions exist regarding the definition (Trice and Beyer, 1986). Second, the new approach argues that leaders, being different from managers, have to motivate followers rather than arbitrarily impose leader behaviors on followers or passively fit leadership styles to followers and situations.
The new leadership seeks to achieve high level of motivational outcomes by transforming/changing follower self-concepts. For example, House and his colleagues (House and Shamir, 1993; Shamir, Zakay, Breinin, and Popper, 1998) proposed mechanisms that leaders use to transform follower self-concepts, which includes offering an appealing vision; making task and mission meaningful; developing a collective identity; and heightening both individual and collective self-efficacy. In the literature, the vision advocated by leaders received a large amount of attention (e. . , Bass, 1996; Shamir et al. , 1993). Third, compared to the above-mentioned approaches, the new approach better describes the underlying influence process. For example, Bass and Avolio (1993) proposed an influence process that transformational leaders evoke and meet followers’ high-order needs, which in turn promote commitment and performance. Shamir and associates’ (1993) research explained the reciprocal nature of mutual influence, in which leaders choose a vision that is congruent with followers’ values and identities.
Reciprocally, followers select the individuals as their leaders who would like to espouse their values. Empirically, many studies have included psychological intervening variables (i. e. , psychological empowerment, social identification, role characteristics) between leader behaviors and outcome variables to examine the dynamics of influence (e. g. , Kark, Shamir, and Chen, 2003; Mackenzie et al. , 2001). Fourth, the new leadership is more change-oriented. Transformational/charismatic leaders are likely to emerge from crisis environment and they serve as change agents in organizations.
Using data from 48 Fortune 500 firms, Waldman et al. (2001) illustrated a clear picture that charismatic leadership can predict performance under conditions with level of environmental uncertainty. Weaknesses and future directions of the approach Accordingly, the transformational/charismatic approach has some problems left for future research to resolve: First, the approach is criticized to regress back to the “one best way” of leadership with too much attention on leader behaviors (sometimes traits) but very little on contextual variables (Beyer, 1999).
Although transformational/charismatic leadership has been examined to be applicable in various contexts, further attention should be paid to identify situational factors that facilitate or limit the effects of transformational/charismatic leadership. On the other hand, the consequences of transformational/charismatic leadership do not always appear positive; rather, some researchers have 6 International Journal of Business and Management August, 2006 identified the “dark side” of the new leadership (e. g. , Conger and Kanungo, 1998).
For example, charismatic leadership tends to make more risky decisions that can result in serious failure. The future empirical studies should seek to identify more specific factors (i. e. , situations, leader qualities, task and follower characteristics) that induce the negative consequences. Second, the approach emphasizes the role the vision (articulated by the leader) plays in motivating followers. The underlying mechanism, however, is rather ambiguous (Conger, 1999). Simple identification and an attractive vision do not fully explain follower commitment and motivation.
Moreover, the structure and content of leadership vision are also only partially understood (Conger, 1999). To resolve these problems, leadership values and the value interexchange between the leader and the follower should be incorporated in the models (Shamir et al. , 1993). Several recently emerged leadership theories (e. g. , Servant leadership, Graham, 1991; Ethical leadership, Ciulla, 1998; Value-based leadership, Fu et al. , 2002), though at the early stage of development and receiving little empirical support, provide solid rationale for remedying the deficiencies through the value viewpoint.
For example, ethical leadership regards that values play a central role in the leadership process; servant leadership (Graham, 1991) argues that leadership values provide a system of rules or principles that guide the pondering, formulating, and communicating of the vision. Future studies should try to develop pragmatic models and empirically examine the effects of leadership values on leader behavior, leader-member relationship, influence process, follower behavioral and attitudinal outcomes, and leadership succession.
Third, the approach has to extend its level of analysis for future development of the theories. As discussed above, transformational/charismatic leadership theories have specified the dynamic of influence process, but only at the individual or dyadic level. Although the specification helps to explain individual behavioral or attitudinal outcomes, it does not adequately address group and organizational processes (Yukl, 2002). The analysis also results in a less consideration of contextual factors involving the leadership process and the suggestion of a “universal” form of leadership.
Future studies should seek to explain leadership’s effects on the interacting group including collective efficiency, group identification, etc. , and influence on the firm including culture, organizational change and so on. Finally, since the theories have addressed much about the role leaders play in organizational change, future models should include more change-related variables. For example, Yukl (2002) identifies a new behavioral dimension, change-oriented, to enrich the two factor model.
Future studies should specify the real impact of the leader behavior on followers at the dyadic level, units at the group level, as well as companies at the organization level. Leader-Member Exchange (LMX) theory Being different from other leadership theories in which a leader treats followers in a collective way (Average Leadership Style, ALS), LMX theory takes a quite different approach and makes the dyadic relationship between the leader and each follower the focal point of the leadership process.
Overall, the development of LMX theory has three stages: In the early studies of LMX, a leader’s relationship to the overall work unit comprises a series of vertical dyads categorized as being of in-group (high LMX relationship) and out-group (low LMX relationship). Subordinates who get well along with the leader and are willing to expand their role responsibilities become in-group members, whereas those who maintain only formal hierarchical relationships with their leader become out-group members.
Subsequent studies of LMX concerned with the relationship between LMX and organizational outcomes. Researchers found that high-quality exchanges between leaders and followers produced multiple positive outcomes (e. g. , less employee turnover, greater organizational commitment, and more OCBs). In general, 7 International Journal of Business and Management August, 2006 researchers determined that good LMX resulted in followers feeling better, accomplishing more, and helping the organization prosper (Graen, Liden, and Hoel, 1982).
The most recent emphasis in LMX research has been on leadership making, which emphasizes that leaders should try to transform the relationship into mutual trust, respect, and obligation to each other. Leadership making develops over time and involves different phases: a stranger phase, acquaintance phase, and partner phase, with the last phase generating the highest LMX quality. By taking on and fulfilling new role responsibilities, followers move through theses three phases to develop mature partnerships with their leaders.
According to Graen and Uhl-Bien (1991), the third phase corresponds to transformational leadership, whereas the first phase corresponds to transactional leadership. Weaknesses and future directions of the approach In correspondence with the above three major stages of LMX theory, the following weaknesses are likely to exist in the literature and future directions emerge for addressing the inadequacies: First, by suggesting that some members of the work unit receive special attention and the others do not, LMX theory provide negative implications to organizational justice (McClane, 1991; Scandura, 1999).
The perceived inequalities between in-group and out-group can have a devastating impact on the feelings, attitudes, and behavior of out-group members. For example, McClane (1991) pointed out the existence of in-group and out-group has undesirable effects on the group as a whole no matter whether the leader actually treats the members fairly or not. Scandura (1999) provided a comprehensive framework for future empirical examining the nature of the relationship through integrating organizational justice and LMX.
Second, given a plethora of empirical studies on its antecedents and consequences, LMX is regarded as a universal theory with little concern for situational variables that may affect the exchange process (Green et al, 1996). An exception is Dunegan, Duchon, and Uhl-Bien’s (1992) work on the role of task analyzability and task variety as moderating the effects of LMX on subordinate performance. They found that the relationship between LMX and performance was significant when tasks have low analyzability/high variety and high analyzability/low variety, whereas the relationship became insignificant in the “low-low” and “high-high” situations.
Such kinds of effort are still desired in LMX research in order to recognize the possible contingency. Third, LMX theory emphasizes the importance of leader-member exchanges, but fails to explain how high-quality relationship is established. Although the recent models highlighted the importance of role making (Graen and Uhl-Bien, 1991), mentoring (Scandura and Schriesheim, 1994), incremental influence, and type of reciprocity, the theory is still criticized for not explaining how the evolution of relationship actually occurred and how mature partnerships are built.
On the other hand, the vertical relationships are likely to horizontally interact with each other. Future work needs to elaborate more about how the differentiated relationships and the interaction among LMXs evolve over time. Finally, there are questions regarding research methodologies, for instance, questions whether the principal measure of LMX theory is sufficiently refined to measure the complexities of the relationship (Garen and Uhl-Bien, 1995), and whether appropriate level of analysis and data-analytic techniques are specified and employed (Schriesheim, et al. 2002). Overall, more rigorous operationalization of the model is called for improving its validity. Conclusion Leadership is a complex process in which the leader, followers, and the situation interact with each other. Theorists have attempted to discover what traits, abilities, behaviors, relationships between leader and 8 International Journal of Business and Management August, 2006 follower, and/or aspects of the situation determine how well a leader is able to influence followers and accomplish collective goals.
Based on my reviews, I have summarized the characteristics of the major theories in each of the four major paradigms and briefly discussed major findings, key research issues, strengths, weaknesses and possible future directions in regard to those theories. The two contemporary theories, namely transformational/charismatic leadership and LMX, are discussed in detail. Summarily, I would like to emphasize the following issues that should receive careful concerns in future studies: Clearly specifying the key components of theories.
For transformational/charismatic leadership, theorists should clarify the nature of charisma, identify specific factors that induce possible negative consequences, set clear boundary for its behavioral dimensions, and the vision’s motivational influence on followers. LMX theory should minimize the inconsistence of defining leader-member relationship, offer clearly defined and reliable constructs, and clarify the exchange process. Extending theories by incorporating contextual factors. It has concluded that leadership is a complex process and there is no one best way to lead.
Thus future studies should seek to adopt more contextual moderating variables and establish contingent model for providing specific implications for the particular situation. Effects of situational factors such as task, follower, work environment, and organizational variables should be systematically investigated in the leadership process. Clarifying the level of analysis. A good theory should include a clear and proper treatment of its level of analysis, which requires a valid and reliable measurement of the construct and choice of analytical methods.
Leadership is a process involving multiple levels of factors and its effects occur at the all levels: individual, dyadic, group, and organization. Analyzing the phenomenon in a framework with clear levels and specification within- and cross-level relationships will definitely help us understand the influence process and the dynamics of leadership. The contemporary theories, however, appear to be rather obtuse to this issue. For example, transformational/charismatic leadership has not explicitly formulated its enquiring level with some appearing to be strictly at individual level (e. . , individualized consideration; Bass, 1985) while others being at higher levels (e. g. , leader behaviors that foster the acceptance of group goals; Podsakoff et al. , 1990). On the other hand, LMX research has been criticized to be fundamentally uninformative about the LMX process because its analysis is totally confused in terms of levels (Schriesheim, et al. , 2002). Future research on leadership should pay special attention to the issue. References Bass, B. M. and B. Avolio, (1993), “Transformational leadership: a response to critiques,” In M.
M. Chemers and R. Ayman (eds. ), Leadership Theory and Research: Perspectives and Directions (pp. 49-80), New York: Academic Press. Bass, B. M. , (1985), Leadership and performance beyond expectations. New York: Free Press. Bass, B. M. , (1996), A new paradigm of leadership: An inquiry into transformational leadership. Alexandria, VA: U. S. Army Research Institute for the Behavioral and Social Sciences. Beyer, J. M. , (1999), “Taming and promoting charisma to change organizations,” Leadership Quarterly, 10(2), 307-330. Blake, R. R. and J. S.
Mouton, (1964), The Managerial Grid. Houston: Golf Publishing. Bryman, A. , (1992), Charisma and Leadership in Organizations. London: SAGE. Burns, J. M. , (1978), Leadership. New York: Harper & Row. Ciulla, J. B. , (1998), Ethics, the Heart of Leadership. Westport, CT: Greenwood. 9 International Journal of Business and Management August, 2006 Conger, J. A. , (1999), “Charismatic and transformational leadership in organizations: An insider’s perspective on these developing streams on research,” Leadership Quarterly, 10(2), 145-179. Conger, J. A. and R. N. Kanungo, (1998), Charismatic Leadership in Organizations. Thousand Oaks, CA: SAGE. Day, D. V. , (2000), “Leadership development: A review in context,” Leadership Quarterly, 11(4), 581-613. Dunegan, K. J. , D. Duchon, and M. Uhl-Bien, (1992), “Examining the link between leader-member exchange and subordinate performance: The role of task analyzability and variety as moderators,” Journal of Management, 18(1), 59-79. Fiedler, F. E. , (1967), A theory of leadership effectiveness. New York: McGraw-Hill. Fu, P. P. , R. J. House, J. Liu, et al. (2002), “Study of Chinese business Executives: A value-based approach,” Paper presented at the 62nd annual meeting of the Academy of Management, Denver, Colorado, August 9-14, 2002. Graen, G. , R. C. Liden, and W. Hoel, (1982), “The role of leadership in the employee withdrawal process,” Journal of Applied Psychology, 67(6), 868–872. Graen, G. B. and M. Ulh-Bien, (1991), “The transformation of professionals into self-managing and partially self-designing contributions: Toward a theory of leader-making,” Journal of Management Systems, 3(3), 33-48. Graen, G. B. , and M.
Ulh-Bien, (1995), “Relationship-based approach to leadership: Development of leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain perspective,” Leadership Quarterly, 6(2), 219-247. Graham, J. W. , (1991), “Servant-leadership in organizations: Inspirational and moral,” Leadership Quarterly, 2(2), 105-119. Green, S. G. , S. E. Anderson, and S. L. Shivers, (1996), “Demographic and organizational influences on leader-member exchange and related work attitudes,” Organizational Behavior and Human Decision Processes, 66(2), 203-214. Hersey, P. , and K.
Blanchard, (1982), Management of organizational behavior: Utilizing human resources. Englewood Cliffs, NJ: Prentice-Hall. House, R. J. , and B. Shamir, (1993), “Toward the integration of transformational, charismatic and visionary theories of leadership,” In M. C. R. Ayman (eds. ) Leadership Theory and Research: Perspectives and directions. New York: Academic Press. House, R. J. , and R. R. Mitchell, (1974), “Path-goal theory of leadership,” Journal of Contemporary Business, 3, 81-97. Hughes, R. L. , R. C. Ginnett, and G. J. Curphy, (1996), Leadership: Enhancing the Lessons of Experience.
Chicago: Irwin. Jermier, J. M. , and S. Kerr, (1997), “Substitutes for leadership: their meaning and measurement contextual recollections and current observations,” Leader