managment

Transactors, transformers and beyond

A multi-method development of a theoretical typology of leadership

Craig L. Pearce Peter F. Drucker Graduate School of Management, Claremont Graduate

University, Claremont, California, USA Henry P. Sims Jr

R. H. Smith School of Business, University of Maryland, College Park, Maryland, USA

Jonathan F. Cox Center for the Study of Work Teams, University of North Texas, Denton,

Texas, USA Gail Ball

Department of Distance Education, The Pennsylvania State University, University Park, Pennsylvania, USA

Eugene Schnell Johns Hopkins University, Baltimore, Maryland, USA

Ken A. Smith School of Business, Syracuse University, Syracuse, New York, USA

Linda Trevino Frank and Mary Jean Smeal School of Business, Pennsylvania State

University, University Park, Pennsylvania, USA

Keywords Leadership, Directives, Empowerment

Abstract Extends the transactional-transformational model of leadership by deductively developing four theoretical behavioral types of leadership based on a historical analysis of leadership literature. Then, in an exploratory empirical phase, uses two data sets to inductively develop alternative models of leadership types. Finally, with a third data set, tests several theoretically plausible typologies using second-order confirmatory factor analysis (CFA). The results of the CFA generally support the existence of four leadership types: directive leadership, transactional leadership, transformational leadership, and empowering leadership.

The Emerald Research Register for this journal is available at The current issue and full text archive of this journal is available at

http://www .emeraldinsight .com/researchregister http:// www.emeraldinsigh t.com/0262-171 1.htm

The authors would like to thank James McGregor Burns, Robert J. House, James (Jerry) Hunt and Ken G. Smith for their helpful comments on earlier drafts of this manuscript.Craig L. Pearce and Henry P. Sims Jr participated in studies one, two and three; Eugene Schnell and Ken A. Smith participated in study one; Gail Ball and Linda Tevino participated in study two; Jonathan F. Cox participated in study three.

Transactors, transformers and

beyond

273

Received October 2002 Accepted October 2002

Journal of Management Development Vol. 22 No. 4, 2003

pp. 273-307 q MCB UP Limited

0262-1711 DOI 10.1108/02621710310467587

The purpose of this research is to investigate and build a typology of leadership that has foundations in, and extends, the current dominant transactional-transformational paradigm (Sashkin and Rosenbach, 1993; Yukl, 1989). Typologies are theories that identify multiple interrelated “ideal types” at different levels of abstraction. According to Doty and Glick (1994) typologies are essential to clear and parsimonious understanding of organizational phenomena. As Phillips and Lord (1986) and Hunt (1996) suggest, it is critical to our understanding of leadership that we clearly delineate the behavioral level accuracy of the typologies we use.

The current dominant typology in leadership research is the transactional- transformational typology. However, one may wonder if scholars have coalesced too narrowly on this two factor theory of leadership. Yukl (1989, p. 212) stated:

. . . [the transactional-transformational paradigm] is fast becoming a two factor theory of leadership processes, which is an unwarranted oversimplification of a complex phenomenon.

Further, Yukl (1989, p. 212) declared:

. . . the distinction between the two types of leadership is not as clear as some theorists would have us believe . . . these differences should be determined by empirical research, not predetermined theoretical definitions that make unnecessary assumptions and bias subsequent research.

Further, Bass and Avolio (1993, p. 76) stated:

. . . we invite critics and supporters to join in the effort to shape a theory and model of leadership that captures a broader array of leadership behaviors and attributes than previously studied.

In this research we attempt to expand the range of leadership examined by using historical analysis to extend the transactional-transformational paradigm.

In this paper we take a theory building approach (Doty and Glick, 1994; Dubin, 1978). Yukl (1998) defined three levels of abstraction in leader behavior research:

(1) narrow (specific items);

(2) mid-level (behavioral scales); and

(3) broad level (groupings of behavioral scales).

Our focus cuts across these three levels but is ultimately aimed at the delineation of a broad level typology of the grouping of leader behaviors for use in future research. Yukl (1998, p. 63) also asserted that:

. . . it is essential to pay more attention to the overall pattern of leadership behavior rather than becoming too preoccupied with any particular component of it.

Thus, in phase I, we first integrate major historical findings from the field of leadership to deductively develop a broad level theoretical model of leader

JMD 22,4

274

behavior consisting of four behavioral types. These behavioral types are groupings of specific leader behaviors, and can alternatively be thought of as behavioral sets, behavioral strategies, behavioral styles, archetypes, or second- order factors. In phase II, we empirically examine two independent data sets, using exploratory factor analytic (EFA) techniques, to inductively generate alternative theoretically plausible models of the broad level grouping of leader behavior. Subsequently, in phase III, we test each of the competing models’ ability to account for the underlying structure of the data, with a third independent data set, using confirmatory factor analytic (CFA) techniques. This analysis leads to a final model of the broad level types of leader behavior. As Bryman (1992, p. 113) notes “There is little doubt that there is the possibility of conceptual confusion becoming rife in this field”. Thus, this research is intended to help establish some clarity regarding the current status of the field and to help guide future research on leader behavior and its impact in organizations.

Phase I. Historical analysis Our historical analysis of the leadership literature essentially yielded an adapted and extended version of the model of leadership first proposed by Bass et al. (1987). The model is also adapted from the theoretical development of Manz and Sims (1991) and also builds on the work of Yukl (1987) and Quinn (1988)[1]. The fundamental theoretical and research bases of this model are derived from a historical analysis of the leadership literature. For each type, we draw from several theoretical roots of specific leader behaviors as well theoretical traditions related to leader behavior. These roots are summarized in Table I and are described in greater detail below.

Directive leadership The first behavioral type from our historical analysis of leadership is “directive leadership”. The directive leadership behavioral type describes leadership that primarily relies on position power, which at times has been referred to as legitimate power (French and Raven, 1959). Directive leadership also relies on coercive power (French and Raven, 1959). This type of leadership includes direction, command, assigned goals, intimidation and reprimand as the primary mechanisms to influence subordinate behavior.

The roots of the directive leadership behavioral type lie in theory X management style (McGregor, 1960), initiating structure types of leader behavior from the Ohio State studies (Fleishman, 1953), task oriented types of leader behavior from the Michigan studies (Katz et al., 1950), and punishment research (Arvey and Ivancevitch, 1980; Korukonda and Hunt, 1989; Sims, 1980) and is similar to what Quinn et al. (1990) refer to as a consolidation orientation. Each theoretical tradition is discussed, in turn, below.

Transactors, transformers and

beyond

275

Theory X. McGregor (1960) defined a two factor theory of leadership; theory X- theory Y. The underlying tenets of the theory X leader are that the leader believes that the average adult:

. is lazy – and works as little as possible;

. lacks ambition;

. dislikes responsibility;

. prefers to be led by others;

. is inherently selfish;

. is indifferent to organizational needs and goals;

. is resistant to change; and

. is gullible, not very intelligent, and easily duped.

Holding these beliefs about individuals the theory X manager resorts to policies and practices that contain tight control mechanisms and exert coercive power over subordinates (Pinder, 1984). These types of stringent control are the first basis of the directive leadership behavioral type.

Initiating structure. The initiating structure grouping of leader behaviors was identified in the Ohio State studies (Fleishman, 1953; Halpin and Winer,

Leadership type Theoretical and research bases

Directive leadership Theory X leadership (McGregor, 1960) Initiating structure from Ohio State studies

(e.g. Fleishman, 1953) Task oriented behavior from Michigan studies

(e.g. Katz et al., 1950) Punishment research (e.g. Arvey and Ivancevitch, 1980)

Transactional leadership Expectancy theory (e.g. Vroom, 1964) Path-goal theory (e.g. House, 1971) Equity theory (e.g. Adams, 1963) Exchange theory (e.g. Homans, 1961) Reinforcement theory (e.g. Luthans and Kreitner, 1985;

Sims, 1977; Thorndike, 1911) Reward research (Podsakoff et al., 1982)

Transformational leadership Sociology of charisma (e.g. Weber, 1946) Charismatic leadership theory (e.g. House, 1977) Transformational leadership (e.g. Bass, 1985; Burns, 1978)

Empowering leadership Behavioral self-management (e.g. Thorenson and Mahoney, 1974)

Social cognitive theory (e.g. Bandura, 1986) Cognitive behavior modification (e.g. Meichenbaum, 1977) Participative management and participative goal setting

research (e.g. Likert, 1961, 1967; Locke and Latham, 1990)

Table I. Theoretical and research bases of the historically- derived model of leadership types

JMD 22,4

276

1957). The Ohio State studies identified two broad types of leader behavior through the use of factor analysis (Yukl, 1989). The initiating structure group of leader behavior includes two general types of leader behavior:

(1) directive structuring behavior; and

(2) autocratic and punitive behavior (Schriesheim et al., 1976).

Bass (1990) and Yukl (1989) reviewed the initiating structure literature and defined the following directive/structuring and autocratic/punitive leader behaviors. Directive leader behaviors included:

. initiates activity of the work group;

. organizes group activity;

. defines, in detail, the way work is done;

. establishes clear channels of communication;

. emphasizes goal attainment;

. assigns subordinates to tasks;

. offers new approaches to problems; and

. coordinates activities of subordinates.

Also, the autocratic/punitive leader behaviors contained in some of the initiating structure studies included:

. making important decisions without consulting subordinates;

. criticizing poor work;

. ensuring that subordinates follow procedures; and

. ensuring that subordinates work up to their capacity.

Thus, initiating structure is the second basis of directive leadership. Task oriented behavior. Parallel to the Ohio State studies were the Michigan

leadership studies. These two groups of studies arrived at quite similar conclusions (Bass, 1990; Yukl, 1989). Likert (1961, 1967) summarized much of the Michigan studies and found three primary groupings of managerial behavior:

(1) task oriented behavior;

(2) relationship oriented behavior; and

(3) participative leadership.

In the Michigan studies task oriented behavior included such behaviors as: . planning and scheduling work; . coordinating subordinate work; and . providing supplies, tools and technical assistance.

Transactors, transformers and

beyond

277

As in the case of the initiating structure leader behaviors, the task-oriented leader behaviors also form a backdrop for the directive leadership behavioral type.

Punishment research. There are, broadly speaking, two types of punishment, contingent and non-contingent. In this sense, Kazdin (1975, pp. 33-4) defines contingent punishment as:

. . . the presentation of an aversive event or the removal of a positive event following a response which decreases the frequency of that response.

Non-contingent punishment, on the other hand, is the presentation of an aversive event in some random manner. In an organizational setting there are many types of punishments that are routinely dispersed, ranging from an angry word to dismissal. Researchers have also investigated the effects of non-contingent reprimand and punishment behavior on subordinate satisfaction and performance (e.g. Arvey and Ivancevitch, 1980; Ball et al., 1994; Korukonda and Hunt, 1989; Pearce and Sims, 2002; Podsakoff et al., 1982; Sims, 1980). Non- contingent reprimand tends to yield a negative effect on subordinate satisfaction but little effect on performance, and contingent punishment has little effect on subordinate performance (Cox, 1994). Both contingent and non-contingent reprimand behaviors form part of the directive leadership behavioral type.

Summary. Based on our historical analysis of the theoretical and research bases of leadership it seems likely that directive leadership will prove to be a stable and distinct broad level type of leader behavior and will thus prove to be an extension of the transactional-transformational paradigm. Thus:

H1a. Directive leadership exists as a distinct behavioral type and includes such behaviors as: instruction and command; assigned goals; contingent reprimand, and intimidation and non-contingent reprimand.

Transactional leadership The second behavioral type from our historical analysis of leadership is transactional leadership. The transactional leadership behavioral type is generally consistent with the components of the transactional-transformational paradigm of leadership. The bases of this type lie in expectancy theory (Vroom, 1964), path-goal theory (House, 1971; House and Mitchell, 1974), exchange/equity theory (Adams, 1963; Homans, 1958, 1961), and reinforcement theory (Luthans and Kreitner, 1985; Scott and Podsakoff, 1982) and is somewhat similar to what Quinn et al. (1990) termed output maximization orientation. Each theoretical tradition is reviewed, in turn, below.

Expectancy theory. According to expectancy theory, a cognitive-rational model of human behavior, individuals assess situations according to three variables:

(1) valence – the attractiveness of potential outcomes for engaging in certain behaviors;

(2) instrumentality – the perceived linkage between a behavior and the outcome; and

JMD 22,4

278

(3) expectancy – the perceived likelihood of effort resulting in the behavior necessary to obtain the outcome.

Subsequently, individuals engage in behaviors that will maximize their expected return from performance. In line with equity theory, transactional leadership is focused on clarifying the effort-reward relationships and using the reward systems to achieve maximal motivation. Thus, rewards are part of the “exchange” equation.

Path-goal theory. Building on the concepts of expectancy theory (Vroom, 1964), House (1971) House and Dessler (1974) and House and Mitchell (1974) developed path-goal theory. Path-goal theory was developed to explain how various leader behaviors (contained in consideration-initiating structure research) influence subordinate satisfaction and performance by clarifying the “path” to desired rewards.

According to House and Dessler (1974, p. 13):

. . . leader behavior will be viewed as acceptable to subordinates to the extent that the subordinates see such behavior as either an immediate source of satisfaction or as instrumental to future satisfaction.

This model of leadership espouses the view that the leader needs to modify the way subordinates perceive the contingency relationships between effort and subsequent satisfaction and that the means for doing this is dependent on the subordinate task environment. Thus, the path-goal theory of leadership emphasizes the transactional nature of leadership.

Exchange/equity theory. Homans (1961), and Adams (1963) are generally credited with the development of the exchange or equity group of theories. While several versions exist, the basic tenet of this class of theories is that individuals seek to maintain equity between what they give vis-à-vis what they obtain in an exchange (Landy, 1985; Pinder, 1984). Like expectancy theory, this group of theories is cognitive-rational. Individuals in the exchange relationship are thought to assess four fundamental pieces of information:

(1) what they give in an exchange;

(2) what they receive in an exchange;

(3) what others give in a similar exchange; and

(4) what others receive in a similar exchange.

Based on these four pieces of information, individuals are thought to strive to maintain an equitable ratio between what they give and receive from an exchange in comparison to what others give and receive from a comparable exchange. The theory predicts that individuals will attempt to rectify situations where these exchange ratios are out of balance. Research has demonstrated that individuals are more likely to engage in corrective actions when they perceive negative inequity than when they perceive positive inequity (Landy, 1985; Pinder, 1984).

Transactors, transformers and

beyond

279

Based on this theory of motivation, prescriptions are made for leadership. The prescriptions center on motivating subordinate performance through providing equitable rewards for inputs and thus predicts that higher levels of subordinate input can be generated through higher levels of reward. Therefore, this class of motivation theories serves as another basis for the transactional leadership behavioral type.

Reinforcement theory. Reinforcement theory is summarized by the law of effect (Thorndike, 1911), which suggests that the consequence of a behavior is an important determinant of whether the behavior will be repeated. Using the law of effect transactional leader behavior influences subordinate behavior by reinforcing (rewarding) those behaviors that are desired (Luthans and Kreitner, 1985; Sims, 1977).

Summary. Based on our historical analysis of the theoretical and research bases of leadership it seems likely that transactional leadership will prove to be a stable and distinct type of leader behavior and will thus prove to reinforce the transactional-transformational paradigm. Thus:

H1b. Transactional leadership exists as a distinct behavioral type of leadership and includes such behaviors as: contingent material reward and contingent personal reward.

Transformational leadership The third behavioral type from our historical analysis of leadership is “transformational leadership”. This behavioral type is similar to the transactional-transformational paradigm, but as Bryman (1992) notes there is some conceptual confusion as to what is contained in transformational leadership. The historical bases of the transformational leadership behavioral type are drawn from the sociology of charisma (Weber, 1924/1947, 1946), charismatic leadership theory (House, 1977), and transformational leadership (Bass, 1985; Burns, 1978) and is closely related to what Quinn et al. (1990) term change orientation.

Sociology of charisma. Weber (1924/1947, 1946) introduced the concept of charisma into the study of leadership. However, much of this work laid dormant and was only studied in the fields of sociology rather than leadership per se, until the late 1970s (Bass, 1990). Weber’s (1924/1947, 1946) original concept of charismatic leadership was focused on the leaders who inspired new forms of organization.

Trice and Beyer (1986) summarized Weber’s (1924/1947, 1946) work as containing five components necessary for the emergence of a charismatic leader. These five components were:

(1) the leader must have extraordinary (mystical) gifts;

(2) there must be some type of crisis;

(3) the leader must espouse a radical solution to the crisis;

JMD 22,4

280

(4) followers must believe they are linked to the leader through some type of transcendent powers; and

(5) followers validate the transcendence of the leader’s gifts through repeated successes.

Charismatic leadership theory. House (1977) proposed “A 1976 theory of charismatic leadership” to address a long-standing gap in the formal study of leadership. This theory catalyzed the beginning of a new genre in leadership studies. The basic tenets of the original House model are defined in terms of leader traits, leader effects on followers, and leader behaviors. House’s (1977) viewpoints continued to develop, and later revisions of the theory (House and Shamir, 1993; House et al., 1993) proposed that charismatic leaders engage in the following six behaviors in order to achieve charismatic effects:

(1) impression management (behaviors designed to display leader competence);

(2) articulation of ideological goals;

(3) defining subordinate roles in terms of ideological values;

(4) engaging in role modeling behavior;

(5) communicating high expectations and confidence in subordinates; and

(6) engaging in behavior designed to arouse appropriate follower motives (e.g., need for achievement).

Charismatic leadership theory provides a basis for the transformational leadership behavioral type.

Transformational leadership. Burns (1978) more clearly explicated the distinction between transactional and transformational leadership behaviors. He depicted transactional leaders as those who focus on reward contingencies and “approach followers with an eye to exchanging one thing for another” (Burns, 1978, p. 3). He contrasted the transactional leader with the transformational leader. He depicted transformational leaders as those who focus on transforming followers’ motivational state to higher level needs such as self-actualization (Maslow, 1954).

Bass (1985, 1990), Avolio and Bass (1988), Avolio et al. (1999), Bass et al. (1987), Hatter and Bass (1988), Waldman et al. (1987), and Yammarino and Bass (1990) operationalized and empirically tested Burns’ (1978) leadership concepts. Bass (1985, 1990) also extended the Burns’ (1978) model by including leaders who do not necessarily appeal to only higher moral values. The leader behaviors included in Bass’ model of transformational leadership include:

. transmitting a sense of mission;

. delegation of authority, coaching and teaching; and

. emphasizing problem solving and use of reasoning.

Transactors, transformers and

beyond

281

Summary. Based on our historical analysis of the theoretical and research bases of leadership it seems likely that transformational leadership will prove to be a stable and distinct type of leader behavior and will thus prove to reinforce the transactional-transformational paradigm. Thus:

H1c. Transformational leadership exists as a distinct behavioral type and includes such behaviors as: stimulation and inspiration; vision; idealism; and challenge to status quo.

Empowering leadership The fourth behavioral type from our historical analysis is empowering leadership. Empowering leadership emphasizes the development of follower self-management or self-leadership skills. Manz and Sims (1990, 1991, 2001) have called this “superleadership”. The historical bases of the empowering leadership are found in behavioral self-management (Thorenson and Mahoney, 1974), social cognitive theory (Bandura, 1986), cognitive behavior modification research (Meichenbaum, 1977) and participative goal setting research (Erez and Arad, 1986) and is closely related to what Quinn et al. (1990) term a human resource development orientation.

Behavioral self-management. Behavioral self-management has its roots in clinical psychology (Mahoney and Arnkoff, 1978). Manz and Sims (1980) expanded the clinical applications to organizational environments by defining self-management as a substitute for leadership. Behavioral self-management includes several self-controlled behavior management strategies (Manz and Sims, 1990) including:

. self-observation;

. environmental cueing;

. self-goal setting;

. self-reward;

. self-punishment; and

. rehearsal.

Encouraging subordinate behavioral self-management is the first basis of the empowering leadership behavioral type.

Social cognitive theory. The basic proposition of social cognitive theory is that of triadic reciprocality (Bandura, 1986). According to triadic reciprocality individuals influence their environment through their behavior both of which (environment and behavior) in turn influence the individual. One of the key contributions of social cognitive theory is a framework for understanding how modeling influences individual behavior. In terms of the empowering leadership behavioral type, it is proposed that the leader models appropriate self-leadership behavior which is subsequently adopted by the subordinate.

JMD 22,4

282

Cognitive behavior modification. Similar to social cognitive theory (Bandura, 1986) cognitive behavior modification research has focused on “conceptualiz(ing) cognitive events and . . . understand(ing) their role in behavior change” (Meichenbaum, 1977, p. 11). Cognitive behavior modification is similar to the cognitive self-leadership strategies of “superleadership” such as the strategy of re-conceptualizing performance obstacles not as problems, but rather as opportunities for learning (Cox, 1994; Manz and Sims, 1990), and thus provides another basis of the empowering leadership behavioral type.

Participative goal setting research[2]. Locke and Latham (1990) reviewed 25 years of goal setting research. The major findings indicate that specific, difficult goals lead to higher performance, and that, in general, it does not matter if the goals are participatively or unilaterally set. Erez and Arad (1986) have found some instances where participatively set goals can lead to higher performance and satisfaction. However, while the directive leadership behavioral type includes assigning goals, the empowering leadership behavioral type emphasis on developing subordinate self-management skills is more in keeping with participative goal setting as in the case of an ideal “management by objectives” system (Drucker, 1954).

Summary. Based on our historical analysis of the theoretical and research bases of leadership it seems likely that empowering leadership will prove to be a stable and distinct type of leader behavior and will thus prove to be an extension of the transactional-transformational paradigm. Thus:

H1d. Empowering leadership exists as a distinct behavioral type and includes such behaviors as: encourages opportunity thinking; encourages self-reward; encourages self-leadership; encourages participative goal setting; and encourages teamwork.

Historical analysis summary Bass and Avolio (1993) state that one of the major problems in the study of leadership is that there is a tendency, in leadership research, to discount previously existing theories in an effort to introduce a “new way of thinking”. We agree. In this section we attempted to build on multiple theoretical and empirical lines from the leadership literature, and we also attempted to build on, and extend, the current leadership paradigm rather than discount and discard it in the development of our deductively derived model of the behavioral types of leadership.

From our historical analysis we generated four hypotheses regarding behavioral types of leadership:

(1) directive leadership;

(2) transactional leadership;

(3) transformational leadership; and

(4) empowering leadership.

Transactors, transformers and

beyond

283

Together these four hypotheses account for our first theoretically plausible typology (Model One) that we test later with CFA techniques. In the next section we inductively derive additional theoretically plausible typologies using EFA techniques.

Phase II. Preliminary empirical examination Method In this section we empirically explore the behavioral types of leadership, using EFA techniques, with varimax rotation, with data from two independent research projects. Both of these projects, as well as previous empirical research (e.g. Podsakoff et al., 1982; Bass et al., 1987; Manz and Sims, 1987), relied on our deductively-derived typology as inspiration to generate questionnaire items to describe leader behavior. The specific questionnaire items from each study differ slightly due to improvements made to the instrument over time and because of differing emphasis on leadership in the projects.

Both of the studies conducted first order factor analyses of leader behaviors with data collected from different organizational environments. The first order factor analysis of these data sets can be obtained from Scully et al. (1994) and Ball (1991), respectively. The results of the first order factor analyses are the input for our empirically-based inductive generation of leader behavior types.

We then used higher order factor analysis where we conduct factor analyses of the dimensions produced from the first order factor analyses. The higher order factor analyses were generated by evaluating factor solutions (from each data set) over a range of a specified number of factors. The scree points of the factor solutions were evaluated as input, but were not blindly utilized in selecting the final solutions. As is common practice with EFA (see Gorsuch, 1983) the final solution in each set was made on the basis of examining iterative factor solutions until the most sensible solution became apparent. Rummel (1970, p. 357) referred to this procedure as using a “research sense” criterion of selection.

Thus, with both data sets, we used an iterative exploratory approach following the procedures laid out by Gorsuch (1983). That is, we examined several factor solutions with a range of factor numbers, and eventually a slightly reduced set of variables. For example, laissez-faire was eventually dropped from data set one, and “non-contingent reward” from data set two, because they did not load with any other leader behavior dimensions. The results of higher order factor analysis allow for wider scope of generalization of results (Gorsuch, 1983) and the identification of what Yukl (1998) termed a broad level of abstraction.

Overview of the data sets Data set one. This study involved the description, by the top management team members (n ¼ 253), of the leader behavior of the CEO. The

JMD 22,4

284

organizations included in this study came from high-tech firms located in the mid-Atlantic USA region. Of the firms, 63 percent were publicly held. The final sample consisted of individual members of 56 top management teams. The age of the subjects ranged from 27 to 60 years, and 87.5 percent of the sample was male.

The first order factor analysis of this data resulted in 15 distinct leader behaviors. All of these dimensions demonstrated excellent reliability. Table II lists the behaviors from both data set one and data set two, as well as the behaviors from data set three (which we later use in the model testing portion of this research).

Leadership type Study one Study two Study three

Directive Instruction Instruction and command

Instruction and command

Command Assigned goals Assigned goals Assigned goals

Contingent reprimand Contingent reprimand Intimidation Intimidation and

non-contingent reprimand

Non-contingent reprimand

Non-contingent reprimand

Transactional Contingent material reward

Contingent material reward

Contingent material reward

Contingent personal reward

Contingent material reward

Contingent material reward

Stimulation and inspiration

Stimulation and inspiration

Stimulation and inspiration

Transformational Vision Vision Vision Idealism Challenge to status quo Encourages

opportunity thinking Encourages self-reward

Encourages self- efficacy

Encourages self- efficacy

Empowering Encourages initiative Encourages initiative Encourages self- leadership

Encourages self problem solving

Encourages self problem solving

Encourages self goal setting

Encourages self goal setting

Participative goal setting

Encourages self evaluation and criticism

Encourages teamwork

Table II. Summary of leader

behaviors examined across the three

studies

Transactors, transformers and

beyond

285

Data set two. This study (Ball, 1991) involved the description, by the subordinates (n ¼ 208), of leader behavior of supervisors. The description of leader behavior was conducted with a related version of the instrument used in data set one. Of the organizations, 62 percent were government agencies and 38 percent were private businesses representing a variety of industries. These organizations were primarily located in the northeastern and southeastern USA.

The supervisors in the sample ranged in age from 26 years to 61 years with a mean age of 42.3 years. Of the supervisors, 30 percent were female. The work experience of these supervisors ranged from two to 46 years with a mean of 27.6 years, and their managerial experience ranged from one to 25 years with a mean of 9.6 years.

Summary. These two data sets provide a diverse set of leaders from which to explore the underlying structure of leader behavior. Participants represent multiple industries and government agencies. Also, target leaders ranged from supervisors to CEOs, thus enabling a broader generalization of the results.

In summary, the main research approach, in this section, was to conduct higher order EFA, with varimax rotation, on data from two independent data sets. The scores on the first order leader behaviors were used as manifest indicators to the higher order factor analyses. The main objective, for the analysis of each data set, was to discover how the more specific leader behaviors grouped together, and thus to inductively derive broad level behavioral types of leadership based on the results across these two data sets.

EFA results The final factor structures from data sets one and two are presented in Tables III and IV, respectively. For ease of interpretation, the factors are presented in the order presented in the historically-derived model rather than the order in which the factors were extracted, however, the order of extraction is indicated by Roman numerals. Specific results from analysis of each data set are presented below.

Data set one. The factor analysis of the data from study one (see Table III) yielded four factors. Factor one contains three of the four leader behaviors from the historical analysis directive leadership behavioral type. However, one of the behaviors, contingent reprimand, also had a high loading on another factor. Therefore, it appears that the directive leadership may indeed be a distinct behavioral type of leadership.

Factor two contains one of the leader behaviors from the historical analysis directive leadership behavioral type (non-contingent reprimand), all of the leader behaviors of transactional leadership, and all of the leader behaviors of transformational leadership. However, “non-contingent reprimand” and “stimulation and inspiration” both had high cross- loadings on other factors. Therefore, it would appear that transactional leadership and transformational leadership may collapse into one

JMD 22,4

286

Extraction order Factor name III I IV II

Directive Instruction and command 0.79 20.13 20.21 20.03 Contingent reprimand 0.62 20.08 0.16 0.55 Assigned goals 0.74 0.25 0.01 0.08 Transactor plus Stimulation and inspiration 0.13 0.79 0.04 0.37 Vision 0.01 0.74 0.08 0.28 Contingent personal reward 20.04 0.69 0.10 0.33 Contingent material reward 0.35 0.66 0.18 0.01 Non-contingent reprimand 0.45 20.71 20.11 0.01 Empowering one Encourages self-efficacy 20.15 0.34 0.70 0.24 Encourages initiative 0.01 20.02 0.85 20.07 Empowering two Encourages self evaluation and criticism 0.36 0.26 0.00 0.76 Encourages self problem solving 20.09 0.29 0.49 0.59 Encourages self goal setting 20.07 0.40 20.01 0.74 Eigen value 1.26 4.50 0.89 2.29 Variance explained (%) 9.70 34.60 6.90 17.70 Cumulative variance explained (%) 9.70 44.30 51.20 58.90

Table III. Factor structure from study one

Extraction order Factor name IV I III II

Directive Command 0.84 20.28 20.08 0.08 Instruction 0.82 0.19 0.21 0.12 Assigned goals 0.59 0.18 0.35 0.39 Aversive Intimidation 20.05 20.93 0.03 20.11 Non-contingent reprimand 0.15 20.82 20.20 20.17 Contingent personal reward 0.15 0.55 0.53 0.37 Encourages self-efficacy 0.22 0.55 0.51 0.52 Transformational Idealism 0.05 0.00 0.91 0.17 Vision 0.37 0.47 0.56 0.34 Stimulation and inspiration 0.26 0.48 0.54 0.48 Empowering Encourages self problem solving 0.10 0.06 0.06 0.93 Encourages initiative 0.09 0.45 0.42 0.67 Encourages self goal setting 0.25 0.46 0.45 0.61 Contingent material reward 0.24 0.16 0.37 0.56 Eigen value 0.77 7.44 0.97 1.89 Variance explained (%) 5.50 53.10 6.90 13.50 Cumulative variance explained (%) 5.50 58.60 65.50 79.00

Table IV. Factor structure from study two

Transactors, transformers and

beyond

287

behavioral type. Perhaps Yukl (1989) was correct in declaring that these two types are potentially more of a statistical artifact of a restricted range of behavioral sampling.

Factors three and four account for all five leader behaviors from the historically-derived empowering leadership behavioral type. It is curious, however, that the empowering leadership behaviors are divided into two factors. Together these two factors suggest that a distinct empowering leadership behavioral type may exist.

Data set two. The factor analysis of data set two (see Table IV) yielded four factors. Factor one contains three of the five leader behaviors from the historically-derived directive leadership behavioral type. One of the behaviors in this factor also had a high cross-loading on another factor. However, two behaviors from the historically-derived directive leadership behavioral type did not load on this factor. On closer examination it is clear that the three behaviors that did load were more “directive” in nature, while the two behaviors that did not load were more “aversive” in nature. Thus, it seems plausible that this historically-derived behavioral type may actually split into two distinct types.

Factor two is comprised one of the two leader behaviors from the historically-derived transactional leadership behavioral type, one behavior from empowering leadership and two behaviors (negatively loaded) from directive leadership (those behaviors that are more “aversive” in nature). However, “contingent personal reward” and “encourages self-efficacy” both had high cross-loadings. The only two behaviors that clearly load on this factor were the two “aversive” leader behaviors from the historically-derived directive leadership behavioral type, thus lending further credence to the idea that this historically-derived type may indeed split into two distinct types.

Factor three is comprised all of the leader behaviors from the historically- derived transformational leadership behavioral type. While, “stimulation and inspiration” and “vision” also had high cross-loadings on another factor, these results clearly suggest the existence of a distinct transformational leadership type.

Factor four is comprised three of the four leader behaviors from the historically-derived empowering leadership behavioral type and one of the transactional leadership behaviors. However, “contingent material reward”, “encourages initiative”, and “encourages self-goal setting” all had high cross- loadings. The only behavior that clearly loads on this factor is one of the behaviors from the historically-derived empowering leadership type, thus further suggesting the potential existence of a distinct empowering leadership behavioral type.

Empirically-based, inductively-derived hypothesis generation. Yukl (1989) implored researchers to determine the broad level types of leadership with empirical analyses and not merely to rely on “predetermined theoretical definitions that make unnecessary assumptions and bias subsequent results” (Yukl, 1989, p. 212). Our historical analysis generated a theoretical model of

JMD 22,4

288

leadership that contained four behavioral types. However, in pursuing Yukl’s (1989) advice we see that our EFA results provide less than complete concurrence with our historical model.

At this stage, following are our overall conclusions. Results across data sets one and two suggest that a distinct behavioral type of directive leadership does exist. However, this type may be comprised only a sub-set of the leader behaviors identified in the historical analysis – those that are clearly “directive” in nature, while those that are more “aversive” in nature may form a second behavioral type. This is consistent with the findings of Schriesheim et al. (1976) who identified two sub-groups of initiating structure behavior. We refer to these two theoretically plausible behavioral types as “directive leadership” and “aversive leadership”, respectively.

The second and third historically-derived behavioral types were transactional leadership and transformational leadership. The results across data sets one and two suggest that transactional leadership and transformational leadership may not be as distinct as the historically-derived model indicates. Thus, these results suggest that Sashkin and Rosenbach (1992) may have been on track when asking if transformational leadership may merely entail a qualitatively different type of transaction and exchange than that associated with transactional leadership. In data set one these two types seemed to collapse into one, and we refer to this theoretically plausible behavioral type as “transactional-transformational leadership.” However, in data set two the transactional leadership type did not seem to be supported, while the transformational leadership type appeared to receive very good support. Therefore, these results suggest that transactional leadership and transformational leadership may potentially be accounted for with either one or two behavioral types.

The fourth historically-derived behavioral type was empowering leadership. The results across data sets one and two suggest that empowering leadership does indeed exist as a distinct behavioral type, and may represent a “paradigm shift” rather than a linear extension of transformational leadership. While the specific dimensions are somewhat mixed across the data sets, at least one empowering leadership factor emerged in each of the data sets. Data set one yielded two factors which accounted for all of the historically-derived empowering leadership behaviors. Data set two contained one empowering leadership factor that was comprised three of the four historically-derived leader behaviors, and if one discounts the high cross-loading behaviors, each of these factors contains only behaviors from the historically-derived behavioral type. These results clearly suggest that a distinct empowering leadership behavioral type does exist.

Therefore, the results across the exploratory factor analyses of data sets one and two suggest several competing typologies, that diverge from the historically-derived typology, that could account for the underlying structure of

Transactors, transformers and

beyond

289

leader behavior. Thus, a second theoretically plausible model suggests that the directive leadership type could be split into two components while the remainder of the historically-derived model remains intact, and therefore leads to the following hypothesis, which we call Model Two:

H2a. Directive leadership exists as a distinct behavioral type (excluding aversive leadership).

H2b. Aversive leadership exists as a distinct behavioral type.

H2c. Transactional leadership exists as a distinct behavioral type.

H2d. Transformational leadership exists as a distinct behavioral type.

H2e. Empowering leadership exists as a distinct behavioral type.

However, a third theoretically plausible model suggests that transactional leadership and transformational leadership may collapse into one behavioral type, while the remainder of the historically-derived model remains intact, and thus leads to the following hypothesis, which we call Model Three:

H3a. Directive leadership exists as a distinct behavioral type.

H3b. Transactional-transformational leadership exists as a distinct behavioral type including all transactional and transformational behaviors.

H3c. Empowering leadership exists as a distinct behavioral type.

Finally, a fourth theoretically plausible model suggests that directive leadership behavioral type could be split into two components, and that transactional leadership and transformational leadership may collapse into one behavioral type, while the remainder of the historically-derived model remains intact, and thus leads to the following hypothesis, which we call Model Four:

H4a. Directive leadership exists as a distinct behavioral type (excluding aversive leadership).

H4b. Aversive leadership exists as a distinct behavioral type.

H4c. Transactional-transformational leadership exists as a distinct behavioral type including all transactional and transformational behaviors.

H4d. Empowering leadership exists as a distinct behavioral type.

In the following section we report our tests of each of the theoretically plausible models using a third independent data set using CFA techniques. Our objective was to test competing models to determine which model most accurately accounts for the underlying structure of the relationships among the variables and to use this in the final theory building component of this research.

JMD 22,4

290

Phase III. Testing the theoretically plausible models Sample Cox (1994) studied the effects of leadership training on leader behavior and subordinate behavior. The description of leader behavior was conducted with a further refined version of the instrument used in data sets one and two. This study was conducted in one organization with a group of mid-level managers and their subordinates. The organization was located in the mid-Atlantic USA region and operates in the civilian defense electronics industry. The sample consisted of 702 participants. The first order factor analysis of this data set resulted in 15 distinct leader behaviors and is reported in Cox (1994). All dimensions demonstrated excellent reliability.

Analysis The method for analysis of the multiple theoretically plausible typologies identified in this research was CFA using LISREL. To examine model fit for each of the theoretically plausible models the x2 statistic, and the comparative fit index (CFI) statistic were used. In order to choose the model that provides the best fit to the data, we used the Akaike information criterion[3] (AIC) (Akaike, 1987).

Model One proposed that directive leadership, transactional leadership, transformational leadership, and empowering leadership, would exist as four distinct behavioral types. Model Two proposed that directive leadership, aversive leadership, transactional leadership, transformational leadership, and empowering leadership, would exist as five distinct behavioral types. Model Three, proposed that directive leadership, transactional-transformational leadership, and empowering leadership, would exist as three distinct behavioral types. Model Four proposed that directive leadership, aversive leadership, transactional-transformational leadership, and empowering leadership, would exist as four distinct behavioral types.

CFA results A total of 15 leader behaviors were analyzed using CFA techniques. Four theoretically plausible models were specified. The first model was the Phase I model derived from our historical analysis of leadership literature. This model proposed that directive leadership, transactional leadership, transformational leadership, and empowering leadership, would exist as four distinct behavioral types. Using CFA we found that this model did provide a reasonable overall account of the underlying structure of the data with a CFI score of 0.84, a x2

score (with 86 degrees of freedom) of 860.98 (p ¼ 0:00) where the x2 of the independence model was 4949.78, and an AIC of 928.98, with all path loadings achieving significance. Thus, we can conclude overall support for Model One.

The next three models were derived using exploratory factor analytic techniques in Phase II. Model Two proposed that directive leadership, aversive leadership, transactional leadership, transformational leadership, and

Transactors, transformers and

beyond

291

empowering leadership, would exist as five distinct behavioral types. Using CFA we found that this model did not provide a reasonable overall account of the underlying structure of the data. In fact, the model failed to converge. Thus, we rejected Model Two.

Model Three proposed that directive leadership, transactional- transformational leadership, and empowering leadership, would exist as three distinct behavioral types. Using CFA techniques we found that this model did provide a reasonable account of the underlying structure of the data with a CFI score of 0.82, a x2 score (with 87 degrees of freedom) of 956.92 (p ¼ 0:00) where the x2 of the independence model was 1022.92 with all path loadings achieving significance. Thus, we can conclude overall support for Model Three.

Model Four proposed that directive leadership, aversive leadership, transactional-transformational leadership, and empowering leadership, would exist as four distinct behavioral types. Using CFA techniques we found that this model did not provide a reasonable account of the underlying structure of the data. In fact, this model also failed to converge. Thus, we rejected Model Four.

While two models did provide reasonable overall accounts of the underlying structure of the data, two of the models failed to converge. The models that failed to converge were models that specified aversive leadership as a distinct behavioral type. It is possible that these models failed to converge simply because there was only one manifest indicator for the factor, “intimidation and non-contingent reprimand”. Moreover, it is interesting that the “intimidation and non-contingent reprimand” behavior seemed to cause empirical problems in both of the models that did converge (e.g. produced extremely high residuals).

Therefore, to overcome the empirical problems associated with the “intimidation and non-contingent reprimand” behavior we removed this dimension and re-estimated Models One and Three, now re-labeled Models Five and Six, respectively. Model five proposed that directive leadership, transactional leadership, transformational leadership, and empowering leadership, would exist as four distinct behavioral types. Again using CFA we found that this model did provide a reasonable overall account of the underlying structure of the data with a CFI score of 0.93, a x2 score (with 73 degrees of freedom) of 363.33 (p ¼ 0:00) where the x2 of the independence model was 4407.76, and an AIC of 427.33, with all path loadings achieving significance. Thus, we can conclude overall support for Model Five.

Model Six proposed that directive leadership, transactional-transformational leadership, and empowering leadership, would exist as three distinct behavioral types. Using CFA we found that this model did provide a reasonable overall account of the underlying structure of the data with a CFI score of 0.91, a x2 score (with 74 degrees of freedom) of 460.11 (p ¼ 0:00) where the x2 of the independence model was 4407.76, and an AIC of 522.11, with all

JMD 22,4

292

path loadings achieving significance. Thus, we can conclude overall support for Model Six.

Thus, we have found several models that could account for the underlying structure of leader behavior. However, our interest is to determine the best fitting of these theoretically plausible models for the construction of our final theoretical typology. To do this we examined the AIC statistic associated with each of the models that was supported. The smaller the AIC, the better the model accounts for the underlying structure of the data. Using the AIC statistic we can see that Model Five, which proposed four behavioral types of leadership – directive leadership, transactional leadership, transformational leadership, and empowering leadership – provides the best fit to the data, and is chosen as the best theoretical account of the behavioral types of leadership. Model Five is presented in Figure 1.

Discussion This research, as a whole, raises many questions about typologies and the content of those typologies. The following sections deal with these questions. Later, we will return to the specific results of this study.

Why leadership typologies? Why is research into leadership typologies important? First, both researchers and practitioners have long expressed a curiosity as to how leadership can be defined and conceptualized. For example, we often find managers referring to a particular executive as an “X” type of leader. The more one can capture real leadership with a conceptual representation, the more we can guide and channel real leaders to behave in more effective ways.

Researchers are most often interested in how particular kinds of leadership relate to individual, group, and organizational effectiveness. In fact, most scholarly inquiries into leadership take the form of theories that relate leadership to some set of outcome variables, especially performance. In a more practical vein, trainers are interested in systematic definitions of leader behavior as a framework for guiding their developmental efforts (Cox et al., in press). Indeed, theories of leadership are often the fundamental framework guiding selection, training, appraisal, and development interventions. Given that leadership is considered such an important independent variable in organizational research it is critical that we develop as clear and accurate representation of it as possible (Hunt, 1996). As Yukl (1998) stated so clearly, it is imperative that we understand the overall patterns of leader behavior and how they interrelate.

The emergence and disappearance of aversive leadership Aversive leadership was not an explicit part of our original theoretical formulation. We hypothesized that aversive behavior would manifest together with the directive behaviors. However, in the early phases of our analyses, we

Transactors, transformers and

beyond

293

found that aversive leadership emerged as a separate type. Thus, in the CFA, we hypothesized that directive and aversive may form separate behavioral types. Empirically, this model was not confirmed, and we dropped aversive behavior from our final best-fitting model.

Yet, we would not recommend that aversive behavior be dropped as a separate type in all future theory and analysis. One explanation of the failure to confirm is that we only had one manifest indicator of aversive behavior, i.e. the measurement of aversive behavior may not have been sufficiently developed to utilize a CFA framework. In essence, the failure to confirm might be considered

Figure 1. Graphical representation of the final CFA model

JMD 22,4

294

a “technical” or measurement problem. We suggest more work in the measurement of aversive behavior is needed.

Also, it is instructive to look back at the historical initiating structure analyses as a useful precedent. In the early stages, initiating structure was thought to be a holistic unified dimension. Later analysis showed differences between the different versions: The Stogdill (1963) version was more “directive” in nature, the Fleishman (1953) version was more “punitive” in nature. The excellent analysis of Schriesheim et al. (1976) delineated these differences and helped to clarify what initiating structure really measured. When we use the Schriesheim et al. (1976) findings as background, we are more prone to conclude that directive leadership is distinct from aversive (punitive) leadership.

An exhaustive and complete typology? One should ask if the typology presented here is exhaustive and complete. The obvious answer is no. This conceptual framework is only one of many that have been proposed to represent leadership. In 1991, for example, Fleishman et al. (1991) provided a superb review of many leadership taxonomies, and listed over 65 taxonomies of leadership. Hunt (1996) and Yukl (1998) also address this issue. Yukl (1998) presents 12 separate taxonomies as representative of the diversity of approaches to leadership[4], while Hunt (1996) presents an integration of ten separate typologies.

Among the more prominent typologies are the work of Podsakoff et al. (1982), Quinn (1988), Quinn et al. (1990), Lord and Maher (1991), Avolio et al. (1999), Bass and Avolio (1993), Kim and Yukl (1995), Hunt (1996) and Yukl (1998). It is beyond the scope of this paper to review and contrast these typologies in detail, but we do suggest that each typology has a particular advantage.

Lord and Maher (1991), for example, focus on the cognitive aspects of leadership. Hunt (1996) provides an alternative angle from which to view leadership – from that of a multi-level perspective. The work of Avolio et al. (1999) and Bass and Avolio (1993) deserves special mention because it seems to have attained a prominent centrality in current leadership research (e.g., typical quote from journal reviewers: “Why didn’t you use the Bass and Avolio instrument?”). Obviously, we support the overall intent of the Bass and Avolio instrument, since we also focus on the two main types of leadership included in that instrument, the transactional and transformational leader behaviors. In the work reported here, however, we have attempted to extend this framework by conceptualizing empowering leadership as a distinct type, and by having a stronger historical focus as represented by the directive and aversive types.

We suggest that empowering leadership is an example of how practice sometimes precedes theory. While the roots of empowering leadership in the

Transactors, transformers and

beyond

295

literature stretch back to Manz and Sims (1980, 1987), empowering leadership did not truly emerge as a distinct concept in the literature until the decade of the 1990s. Yet, clearly, managers were developing a form of this leadership in earlier decades (Manz and Sims, 1987). Thus, an important question is “How does the typology presented in this research compare to other recent typologies?” To address this question we compare and contrast the current typology with the typologies of Quinn (1988), Quinn et al. (1990) and Yukl (1998). We selected these two typologies because they are both broad in scope and rooted in years of empirical and theoretical tradition. Quinn (1988), Quinn et al. (1990) developed a competing values model of leadership based on a series of polar-opposite values (e.g. flexibility vs control, internal focus vs external focus). They assert that leaders tend to engage in behaviors at one end of a continuum to the exclusion (and potential lack of awareness) of the behavioral options at the opposite end. The four quadrants in the Quinn et al. (1990) framework are:

(1) consolidation orientation – which focuses on monitoring and coordinating;

(2) output maximization orientation – which focuses on directing and producing;

(3) change orientation – which focuses on brokering and innovating; and

(4) human resource development orientation – which focuses on mentoring and facilitating.

Within each quadrant they list six representative behaviors. Generally speaking the four quadrants demonstrate considerable conceptual

overlap with the typology presented in this research (see Table V). For example, all of the human resource development orientation behaviors are clearly aligned with the empowering leadership type. Moreover, five of the six consolidation orientation behaviors are closely aligned with the directive leadership type and five of the change orientation behaviors are closely aligned with the transformational leadership type. The behaviors from the output maximization orientation, on the other hand, are not extremely well aligned with the transactional leadership type: only one of the behaviors is clearly transactional in nature (motivating others). In fact, two of the output maximization orientation behaviors (personal productivity and motivation, and time and stress management) are not at all related to the model we present, but rather could be termed self-leadership behaviors. Nonetheless, 17 of the 24 behaviors, or more than 70 percent, demonstrate conceptual convergence with our model. Particularly noteworthy is the convergence on the importance of empowering leadership.

Yukl (1998) is another useful benchmark for the current study. This typology has a long and rich history of theoretical and empirical refinement (Kim and Yukl, 1995; Yukl, 1987). It is interesting to note that multiple

JMD 22,4

296

behaviors from the Yukl typology are clearly aligned with each of the leadership types in our model (see Table V). For example, directive leadership is clearly aligned with:

. organizing;

. problem solving;

. clarifying roles and objectives;

. informing; and

. monitoring.

Transactional leadership is clearly aligned with: . recognizing; and . rewarding.

Transformational leadership is clearly aligned with: . planning; . motivating and inspiring; and . networking.

Empowering leadership is clearly aligned with: . consulting; . delegating; . supporting; . developing and mentoring; and . managing conflict and team building.

It is interesting to note the considerable emphasis Yukl (1987) placed on behaviors outside of the transactional-transformational framework. Taken together, these two notable typologies lend further credence to the typology identified in this research. However, this is not to say that this research is the last word on leadership typologies.

Historically, leadership theorists have tended to pursue a holy grail of one- best-theory of leadership (we would include ourselves in this group). However, we suggest that not only is the one-best-theory elusive, it may be impossible because the way the environment in which leaders function is changing. That is, different times may require different forms of leadership. We have come to believe that defining a leadership typology is like shooting at a moving target. Just when it seems to finally fit well, the business, economic, and competitive environment changes, and new forms of leadership emerge in a natural manner. One example is the current interest in “shared leadership” (e.g. Avolio et al., 1996; Pearce, 1997, in press; Pearce and Conger, 2003; Pearce et al., 2001; Pearce and Sims, 2000, 2002; Perry et al., 1999; Seers, 1996).

Transactors, transformers and

beyond

297

C u

rr en

t st

u d

y ty

p o lo

g y

Q u

in n

a n

d co

ll ea

g u

es ty

p o lo

g y

Y u

k l

ty p

o lo

g y

C o n

so li

d a ti

o n

o ri

en ta

ti o n

O u

tp u

t m

a x

im iz

e o ri

en ta

ti o n

C h

a n

g e

o ri

en ta

ti o n

H R

d ev

el o p

m en

t o ri

en ta

ti o n

D ir ec ti v e

In st

ru ct

io n

a n

d co

m m

a n

d R

ec ei

v in

g a n

d o rg

a n

iz in

g in

fo rm

a ti

o n

T a k

in g

in it

ia ti

v e

O rg

a n

iz in

g A

ss ig

n ed

g o a ls

E v

a lu

a ti

n g

ro u

ti n

e in

fo rm

a ti

o n

G o a l

se tt

in g

P ro

b le

m so

lv in

g

C o n

ti n

g en

t re

p ri

m a n

d

R es

p o n

d in

g to

ro u

ti n

e in

fo rm

a ti

o n

C la

ri fy

in g

ro le

s a n

d o b

je ct

iv es

O rg

a n

iz in

g

In fo

rm in

g

C o n

tr o ll

in g

M o n

it o ri

n g

T ra n sa ct io n a l

C o n

ti n

g en

t m

a te

ri a l

re w

a rd

M o ti

v a ti

n g

o th

er s

N eg

o ti

a ti

n g

a g

re em

en t

a n

d co

m m

it m

en t

R ec

o g

n iz

in g

C o n

ti n

g en

t p

er so

n a l

re w

a rd

R ew

a rd

in g

T ra n sf o rm

a ti o n a l

S ti

m u

la ti

o n

a n

d in

sp ir

a ti

o n

P la

n n

in g

B u

il d

in g

a n

d m

a in

ta in

in g

a p

o w

er b

a se

P la

n n

in g

V is

io n

N eg

o ti

a ti

n g

a n

d se

ll in

g id

ea s

M o ti

v a ti

n g

a n

d in

sp ir

in g

Id ea

li sm

L iv

in g

w it

h ch

a n

g e

N et

w o rk

in g

C h

a ll

en g

e to

st a tu

s q

u o

C re

a ti

v e

th in

k in

g M

a n

a g

in g

ch a n

g e

E m p o w er in g

E n

co u

ra g

es o p

p o rt

u n

it y

th in

k in

g D

el eg

a ti

n g

ef fe

ct iv

el y

U n

d er

st a n

d in

g se

lf a n

d o th

er s

C o n

su lt

in g

E n

co u

ra g

es se

lf -r

ew a rd

In te

rp er

so n

a l

co m

m u

n ic

a ti

o n

D el

eg a ti

n g

E n

co u

ra g

es se

lf -l

ea d

er sh

ip D

ev el

o p

in g

su b

o rd

in a te

s

S u

p p

o rt

in g

P a rt

ic ip

a ti

v e

g o a l

se tt

in g

T ea

m b

u il

d in

g

D ev

el o p

in g

a n

d m

en to

ri n

g E

n co

u ra

g es

te a m

w o rk

P a rt

ic ip

a ti

v e

d ec

is io

n m

a k

in g

C o n

fl ic

t m

a n

a g

em en

t

M a n

a g

in g

co n

fl ic

t a n

d te

a m

b u

il d

in g

B eh a v io rs

n o t co n ta in ed

in th e

cu rr en t st u d y ty p o lo g y

P er

so n

a l

p ro

d u

ct iv

it y

a n

d m

o ti

v a ti

o n

T im

e a n

d st

re ss

m a n

a g

em en

t

Table V. Comparison of the Quinn and colleagues and Yukl typologies with the typology of the current study

JMD 22,4

298

In the end, we suggest that there is no “one best” typology of leadership. We agree with Yukl (1998, p. 63), when he wrote the following:

It is important to remember that all behavior taxonomies are arbitrary and have no validity in any absolute sense. Unfortunately, there has been too much preoccupation with finding and using the “correct” set of behavior categories. It is essential to be flexible about the behavior constructs used in analyzing patterns of leadership behavior, rather than assuming that we already know what constructs will be most useful.

In the spirit of Yukl (1998), we believe we have presented one typology that represents one step in our development of leadership theory. The typology presented here draws deeply on historical theories, yet breaks new ground because of its hierarchical nature, and its emphasis on empowering leadership. Nonetheless, it is only one view of how we might conceptualize leadership. Other views will continue to emerge and flourish.

Conclusions Using historical analysis of leadership theory and research, we deductively derived a model of leadership that contained four broad level types of leadership. We subsequently, inductively derived alternative, theoretically plausible models of the types of leadership using EFA techniques, with two independent data sets. Finally, we empirically tested each of the competing, theoretically plausible models, using CFA techniques, with a third, independent data set. Our results clearly suggest that a two factor theory of leadership does not adequately account for the underlying nature of the phenomenon we call leadership. This research suggests that there are at least four types of leadership:

(1) The directive leadership type which includes the following behavioral set: . issuing instructions and commands; . assigning goals; and . contingent reprimand.

(2) The transactional leadership type which includes the following behavioral set: . dispensing contingent material rewards; and . dispensing contingent personal rewards.

(3) The transformational leadership type which includes the following behavioral set: . providing a sense of vision; . challenging the status quo; . engaging in idealism; and . providing stimulation and inspiration.

Transactors, transformers and

beyond

299

(4) The empowering leadership type which includes the following behavioral set: . encouraging opportunity thinking; . encouraging self-rewards; . encouraging self-leadership; . engaging in participative goal setting; and . encouraging teamwork.

This research, which has extended over more than a decade, has made some significant contributions. For example, the historically-derived directive leadership behavioral type, which has its roots in theory X and initiating structure research, had lost the attention of leadership researchers during the recent past. It is unfortunate that this grouping of leader behaviors seems to have gone astray when the field shifted away from the consideration-initiating structure paradigm. Our research suggests that directive leadership does indeed exist as an independent type of leadership, and as such provides a link to the more distant historical roots of leadership research. However, while the historically-derived model suggested that this type included both “directive” and “aversive” behaviors, our further analyses suggested that directive and aversive leader behaviors may form two distinct types of leadership. This is an issue for further investigation.

From a conceptual viewpoint, our earlier theoretical discussion presented a case for distinguishing between directive, transactional, transformational and empowering leader behaviors. The conceptual/theoretical differentiation between these types of leadership is clear. Directive leadership refers to behaviors that are primarily associated with the leader giving orders on how work is to be done. Transactional leadership refers to the behaviors that establish the conditions of the exchange relationship between leader and follower. Transformational leadership refers to behaviors that define and sustain vision, elicit inspiration from followers, and stimulate change. In contrast, empowering leadership is mainly aimed at the development of followers, eventually molding them into effective self-leaders that are capable of initiative, creativity, and acting on their own.

As discussed in the previous section, the theoretical conceptualization of Quinn (1988) and Quinn et al. (1990) and our results converge on the notion of empowerment as a distinct type. Yet, in our earlier analyses, the leadership types did not always remain entirely separate. In study one, for example, many of the transactional dimensions grouped with the transformational behaviors. In the final study, however, using the CFA methodology, the types were clearly different. Other empirical research reported by Cohen et al. (1997), Pearce and Sims (2002) and Pearce et al. (in press) also found this empirical distinction.

JMD 22,4

300

While the Quinn (1988) and Quinn et al. (1990) model purports that leaders generally display behaviors indicative of one, but not both, ends of various competing values continuums, they also suggested that “master managers” could be high on all of the orientations. Thus, while Quinn (1988) and Quinn et al. (1990) defined different types of leaders they also recognized that leaders could engage in multiple types of leader behavior either simultaneously or sequentially. In other unpublished research, not reported here, Pearce et al. (2002) pursued the question of whether a specific leader could be both transactional and transformational at the same time, or if these two types of leadership are mutually exclusive. Using latent class modeling techniques, they concluded that some leaders were both transactional and transformational; others were transactional only; others were transformational only; and some were neither transactional nor transformational. In essence, transactional and transformational leadership were distinct, yet not mutually exclusive.

The purpose of a typology, however, is not the definition of mutually exclusive categories, but rather the identification of interrelated sets of phenomena (Doty and Glick, 1994) Thus, our data clearly show that a leader who is high on transactional leader behaviors is likely to also be high on transformational and empowering behaviors as well. Thus, we come down on the side that presents the types as non-mutually exclusive.

Most of our analyses led us to the conclusion that the various types of leadership are conceptually distinct, and often empirically distinct, yet are highly related. Most of all, these forms of leadership are not mutually exclusive, and a specific leader is not required to choose between being one vs another. Indeed, a specific leader may actually use behaviors from more than one type. Clearly, more research is needed to explore and articulate the boundaries between the various types of leadership.

Most of all, this research provides a benchmark that facilitates examination of leadership typologies. Moreover, the study is a demonstration of the utility of a sequential, inductive, deductive approach to theory building, combined with confirmatory testing of multiple theoretically plausible models to arrive at a final theoretical framework. As Yukl (1989) noted, we should not let predetermined theoretical notions bias our subsequent research, but rather we should attempt to explore and determine the dimensions and types of leadership with rigorous empirical investigations.

Leadership theory is a dynamic evolving paradigm that moves forward in fits and starts rather than in a linear fashion. A focus on empowering leadership, as provided here, is not an end, but is an important way station in a continuing stream of theory development. Leadership theory has been stuck in the transactional/transformational mode for more than a decade. It’s time to move beyond, while not forgetting what has come before.

Transactors, transformers and

beyond

301

Notes

1. An early version of this historical analysis can be found in Cox (1994).

2. A priori, we felt that “participative goal setting” may actually be part of the transactional leadership type, but only presented it as part of the empowering leadership type in order to simplify the text and figures. However, we also tested each of the theoretically plausible models with “participative goal setting” as part of the transactional leadership type. None of these models fit as well as when we specified “participative goal setting” to be part of the empowering leadership type.

3. Note that these models are not subsuming or nested. Therefore, it is not possible to compare the relative fit across models using the x 2 statistic. Thus, in order to choose the model that provides the best fit to the data, we used the Akaike information criterion (AIC) (Akaike, 1987). The AIC permits the examination of non-nested models for relative fit to permit the researcher to choose from among multiple theoretical models (McCutcheon, 1987). “This approach can be used regardless of whether the models can be ordered in a nested sequence or not” (Jöreskog, 1993, p. 306). The AIC has been utilized extensively in educational research (e.g. Macready, 1982) and is the preferred method for selecting among competing non-nested models, particularly when complex models are specified.

4. In a recent work, Doty and Glick (1994) distinguished between a taxonomy and a typology. A taxonomy is a classification system that categorizes “phenomena into mutually exclusive and exhaustive sets with a series of discrete rules” whereas a typology contains “multiple ideal types, each of which represents a unique combination of the organizational attributes that are believed to determine relevant outcomes(s)” (Doty and Glick, 1994, p. 232). According to this more restricted definition, we believe that Fleishman et al. (1991) and Yukl (1998) actually focus on typologies.

References

Adams, J.S. (1963), “Wage inequities, productivity and work quality”, Industrial Relations, Vol. 3 p. 916.

Akaike, H. (1987), “Factor analysis and AIC”, Psychometrika, Vol. 52, pp. 317-32.

Arvey, R.D. and Ivancevitch, J.M. (1980), “Punishment in organizations: a review, propositions, and research suggestions”, Academy of Management Review, Vol. 5, pp. 123-32.

Avolio, B.J. and Bass, B.M. (1988), “Transformation leadership, charisma, and beyond”, in Hunt, J.G., Baliga, B.R., Dachler, H.P. and Schreisheimm, C.A. (Eds), Emerging Leadership Vistas, Lexington, MA, pp. 29-50.

Avolio, B.J., Bass, B.M. and Jung, D. (1999), “Re-examining the components of transactional and transformational leadership using the Multifactor Leadership Questionnaire”, Journal of Occupational and Organizational Psychology, Vol. 72, pp. 441-62.

Avolio, B.J., Jung, D.I., Murry, W. and Suvasubramaniam, N. (1996), “Building highly developed teams: focusing on shared leadership processes, efficacy, trust, and performance”, Advances in Interdiciplinary Studies of Work Teams, Vol. 3, pp. 173-209.

Ball, G.A. (1991), “Outcomes of punishment incidents: the role of subordinate perceptions, individual differences, and leader behavior”, PhD dissertation, The Pennsylvania State University, University Park, PA

Ball, G.A., Sims, H.P. and Trevino, L.K. (1994), “Just and unjust punishment: influences on subordinate performance and citizenship”, Academy of Management Journal, Vol. 37, pp. 299-322.

Bandura, A. (1986), Social Foundations of Thought and Action: A Social Cognitive Theory, Prentice-Hall, Englewood Cliffs, NJ.

JMD 22,4

302

Bass, B.M. (1985), Leadership and Performance Beyond Expectations, Free Press, New York, NY.

Bass, B.M. (1990), Bass and Stogdill’s Handbook of Leadership: Theory, Research and Managerial Applications, 3rd ed., Free Press, New York, NY.

Bass, B.M. and Avolio, B.J. (1993), “Transformational leadership: a response to critiques”, in Hunt, J.G., Baliga, B.R., Dachler, H.P. and Schriesheim, C.A. (Eds), Emerging Leadership Vistas, D.C. Health, Lexington, MA, pp. 29-40.

Bass, B.M., Avolio, B.J. and Goodheim, L. (1987), “Biography and the assessment of transformational leadership at the world-class level”, Journal of Management, Vol. 13, pp. 7-19.

Bryman, A. (1992), Charisma and Leadership in Organizations, Sage Publications, Newbury Park, CA.

Burns, J.M. (1978), Leadership, Harper & Row, New York, NY.

Cohen, S.G., Chang, L. and Ledford, G.E. (1997), “A hierarchical construct of self-management leadership and its relationship to quality of work life and perceived group effectiveness”, Personnel Psychology, Vol. 50, pp. 275-308.

Cox, J.F. (1994), “The effects of superleadership training on leader behavior, subordinate self- leadership behavior, and subordinate citizenship”, Dissertation Abstracts International, Vol. 55.

Cox, J.F., Pearce, C.L. and Sims, H.P. Jr (in press), “Toward a broader agenda for leadership development: extending the traditional transactional-transformational duality by developing directive, empowering and shared leadership skills”, in Riggio, R.E. and Murphy, S. (Eds), The Future of Leadership Development, Lawrence Earlbaum, Mahwah, NJ.

Doty, D.H. and Glick, W.H. (1994), “Typologies as a unique form of theory building: toward improved understanding and modeling”, Academy of Management Review, Vol. 19 No. 2, pp. 230-51.

Drucker, P.F. (1954), The Practice of Management, Harper & Row, New York, NY.

Dubin, R. (1978), Theory Building, rev ed.,, The Free Press, New York, NY.

Erez, M. and Arad, R. (1986), “Participative goal-setting: social, motivational, and cognitive factors”, Journal of Applied Psychology, Vol. 71, pp. 591-7.

Fleishman, E.A. (1953), “The description of supervisory behavior”, Personnel Psychology, Vol. 37, pp. 1-6.

Fleishman, E.A., Mumford, M.D., Zaccaro, S.J., Levin, K.Y., Korotkin, A.L. and Hein, M.B. (1991), “Taxonomic efforts in the description of leadership behavior: a systhesis and functional interpretation”, The Leadership Quarterly, Vol. 2 No. 4, pp. 245-87.

French, J.R.P. and Raven, B. (1959), “The bases of social power”, in Cartwright, D. (Ed.), Studies in Social Power, Institute for Social Research, University of Michigan, Ann Arbor, MI.

Gorsuch, R.L. (1983), Factor Analysis, 2nd ed., Lawrence Earlbaum Associates, Hillsdale, NJ.

Halpin, A.W. and Winer, B.J. (1957), “A factorial study of the leader behavior descriptions”, in Stogdill, R.M. and Coons, E.A. (Eds), Leader Behavior: Its Description and Measurement, Ohio State University, Bureau of Business Research, Columbus, OH.

Hatter, J.J. and Bass, B.M. (1988), “Superiors’ evaluations and subordinates’ perceptions of transformational and transactional leadership”, Journal of Applied Psychology, Vol. 73 No. 4, pp. 695-702.

Homans, G.C. (1958), “Social behavior as exchange”, American Journal of Sociology, pp. 597-606.

Homans, G.C. (1961), Social Behavior: Its Elementary Forms, Harcourt, Brace, New York, NY.

Transactors, transformers and

beyond

303

House, R.J. (1971), “A path goal theory of leader effectiveness”, Administrative Science Ouarterly, Vol. 16, pp. 321-38.

House, R.J. (1977), “A 1976 theory of charismatic leadership”, in Hunt, J.G. and Larson, L.L. (Eds), Leadership: The Cutting Edge, Southern Illinois University Press, Carbondale, IL.

House, R.J. and Dessler, G. (1974), “The path goal theory of leadership: some post hoc and a priori tests”, in Hunt, J.G. and Larson, L.L. (Eds), Contingency Approaches to Leadership, Southern Illinois University Press, Carbondale, IL.

House, R.J. and Mitchell, T.R. (1974), “Path-goal theory of leadership”, Journal of Contemporary Business, Vol. 3, pp. 81-97.

House, R.J. and Shamir, B. (1993), “Toward the integration of transformational, charismatic and visionary theories of leadership”, in Chemmers, M. and Ayman, R. (Eds), Leadership: Perspectives and Research Directions, Academic Press, New York, NY.

House, R.J., Howell, J.M., Shamir, B., Smith, B. and Spangler, W.D. (1993), “Charismatic leadership: a 1993 theory and five empirical tests”, unpublished manuscript.

Hunt, J.G. (1996), Leadership: A New Synthesis, Sage Publications, Newbury Park, CA.

Joreskog, K.G. (1993), “Testing structural equation models”, in Bollen, K.A. and Long, J.S. (Eds), Testing Structural Equation Models, Sage Publications, Newbury Park, CA.

Katz, D., Maccoby, N. and Morse, N. (1950), Productivity, Supervision, and Morale in an Office Situation, Institute for Social Research, Ann Arbor, MI.

Kazdin, A.E. (1975), Behavior Modification in Applied Settings, Dorsey, Homewood, IL.

Kim, H. and Yukl, G. (1995), “Relationships of self-reported and subordinate-reported leadership to managerial effectiveness and advancement”, Leadership Quarterly, Vol. 6, pp. 361-77.

Korukonda, A.R. and Hunt, J.G. (1989), “Pat on the back vs kick in the pants: an application of cognitive inference to the study of leader reward and punishment behaviors”, Group & Organization Studies., Vol. 14 No. 3, pp. 299-324.

Landy, F.J. (1985), Psychology of Work Behavior, 3rd ed., Dorsey Press, Homewood, IL.

Likert, R. (1961), “An emerging theory of organizations, leadership and management”, in Petrullo, L. and Bass, B.M. (Eds), Leadership and Interpersonal Behavior, Holt, Reinhart & Winston, New York, NY.

Likert, R. (1967), The Human Organization, McGraw-Hill, New York, NY.

Locke, E.A. and Latham, G.P. (1990), A Theory of Goal Setting and Task Performance, Prentice- Hall, Englewood Cliffs, NJ.

Lord, R.G. and Maher, K.J. (1991), Leadership and Information Processing: Linking Perceptions and Performance, Unwin-Hyman, Boston, MA.

Luthans, F. and Kreitner, R. (1985), Organizational Behavior Modification and Beyond, Scott- Foresman, Glenview, IL.

McCutcheon (1987), Latent Class Analysis, Sage, Newbury Park, CA.

McGregor, D. (1960), The Human Side of Enterprise, McGraw-Hill, New York, NY.

Macready, G.B. (1982), “The use of latent class models for assessing prerequisite relations and transference among traits”, Psychometrika, Vol. 47 No. 4, pp. 477-88.

Mahoney, M.J. and Arnkoff, D.B. (1978), “Cognitive and self-control therapies”, in Garfield, S.L. and Borgin, A.E. (Eds), Handbook of Psychotherapy and Therapy Change, Wiley, New York, NY, pp. 689-722.

Manz, C.C. and Sims, H.P. Jr (1980), “Self-management as a substitute for leadership: a social learning theory perspective”, Academy of Management Review, Vol. 5, pp. 361-7.

JMD 22,4

304

Manz, C.C. and Sims, H.P. Jr (1987), “Leading workers to lead themselves: the external leadership of self-managing work teams”, Administrative Science Quarterly, Vol. 32, pp. 106-28.

Manz, C.C. and Sims, H.P. Jr (1990), SuperLeadership, Berkeley Books, New York, NY.

Manz, C.C. and Sims, H.P. Jr (1991), “SuperLeadership: beyond the myth of heroic leadership”, Organizational Dynamics, pp. 18-35.

Manz, C.C. and Sims, H.P. Jr (2001), The New SuperLeadership, Berrett-Koehler, San Francisco, CA.

Maslow, A.H. (1954), Motivation and Personality, Harper, New York, NY.

Meichenbaum, D. (1977), Cognitive-behavior Modification: An Integrative Approach, Plenum Press, New York, NY.

Pearce, C.L. (1997), “The determinants of change management team effectiveness: a longitudinal investigation”, unpublished doctoral dissertation, University of Maryland, College Park, MA.

Pearce, C.L. (in press), “Más allá del liderazgo heroico: como el buen vino, el liderazgo es algo para ser compartido”, Revista de Empresa.

Pearce, C.L. and Conger, J.A. (Eds) (2003), Shared Leadership: Reframing the Hows and Whys Leadership, Sage Publications, Thousand Oaks, CA.

Pearce, C.L. and Sims, H.P. Jr (2000), “Shared leadership: toward a multi-level theory of leadership”, Advances in Interdisciplinary Studies of Work Teams, Vol. 7, pp. 115-39.

Pearce, C.L. and Sims, H.P. Jr (2002), “Vertical versus shared leadership as predictors of the effectiveness of change management teams: an examination of aversive, directive, transactional, transformational, and empowering leader behaviors”, Group Dynamics: Theory, Research, and Practice, Vol. 6 No. 2, p. 172-97.

Pearce, C.L., Perry, M.L. and Sims, H.P. Jr (2001), “Shared leadership: relationship management to improve NPO effectiveness”, in Connors, T.D. (Ed.), The Nonprofit Handbook: Management, Wiley, New York, NY, pp. 624-41.

Pearce, C.L., Sims H.P. Jr and Cox, J.F. (2002), “Can a leader be both transactional and transformational?: An empirical analysis using latent class modeling”, working paper, Claremont Graduate University, Claremont, CA

Pearce, C.L., Yoo, Y. and Alavi, M. (in press), “Leadership, social work and virtual teams: The relative influence of vertical vs shared leadership in the nonprofit sector”, in Riggio, R.E. and Smith-Orr, S. (Eds), Improving Leadership in Nonprofit Organizations, Jossey-Bass, San Francisco, CA.

Perry, M.L., Pearce, C.L. and Sims, H.P. Jr (1999), “Empowered selling teams: how shared leadership can contribute to selling team outcomes”, Journal of Personal Selling and Sales Management, Vol. 19 No. 3, pp. 35-52.

Phillips, J.S. and Lord, R.G. (1986), “Notes on the practical and theoretical consequences of implicit leadership theories for the future of leadership measurement”, Journal of Management, Vol. 12 No. 1, pp. 31-41.

Pinder, C. (1984), Work Motivation: Theory, Issues and Applications, Scott, Foresman and Company, Glenview, IL.

Podsakoff, P.M., Todor, W.D. and Skov, R. (1982), “Effect of leader contingent and noncontingent reward and punishment behaviors on subordinate performance and satisfaction”, Academy of Management Journal, Vol. 25, pp. 810-21.

Quinn, R.E. (1988), Beyond Rational Management: Mastering the Paradoxes and Competing Demands of High Performance, Jossey-Bass, San Francisco, CA.

Transactors, transformers and

beyond

305

Quinn, R.E., Faerman, S.R., Thompson, M.P. and McGrath, M.R. (1990), Becoming a Master Manager, Wiley, New York, NY.

Rummel, R.J. (1970), Applied Factor Analysis, Northwestern University Press, Evanston, IL.

Sashkin, M. and Rosenbach, W.E. (1993), “A new leadership paradigm”, in Rosenbach, W.E. and Taylor, R.L. (Eds), Contemporary Issues in Leadership, 3rd ed., Westview Press, Boulder, CO.

Schriesheim, C.A., House, R.J. and Kerr, S. (1976), “Leader initiating structure: a reconciliation of discrepant research results and some empirical tests”, Organizational Behavior and Human Performance, Vol. 15, pp. 297-321.

Scott, W.E. and Podsakoff, P.M. (1982), “Leadership, supervision and behavioral control: perspectives from an experimental analysis”, in Frederickson, L. (Ed.), Handbook of Organizational Behavior Management, Wiley, New York, NY.

Scully, J.S., Sims, H.P. Jr, Olian, J.D., Smith, K.G., Schnell, E.R. and Smith, K.A. (1994), “Tough times make tough bosses: a meso-analysis of CEO leader behavior”, Leadership Quarterly, Vol. 5 No. 1, pp. 59-83.

Seers, A. (1996), “Better leadership through chemistry: toward a model of emergent shared team leadership”, Advances in Interdisciplinary Studies of Work Teams, Vol. 3, pp. 145-72.

Sims, H.P. Jr (1977), “The leader as manager of reinforcement contingencies: an empirical example and a model”, in Hunt, J.G. and Larson, L.L. (Eds), Leadership: The Cutting Edge, Southern Illinois University Press, Carbondale, IL.

Sims, H.P. Jr (1980), “Further thoughts on punishment in organizations”, Academy of Management Review, Vol. 5, pp. 133-8.

Stogdill, R.M. (1963), Manual for the Leader Behavior Description Questionnaire – Form XII, Ohio State University, Bureau of Business Research, Columbus, OH.

Thorenson, C.E. and Mahoney, M.J. (1974), Behavioral Self-control, Holt, Rinehart & Winston, New York, NY.

Thorndike, E.L. (1911), Animal Intelligence: Experimental Studies, Macmillan, New York, NY.

Trice, H.M. and Beyer, J.M. (1986), “Charisma and its routinization in two social movement organizations”, Research in Organizational Behavior, Vol. 8, pp. 113-64.

Vroom, V.H. (1964), Work and Motivation, Wiley, New York, NY.

Waldman, D.A., Bass, B.M. and Einstein, W.O. (1987), “Leadership and outcomes of performance appraisal process”, Journal of Occupational Psychology, Vol. 60, pp. 177-86.

Weber, M. (1924/1947), The Theory of Social and Economic Organization, Free Press, New York, NY.

Weber, M. (1946), “The sociology of charismatic authority”, in Mills, H.H. and Mills, C.W. (Eds), From Max Weber: Essays in Sociology, Oxford University Press, New York, NY.

Yammarino, F.J. and Bass, B.M. (1990), “Long-term forecasting of transformational leadership and its effects among naval officers”, in Clark, K.E. and Clark, M.B. (Eds), Measures of Leadership, Leadership Library of America, West Orange, NJ, pp. 151-70.

Yukl, G.A. (1987), “A new taxonomy for integrating diverse perspectives on managerial behavior”, paper presented at the American Psychological Association Meeting, New York, NY

Yukl, G.A. (1989), Leadership in Organizations, 2nd ed., Prentice-Hall, Englewood Cliffs, NJ.

Yukl, G.A. (1998), Leadership in Organizations, 4th ed., Prentice-Hall, Englewood Cliffs, NJ.

JMD 22,4

306

Further reading

Brown, D.J. and Lord, R.G. (1999), “The utility of experimental research in the study of transformational/charismatic leadership”, Leadership Quarterly, Vol. 10 No. 4, pp. 531-40.

Conger, J.A. (1999), “Charismatic and transformational leadership in organizations: an insider’s perspective on these developing streams of research”, Leadership Quarterly, Vol. 10 No. 2, pp. 145-70.

Hayduk, L.A. (1987), Structural Equation Modeling With LISREL: Essentials and advances, The Johns Hopkins University Press, Baltimore, MD.

House, R.J. (1999), “Weber and the neo-charismatic leadership paradigm: a response to Beyer”, Leadership Quarterly, Vol. 10 No. 4, pp. 563-75.

Hunt, J.G. (1999), “Transformational/charismatic leadership’s transformation of the field: an historical essay”, Leadership Quarterly, Vol. 10 No. 2, pp. 129-35.

Hunt, J.G. and Conger, J.A. (1999), “From where we sit: an assessment of transformational and charismatic leadership research”, Leadership Quarterly, Vol. 10 No. 3, pp. 335-44.

Joreskog, K. and Sorbom, D. (1993), LISREL 8: Structural Equation Modeling with the SIMPLIS Command Language, Lawrence Earlbaum, London.

Wofford, J.C. (1999), “Laboratory research on charismatic leadership: fruitful or futile?”, Leadership Quarterly, Vol. 10 No. 4, pp. 523-30.

Yukl, G.A. and Howell, J.M. (1999), “Organizational and contextual influences on the emergence and effectiveness of charismatic leadership”, Leadership Quarterly, Vol. 10 No. 2, pp. 257-84.

Transactors, transformers and

beyond

307