Conger Et Al 2000

Journal of Organizational Behavior
J. Organiz. Behav. 21, 747±767 (2000)
Charismatic leadership and follower eects

JAY A. CONGER1*, RABINDRA N. KANUNGO2
AND SANJAY T. MENON2
1Center for Eective Organizations, Marshall School of Business, Los Angeles, California, U.S.A.
2Faculty of Organizational Studies, School of Business, Clarkson University, New York, U.S.A.
Summary On the basis of the current theories of charismatic leadership, several possible follower
eects were identi®ed. It is hypothesized that followers of charismatic leaders could be
distinguished by their greater reverence, trust, and satisfaction with their leader and by a
heightened sense of collective identity, perceived group task performance, and feelings of
empowerment. Using the Conger±Kanungo charismatic leadership scale and measures
of the hypothesized follower eects, an empirical study was conducted on a sample of
252 managers using structural equation modelling. The results show a strong relation-
ship between follower reverence and charismatic leadership. Follower trust and satisfac-
tion, however, are mediated through leader reverence. Followers' sense of collective
identity and perceived group task performance are aected by charismatic leadership.
Feelings of empowerment are mediated through the followers' sense of collective identity
and perceived group task performance. Copyright # 2000 John Wiley & Sons, Ltd.
Introduction
Prior to the 1980's, charismatic leadership was a relatively obscure research topic within the ®eld
of organizational behavior. Only a handful of references are to be found from this period, and
these consist of speculative, formative theories (Berlew, 1974; House, 1977; Katz and Kahn, 1978;
Zaleznik and Kets de Vries, 1975). Since the late 1980's, however, interest in the topic has grown
signi®cantly. More comprehensive theories have been developed (Bass 1985; Bass and Avolio,
1993; Conger and Kanungo, 1987, 1998; Conger, 1989; Shamir et al., 1993) which in turn have
encouraged empirical studies (e.g., Agle and Sonnenfeld, 1994; Bass and Yammarino, 1988;
Conger and Kanungo, 1992, 1994; Conger et al., 1997; House et al., 1991; Howell and Frost,
1989; Podsako et al., 1990; Puer, 1990; Shamir, 1992; Shamir et al., 1998). While theoretical
work has tended to focus principally upon the behavioral dimensions of charismatic leaders, the
existing theories do predict certain follower eects. For example, it has been hypothesized that
charismatic leadership produces higher performance levels among followers as well as more
motivated and satis®ed followers (Bass, 1985; Conger and Kanungo, 1988). In a review of
empirical investigations of charismatic and transformational leadership, Shamir et al. (1993)
found that charismatic leadership was indeed positively correlated with followers' performance
and satisfaction. As Shamir et al. (1993) noted, the eect size of these studies of the behavior of
* Correspondence to: J. A. Conger, Center for Eective Organizations, Marshall School of Business, University of
Southern California, Bridge Hall 308D, Los Angeles, CA 90089, U.S.A.
Copyright # 2000 John Wiley & Sons, Ltd. Received 17 July 1998
Accepted 17 June 1999
748 J. A. CONGER, R. N. KANUNGO AND S. T. MENON
charismatic leaders on follower performance and satisfaction is higher than prior ®eld study
®ndings of other forms of leader behavior, with correlations of 0.50 or better and well below
0.01 probability of error due to chance.
Three important issues, however, remain to be addressed regarding follower eects. First, there
is a need for further validation studies given the limited research attention that the subject has
received (e.g., Bass, 1985; Howell and Frost, 1989; Podsako et al., 1990; Shamir et al., 1998).
Second, there is a need to investigate causal links between individual leader behaviors and
follower eects. As Kirkpatrick and Locke (1996) note, there has been a serious lack of research
which unequivocally demonstrates causal relationships between speci®c charismatic leadership
components and follower eects. This is despite the fact that charismatic leadership is a
multidimensional phenomenon (Conger and Kanungo, 1994) where individual components or
combinations of components may have diering eects. Moreover, two of the three studies which
examined causal relations between follower eects and components of charismatic leadership
were laboratory studies conducted on student samples (Howell and Frost, 1989; Kirkpatrick and
Locke, 1996). As such, there is a need for research involving managerial samples. As importantly,
these studies were experiments built around short time-frames which may not have adequately
fostered the intense leader±follower bonds that typically evolve under charismatic leadership
(Bass, 1985; Conger and Kanungo, 1987; Locke et al., 1991). As a result, neither study may have
allowed for certain relational and performance eects to emerge which aect outcome variables.
The third study, while based on a ®eld sample, was conducted in a military setting (Shamir et al.,
1998). For the above reasons, further study is warranted in assessing follower eects under
charismatic leadership.
The Conger±Kanungo model of

charismatic leadership
Recently, Conger and Kanungo (1987, 1998); Conger, (1989) developed a model of charismatic
leadership within organizational settings, and a measurement scale (Conger and Kanungo, 1994;
Conger et al., 1997). According to the model, charismatic leadership is an attribution based on
follower perceptions of their leader's behavior. The observed behavior of the leader is interpreted
by followers as expressions of charisma in the same sense as a leader's behaviors re¯ect that
individual's participative, people, and task orientations. Charismatic leaders dier from other
leaders by their ability to formulate and articulate an inspirational vision and by behaviors and
actions that foster an impression that they and their mission are extraordinary. As such, indi-
viduals choose to follow such leaders in management settings not simply because of the leader's
formal authority but out of perceptions of their leader's extraordinary character (Weber 1925/
1968). As such, any measurement of charismatic leadership must be based on followers' percep-
tions of the speci®c behavioral attributes of the leader that engender such eects. The Conger±
Kanungo model proposes several distinguishing behavioral components in three distinct stages of
the leadership process.
Speci®cally, in stage one (the environmental assessment stage), the Conger±Kanungo model
distinguishes the charismatic leadership of a manager from other leadership roles along two
dimensions. The ®rst are the followers' perceptions of the manager's greater desire to change the
status quo. The second is a heightened sensitivity on the leader's part to environmental oppor-
tunities, constraints, and followers' needs. Managers who are seen as charismatic will therefore be
more likely to be perceived as both critics of the status quo and as reformers or agents of radical
Copyright # 2000 John Wiley & Sons, Ltd. J. Organiz. Behav. 21, 747±767 (2000)
CHARISMATIC LEADERSHIP AND FOLLOWER EFFECTS 749
reform. In stage two (the vision formulation stage), charismatic leadership is distinguished from
other forms by the followers' perceptions of the manager's formulation of a shared and idealized
future vision as well as his or her eective articulation of this vision in an inspirational manner. It
is the shared perspective of the vision and its potential to satisfy follower needs that form the
basis of attraction to the charismatic leader. It is the idealized aspect, however, that makes such
leaders admirable persons deserving of respect and worthy of identi®cation and imitation by
followers. Finally, in stage three (the implementation stage), managers who are perceived as
charismatic are seen to be engaging in exemplary acts that subordinates interpret as involving
great personal risk and self-sacri®ce. Through these actions, managers are able to empower
subordinates and build trust. Furthermore, in stage three, managers seen as charismatic will also
be perceived to be deploying innovative and unconventional means for achieving their visions.
Their ability to transcend the existing order through unconventional approaches heightens
follower perceptions of the leader's expertize and follower perceptions of control over events.
While research has been conducted on the Conger±Kanungo model (e.g., Conger and
Kanungo, 1992, 1994; Conger et al., 1997), there has been an absence of empirical investigations
examining hypothesized links between the model's individual behavioral dimensions and speci®c
follower eects. For example, Conger and Kanungo (1988) suggested that charismatic leadership
might produce high levels of a collectivist identity among followers, strong emotional attachment
to the leader, and high follower task performance. Similarly, other researchers have proposed
distinct follower eects under charismatic leadership (e.g., Bass, 1985; Shamir et al., 1993). The
purpose of this study is to address this shortcoming.
Hypothesized relationships between charismatic

leader behaviors and follower eects
When subordinates perceive that their manager is exhibiting leadership behaviors (as speci®ed in
the Conger±Kanungo model, 1987), they will not only attribute charisma to him or her ( follower
attribution eect) but also change their attitudes, values, and behavior consistent with what the
manager wants from them. Such leadership in¯uence of a manager on subordinates is referred to
as follower eects. These eects are noticed on two fronts. A manager's leadership behavior can
in¯uence subordinates' reactions (a) towards the manager as a leader and (b) towards themselves
as subordinates and their task eorts. We call these leader focus variables and follower focus
variables.
This study sought to test three hypotheses relating to subordinate beliefs and attitudes toward
the manager as a leader, and three hypotheses relating to follower beliefs and attitudes con-
cerning the task. Speci®cally, under charismatic leadership, we propose that in terms of leader
focus variables we are likely to ®nd follower eects of heightened reverence, trust, and satisfac-
tion with the leader. In terms of follower focus variables, we suggest that charismatic leadership
will be positively related to followers' sense of collective identity, perceived group performance,
and feelings of empowerment. Accordingly, we examined these six outcome components. Five of
the six were chosen because of their consistent mention in the literature on charismatic and
transformational leadership (e.g., Bass, 1985; Bennis and Nanus, 1985; Conger and Kanungo,
1998; Shamir et al., 1993) as hypothesized eects. The sixth±heightened reverence for the
leaderÐwas selected because follower attraction is in large part based on perceptions of the
leader's extraordinary character and the leader's commitment to followers and the goals they
share (Bryman, 1992; Conger and Kanungo, 1998; Weber, 1947). One would assume that the
follower respect and awe engendered by these characteristics would produce a heightened rever-
ence for the leader. A more detailed rationale for the selection of the six outcome variables is
explained under the sections to follow.
Reverence for the leader

The sociologist Max Weber (1925/1968) asserted that the basis for follower attraction under
charismatic leadership was a perception that the leader was extraordinary: `[the charismatic
leader is] set apart from ordinary men and treated as endowed with supernatural, superhuman, or
at least . . . exceptional powers and qualities . . . [which] are not accessible to the ordinary person
but are regarded as of divine or as exemplary, and on the basis of them the individual concerned
is treated as a leader' (Weber, [1925] 1968, pp. 358±359). These perceptions of exceptional powers
are hypothesized to foster a high level of admiration and respect which lead to reverence for the
charismatic leader. Reverence is further reinforced by the leader's inspirational vision which
conveys the sense that his or her mission is extraordinary. During the implementation of the
vision, the charismatic leader is also seen as engaging in exemplary acts that subordinates
interpret as involving great personal risk and sacri®ce on the part of the leader. These acts in
turn heighten follower trust and satisfaction with their leader and deepen the esteem followers
hold for their leader and correspondingly their reverence for him or her. Thus, we propose the
following:
Hypothesis 1. Charismatic leadership behavior will be positively related to the followers' sense
of reverence for that manager.
Trust in the leader

The willingness of followers to be in¯uenced by the charismatic leader is in part based upon their
trust in the leader (Kouzes and Posner, 1987; Yukl, 1989). Given the often lofty visions of
charismatic leaders and the implementation challenges that these visions pose, follower trust is
essential to goal accomplishment and, as such, to the long-term viability of the mission. The
charismatic leader must therefore endeavour to foster deep levels of follower trust. In the only
empirical study speci®cally conducted on follower trust in the leader, Podsako et al. (1990) have
shown that three qualities of a leader contribute signi®cantly to follower trust. These include: (1)
identifying and articulating a vision; (2) setting an example for followers that is consistent with
the values the leader espouses; and (3) promoting group cooperation and the acceptance of group
goals. In addition, we would suggest that the charismatic leader also builds follower trust through
a demonstrated concern for follower needs, risk taking, personal sacri®ces, and unconventional
expertize.
Speci®cally, leaders will be perceived as trustworthy when they advocate their position in a
disinterested manner and demonstrate a concern for followers' needs rather than their own self-
interest (Walster et al., 1966). Charismatic leaders do so by making such qualities appear
extraordinary. They transform their concern for followers' needs into a total dedication and
commitment to a common cause they share, and they express these qualities in a disinterested and
sel¯ess manner. Moreover, charismatic leaders build trust in themselves by engaging in
exemplary acts that are perceived by followers as involving great personal risk, cost, and energy
(Friedland, 1964). In this case, personal risk might include the possible loss of personal ®nances,
the possibility of being ®red or demoted, and the potential loss of formal or informal status,
power, authority, and credibility. The higher the manifest personal cost or sacri®ce for common
goals that the charismatic leader demonstrates, the greater will be their perceived trustworthiness.
In essence, charismatic leaders strive to demonstrate that they are indefatigable workers prepared
to take on high personal risks or incur high personal costs in order to achieve their shared vision.
Through such personal examples, the charismatic leader reinforces perceptions that he or she is
worthy of complete trust. In addition, charismatic leaders appear to followers as knowledgeable
and expert in their areas of in¯uence. The attribution of charisma is generally in¯uenced by the
expertize of these leaders in two areas. First, charismatic leaders use their expertize in demon-
strating the inadequacy of the traditional technology, rules, and regulations of the status quo as a
means of achieving the shared vision (Weber, 1947). Second, charismatic leaders reveal expertize
in transcending the existing order through the use of unconventional or countercultural strategies
and plans of action (Conger, 1985, 1989). Such expertize further engenders trust by followers in
the charismatic leader. As such, we propose the following:
Hypothesis 2. Charismatic leadership behavior will be positively related to the followers' trust
in that manager.
Satisfaction with the leader

Current theory posits that charismatic leadership enhances follower satisfaction with the leader
through the leader's provision of meaningful goals, their exemplary behaviors, and their
empowerment approaches. For example, charismatic leaders oer their followers lofty and
inspiring visions which are highly meaningful to followers and contain signi®cant rewards for
them (Bass, 1985; Conger and Kanungo, 1987; Shamir et al., 1993). Since the leader is seen as the
primary source of the vision, he or she is attributed by followers as bringing meaning into their
lives and as providing them with goals that transcend their own limited existence (Jahoda, 1981)
and which permit them to achieve higher order rewards (Burns, 1978). This heightened sense of
meaningfulness and reward is greatly satisfying for followers. Leader behavior which is inter-
preted by followers as satisfying for them in turn increases follower satisfaction with the leader
him or herself (House and Dessler, 1974). Second, the leader's exemplary acts involving personal
risk and self-sacri®ce build follower satisfaction with the leader by heightening perceptions of the
leader's commitment to realizing their shared vision and the shared rewards that will accompany
the outcomes of the mission. The leader's demonstrated expertize and innovative insight also
build follower satisfaction with the leader by arming followers' perceptions that their personal
and material investments in the cause have a high probability of repeating expected rewards.
Third, through the active use of empowerment strategies (Conger and Kanungo, 1988; Spreitzer,
1995), charismatic leaders enhance the sense of self-ecacy of followers in relationship to task
accomplishment which in turn heightens satisfaction with the leader. Given these various forces,
we therefore propose the following:
Hypothesis 3. Charismatic leadership behavior will be positively related to the followers'
feeling of satisfaction with that manager.
Work-group collective identity

From the vantage point of follower focus variables, charismatic leadership is hypothesized to
transform the self-interests of followers into collective interests. They do so by tying participation
in the organization to a larger collective identity (Shamir et al., 1993). Charismatic leaders aim to
enhance the salience of collective identities in the self-concepts of followers primarily to increase
the probability that followers will engage in self-sacri®cial and cooperative behaviors that
advance the overall mission of the group rather than their own personal aims (Shamir et al.,
1993). As Meindl and Lerner (1983) suggest, a collective or shared identity can heighten the
`heroic motive' among followers and increase the likelihood that pursuits related to self-interests
among individual followers will be abandoned for more collective and altruistic endeavours.
Moreover, a collective identity places signi®cant social and psychological forces on followers
which increase their commitment to the values articulated by the charismatic leader for the
organization and likely heightens their commitment to task eorts (Kanter, 1967; Salancik, 1977;
Shamir et al., 1993. Therefore we propose the following:
Hypothesis 4. Charismatic leadership behavior of a manger will be positively related to the
followers' sense of a collective identity.
Perceived group task performance

Under charismatic leadership, the followers' expectations of successful group performance and
ongoing feelings of task accomplishment play a crucial role in successful outcomes. Returning to
the notion that the goals and visions of charismatic leaders are most often lofty and challenging,
we would expect followers to perceive themselves as a collective entity to possess a capacity to
successfully undertake and accomplish a high level of task eorts (eort-accomplishment
expectancies) (Conger and Kanungo, 1988; Shamir et al., 1993). Otherwise they would experience
the task demands of the charismatic leader with a sense of inadequacy and/or despair which
could lead to either ineectual attempts at task eorts or the abandonment of such eorts
(Conger and Kanungo, 1988). As Bandura (1977) has noted: `The strength of people's conviction
in their own eectiveness is likely to aect whether they would even try to cope with given
situations . . . Ecacy expectations determine how much eort people will expend and how long
they will persist ni the face of obstacles and aversive experiences' ( p. 193±194). A heightened
sense of group task ecacy allows charismatic leaders to mobilize their organizations in the face
of dicult challenges. It enables leaders to set higher performance goals and to have them
accepted. Finally, it helps in motivating subordinates to persist in eorts despite dicult
organizational and environmental obstacles. It follows that accomplishment of such dicult
tasks would lead to perceptions of successful task performance and to a sense of belonging to an
eective, high-performing team. Therefore, followers of charismatic leaders are likely to be highly
satis®ed with their performance as a group and should possess an enhanced sense of group task
performance. Thus.
Hypothesis 5. Charismatic leadership behavior of a manager will be positively related to the
followers' perceptions of group task performance.
Feelings of empowerment
Charismatic leadership is characterized by its emphasis on transforming the values, beliefs, and
attitudes of followers, typically in conjunction with a lofty, stimulating and idealized vision for
the future (Bass, 1985; Conger and Kanungo, 1987). According to Yukl (1989), the eect of such
transformation in¯uence is to empower subordinates to participate in the transformation of the
organization. Empowerment literature also recognizes the empowering eect of an idealized
vision. For example, Burke (1986) proposed that leaders empower by providing clarity of
direction, `but not just any directionÐa direction that encompasses a higher purpose, a worthy
cause, an idea . . . ( p. 69)'. According to Bennis and Nanus (1985), the feeling of making a
dierence for the organization is a critical component of empowerment.
More recent theoretical and empirical work on empowerment also emphasizes the empowering
eect of valued ideals and goals. The Thomas and Velthouse model of empowerment (Thomas
and Velthouse, 1990; Thomas and Tymon, 1994) and Spreitzer's work on empowerment
(Spreitzer, 1995; Spreitzer et al., 1997) both include the dimension of meaning as an essential
component of empowerment. Menon's model of empowerment (Menon, in press; Menon and
Borg, 1995) includes the dimension of goal internalization which captures the empowering
in¯uence of valued organizational goals. According to Kanungo and Mendonca (1996), leaders
formulate and articulate idealized future goals which serve to energize and hence empower
subordinates to the extent that these goals are internalized.
Besides articulating and championing an energizing cause, charismatic leaders also empower
by providing followers with information about their personal ecacy (Conger and Kanungo,
1988). The charismatic leader conveys such information through various means. These include:
(1) expressing con®dence in subordinates accompanied by high performance expectations
(Bandura, 1986; Burke, 1986; Conger, 1989; Conger and Kanungo, 1988; Neilsen, 1986; Shamir
et al., 1993); (2) providing autonomy from bureaucratic constraints (Conger and Kanungo, 1988;
Kanter, 1979); and (3) setting inspirational and highly meaningful goals (Bass, 1985; Conger and
Kanungo, 1988; Bennis and Nanus, 1985; Burke, 1986; McClelland, 1985). Conger and Kanungo
(1988) proposed that such enhanced feelings of self-ecacy are critical to the psychological
experience of empowerment. The empowerment research referred to above also recognizes a
sense of competence as being essential to feelings of empowerment (Bennis and Nanus, 1985;
Menon and Borg, 1995; Spreitzer, 1995; Thomas and Velthouse, 1990).
In the light of the above arguments, therefore.
Hypothesis 6. Charismatic leadership behavior of a manager will be positively related to the
followers' feeling of empowerment.
Methods
Sample and administration
Measures of charismatic leadership and follower eects were obtained from 252 managers in a
large, diversi®ed company. Structural equation modelling was then used to examine the direct
and indirect eects of the charismatic leadership behaviors on follower eects. The participants
in the study were managers attending training sessions in a large manufacturing conglomerate
based in the Northeast. Out of a total of 252 respondents, 94 per cent were male, 86 per cent were
married, and 80 per cent had at least a college degree. The mean age of the sample was 42.5 years
(S.D. 6.89). The mean organizational tenure was 13.62 years (S.D. 8.86) while the mean job
tenure was 2.9 years (S.D. 2.74).
The respondents were asked to answer a questionnaire assessing a supervisor's behavior
which was administered in two parts. Part I contained the charismatic leadership scale and
demographic items, while Part II contained measures to assess six other variables: namely
reverence, trust, satisfaction with leader, and collective identity, perceived group performance,
and empowerment. Part II of the questionnaire was administered 24 hours after the
administration of Part I to minimize same-source bias. Reliability data for all measures are
available in Table 1.
Measures
Charismatic leadership
Charismatic leadership behavior (CHRSMA) was measured using the 20-item Conger±Kanungo
charismatic leadership scale. In a series of empirical studies, the Conger±Kanungo measurement
scale has undergone testing and validation (Conger and Kanungo, 1992, 1994; Conger et al.,
1997). The results of these studies indicate that the Conger±Kanungo scale has sound psycho-
metric properties with adequate reliability and convergent and discriminant validity coecients.
The factor structure of the Conger±Kanungo scale also reveals that the perceived behavioral
components of charismatic leadership are stable. Finally, correlations of each of the Conger±
Kanungo charismatic leadership subscales with other perceived leadership behavior measures
indicate that the models and the scale as its operational measure are distinct from existing
leadership scales (Conger and Kanungo, 1994). The complete scale, which utilized a 6-point `very
characteristic' to `very uncharacteristic' response format, is outlined in Table 4.
Leader focus variables

Reverence for the leader (REV) was measured by a three-item scale developed for this study (see
Appendix 1 for scale items). Trust in the leader (TRU) was measured by a three-item scale. These
items were taken from Bass (1985) and Butler (1991). Satisfaction with the leader (SAT) was
measured by three items based on Bass (1985). A 6-point `strongly agree' to `strongly disagree'
response format was used for all the above items (see Appendix 1 for scale items).
Follower focus variables

Collective identity (CI) was measured by ®ve items speci®cally developed for this study on the
basis of face and content validity. The items re¯ect the sense of shared values, openness, and
broad consensus that characterize cohesive work groups. For the present managerial sample,
respondents were asked to refer to the work group, organizational unit or department that they
belonged to. The actual items are available in Appendix 1. Perceived group performance (GP)
was also measured by ®ve items developed for this study (see Appendix 1). It is important to note
that these items measure current perceptions of group performance with respect to ongoing
performance broadly de®ned rather than expectations of future performance with regard to
speci®c tasks. This approach is consistent with the cross-sectional nature of the study, the nature
of the sample (managers from dierent domains), and the arguments relating to Hypothesis 5.
Items were, therefore, constructed to re¯ect current beliefs about performance eectiveness.
Empowerment (EMP) was measured by Menon's 15-item subjective empowerment scale
(Menon, in press; Menon and Borg, 1995). This scale was preferred over Spreitzer's (1995) scale
because Menon's scale includes items that tap the dimension of goal internalization. We felt that
this is particularly well suited to study the eects of charismatic leadership given the empowering
nature of inspirational goals typically espoused by charismatic leaders. As can be seen from
Appendix 1, the items pertaining to dimensions traditionally associated with empowerment, like
perceived control and perceived competence, are similar to the Spreitzer (1995) scale, the major
dierence being the items related to goal internalization. These convergent and divergent
relationships were empirically demonstrated by Menon (1995) who also reported alpha
reliabilities of 0.87, 0.89, and 0.82 in three dierent samples. Menon (in press) provides evidence
of the construct validity of the scale. A 6-point `strongly agree' to `strongly disagree' response
format was used for all the follower focus measures.
Results
Table 1 shows the means, standard deviations, scale reliabilities, and bivariate correlations of all
variables. As can be seen, all measures have acceptable reliabilities. Structural equation
modelling using the LISREL 7 program was used to test the two sets of hypotheses. Since the
number of items in some of the scales were relatively large (e.g., 20 items in the charisma scale,
15 items in the empowerment scale), it is possible that the number of estimated parameters in a
given model might be too large for the available sample size. To minimize this possibility,
composite indicators were used to represent scales with more than three items (see Anderson and
Gerbing, 1988 for a discussion of this technique). For example, charisma was represented by
three indicators, formed by combining the 20 items through random assignment. First the 20
items were randomly assigned into three groupsÐtwo groups of seven items and one group of six
items. Individual items in each group were then averaged to obtain the three composite indicators
(see Shore et al. (1995) for a recent illustration of this technique).
To con®rm the distinctiveness of the charisma scale and the three leader focus variables, a
con®rmatory factor analysis (CFA) using LISREL 7 was conducted on the 12 items or indicators
measuring charismatic behavior, reverence, trust, and satisfaction with the leader. The
hypothesized four factor model had a chi-square (df 48) of 128.22, goodness-of-®t index
(GFI) of 0.914, and a root mean square residual (rmsr) of 0.039. In comparison, the single factor
model, with all 12 indicators loading on a single factor, had a chi-square (df 54) of 383.53, GFI
of 0.778, and rmsr of 0.051. The chi-square dierence was highly signi®cant, con®rming the
presence of four distinct factors. However, an examination of the correlations in Table 1 revealed
that `satisfaction' and `reverence' are highly correlated (r 0.75) raising the possibility that these
two measures may not be suciently distinct. To test this, the CFA model with four factors was
modi®ed by having the indicators of satisfaction and reverence load on a single factor. The
resulting three factor model had a chi-square (df 51) of 174.99, GFI of 0.884 and rmsr of 0.040.
Compared to the four factor model the chi-square dierence (df 3) was 46.77, which is highly
signi®cant. Thus, it was concluded that the four measures were suciently distinct.
Similarly, to con®rm the distinctiveness of the charisma scale and the three follower focus
variables, a con®rmatory factor analysis using LISREL 7 was conducted on the 12 composite
indicators measuring charismatic behavior, collective identity, empowerment, and group
performance. The hypothesized four factor model had a chi-square (df 48) of 127.63, GFI
of 0.920, and rmsr of 0.021. In comparison, the single factor model, with all 12 indicators loading
on a single factor, had a chi-square (df 54) of 579.72, GFI of 0.695, and rmsr of 0.045. The chi-
square dierence was highly signi®cant. Again, an examination of the correlations in Table 1
revealed that `perceived group performance' and èmpowerment' are highly correlated (r 0.64)
raising the possibility that these two measures may not be suciently distinct. To test this, the
CFA model with four factors was modi®ed by having the indicators of group performance and
empowerment load on a single factor. The resulting three factor model had a chi-square (df 51)
of 255.94, GFI of 0.838 and rmsr of 0.03. Compared to the four factor model the chi-square
dierence (df 3) was 128.31, which is highly signi®cant. Thus, it was concluded that these four
measures were also suciently distinct.
Hypotheses testing
Hypotheses 1 to 6 were tested using the hypothesis testing procedure recommended by Joreskog
and Sorbom (1989, pp. 216±221). First, to test the hypotheses involving charisma and the leader
Table 1. Descriptive statistics and intercorrelations

Variable Mean S.D. Alpha 1 2 3
1. Charisma 4.50 0.50 0.82
2. Reverence 5.13 0.74 0.85 0.50{
3. Trust 3.94 0.48 0.72 0.20{ 0.39{
4. Satisfaction 5.02 0.75 0.81 0.45{ 0.75{ 0.28{
5. Collective Identity 4.71 0.61 0.78 0.34{ 0.41{ 0.19{
6. Group Performance 4.83 0.61 0.85 0.35{ 0.42{ 0.16*
7. Empowerment 5.29 0.46 0.91 0.31{ 0.43{ 0.15*
* p 5 0.05; { p 5 0.01; { p 5 0.001.
focus variables, a saturated structural model with uncorrelated errors and disturbances (Medsker
et al., 1994) was set up with charisma (CHRSMA) as the exogenous latent variable. This model
which allows for all possible unidirectional relationships between the variables charisma
(CHRSMA), reverence (REV), trust (TRU), and satisfaction with the leader (LSAT), had a
chi-square (df 48) of 128.22, with a GFI of 0.914 and rmsr of 0.039.
To test the various hypotheses, the following strategy was adopted (see Joreskog and Sorbom,
1989). First, the coecient corresponding to the hypothesis in question was set to zero. The
resulting model was then compared to the saturated model using a single degree of freedom chi-
square dierence test to see if the two models are signi®cantly dierent. For example, to test
Hypotheses 1, the coecient g11 representing the direct relationship between charisma and
reverence (see Figure 1) was set to zero. The resulting model had a chi-square (df 49) of 197.38,
with a GFI of 0.885 and rmsr of 0.109. Compared to the saturated model, the chi-square
dierence (df 1) is 69.16, which is highly signi®cant. This indicates that the coecient g11 is
signi®cantly dierent from zero. From the sign and magnitude of this coecient in the saturated
model we can then conclude that Hypothesis 1, which predicts a positive relationship between
charismatic behavior and reverence is strongly supported. The complete results of the hypothesis
testing is available in Table 2.
As can be seen from Table 2, Hypotheses 2 and 3 are not supported since the reduced models
obtained by setting the appropriate coecients to zero are not signi®cantly dierent from the
saturated model. Thus, in the present data set there is no direct relationship between charismatic
behavior and trust or satisfaction. Rather, the relationships between charismatic behavior and
these two variables respectively seem to be completely mediated by reverence, in this four variable
model. We also see from the saturated model that the relationship between trust and satisfaction
is non-signi®cant. As can be seen from Table 2, setting the corresponding coecient to zero does
not result in any signi®cant chi-square dierence compared to the saturated model. Thus, the
`true model' or the model that best represents this particular data set is as shown in Figure 1.
A similar procedure was used to test Hypotheses 4 to 6 involving follower focus variables.
First, a saturated structural model with all possible unidirectional relationships between charisma
(CHRSMA) and the variables group collective identity (COH), perceived group performance
(GP), and empowerment (EMP), was set up. This model had a chi-square (df 48) of 127.63,
with GFI of 0.920 and rmsr of 0.021. Hypotheses 4 to 6 were then individually tested by setting
the appropriate coecient of the direct path from CHRSMA to each of the other variables to
zero, one variable at a time. The results of this analysis is shown in Table 3.
As can be seen from Table 3, the chi-square dierence test reveals that the coecients g11 and
g21 , corresponding to Hypotheses 4 and 5 respectively, are signi®cantly dierent from zero.
Therefore, Hypotheses 4 and 5 are supported. On the other hand, since coecient g31 is not
Table 2. Results of hypothesis testing: leader focus variables*

Model/constraints g11 g21 g31 b21 b31 b32 w2 df GFI rmsr w2diff
1. Saturated structural 0.39{ ÿ0.00 ÿ0.04 0.69{ 1.30{ ÿ0.12 128.22 48 0.914 0.039 ±
model
2. g11 0 ± 0.05 0.06 0.63{ 1.16{ ÿ0.06 197.38 49 0.885 0.109 69.16
3. g21 0 0.39{ ± ÿ0.04 0.68{ 1.29{ ÿ0.11 128.23 49 0.914 0.039 0.01{
4. g31 0 0.38{ 0.00 ± 0.68{ 1.23{ ÿ0.09 128.68 49 0.914 0.038 0.48{
5. `True Model' g21 , g31 , 0.38{ ± ± 0.68{ 1.16{ ± 128.88 51 0.913 0.038 0.66{
b32 0
* Corresponding model is depicted in Figure 1; { p 5 0.001; { non-signi®cant.
Table 3. Results of hypothesis testing: follower focus variables*

Model/constraints g11 g21 g31 b21 b31 b32 w2 df GFI rmsr w2diff
1. Saturated structural 0.23{ 0.10{ 0.01 0.64{ 0.29{ 0.35{ 127.63 48 0.920 0.021 ±
model
2. g11 0 ± 0.13{ 0.02 0.64{ 0.28{ 0.35{ 155.74 49 0.906 0.049 28.11
3. g21 0 0.25{ ± 0.01 0.74{ 0.30{ 0.33{ 133.60 49 0.917 0.024 5.97
4. g31 0 0.23{ 0.10{ ± 0.64{ 0.29{ 0.36{ 127.85 49 0.919 0.021 0.22}
* Corresponding model is depicted in Figure 2.
{ p 5 0.05; {p 5 0.001; } non-signi®cant.
Figure 1. Structural model: charisma and leader focus variables (***p 5 0.001. Note: The coecients
correspond to the model with g21 , g31 and b32 constrained to zero)
signi®cantly dierent from zero, Hypothesis 6 is not supported. Given that all the other
coecients are signi®cant in the saturated model, the `true model' or the model that best
represents this data set is depicted in Figure 2. Charismatic leader behavior is directly related to
group collective identity and perceptions of group performance. These two variables in turn
completely mediate the relationship between charismatic behavior and subjective empowerment
in this four variable model.
Thus, charismatic leader behavior seems to directly generate in followers a feeling of reverence,
a sense of group collective identity, and perceptions of group task performance. In the next
section, we shall investigate the causal relationship between the individual behavior components
of charismatic leadership behaviors and speci®c follower eects.
Figure 2. Structural model: charisma and follower focus variables (***p 5 0.001. Note: The coecients
correspond to the model with g31 constrained to zero)
Sub-scale analysis
The Conger±Kanungo scale of charismatic leadership has ®ve sub-scales: strategic vision and
articulation (SVA), sensitivity to the environment (SE), sensitivity to member needs (SMN),
personal risk (PR), and unconventional behavior (UB) (See Conger et al. (1997), for details of
validation studies). To con®rm the existence of these ®ve sub-scales in the present data set, the
20 item Conger±Kanungo scale was subjected to a principal component factor analysis using
varimax rotation with a ®ve factor solution. Table 4 shows the results of the factor analysis. With
a dierent database, the pattern of factor loadings in Table 4 exactly replicates the pattern of
factor loadings reported in Conger et al. (1997)Ði.e., the items load on the ®ve factors exactly as
expected. This was con®rmed by the results of a con®rmatory factor analysis. Compared to a
single factor model [Chi-square (df 170) 979.62; GFI 0.662; rmsr 0.661], the ®ve
factor model [Chi-square (df 160) 348.17; GFI 0.892; rmsr 0.062] provided superior
®t, the chi-square dierence between the two models being highly signi®cant. An examination of
the intercorrelations among the sub-scales (in Table 5) did not reveal unusually high correlations
between any given pair of subs-scales, obviating the need to test the hypothesized ®ve factor
model against potential four or three factor models.
Table 5 shows the means, standard deviations, scale reliabilities, and intercorrelations among
the sub-scales. Bivariate correlations between the sub-scales and the six outcome variables are
also provided in Table 5. The relationships between the outcome variables and each of the ®ve
sub-scales was simultaneously estimated using the LISREL 7 program, with the ®ve sub-scales as
independent variables and the six outcome variables as dependent variables. The results of the
regression analysis are shown in Table 6.
As can be seen from Table 6, the outcome variable reverence is most strongly related to the
sub-scale sensitivity to the environment. Three other sub-scales strategic vision and articulation,
sensitivity to member needs, and personal risk also contribute to feelings of reverence. Trust is
signi®cantly related only to the sub-scale sensitivity to the environment. Satisfaction with the
leader is strongly related to the sub-scales strategic vision and articulation, sensitivity to the
Table 4. Factor loadings for C±K scale items

Item Factor 1 Factor 2 Factor 3 Factor 4 Factor 5
Strategic vision and articulation (SVA)
Has vision; often brings up ideas about possibilities for the 0.79 0.19 0.15 ÿ0.08 0.08
future
Provides inspiring strategic and organizational goals 0.74 ÿ0.01 0.26 ÿ0.04 ÿ0.05
Consistently generates new ideas for the future of the 0.66 0.35 0.05 0.02 0.07
organization
Entrepreneurial; seizes new opportunities in order to 0.65 0.10 0.04 0.00 0.20
achieve goals
Readily recognizes new environmental opportunities 0.58 0.12 0.37 0.09 0.12
( favourable physical and social conditions) that may
facilitate achievement or organizational objectives
Inspirational; able to motivate by articulating eectively 0.58 0.16 0.26 0.20 0.01
the importance of what organizational members are doing
Exciting public speaker 0.54 0.02 ÿ0.16 0.28 0.13
Personal risk (PR)
In pursuing organizational objectives, engages in activities 0.10 0.86 0.05 0.07 0.20
involving considerable personal risk
Takes high personal risks for the sake of the organization 0.23 0.81 0.04 0.10 0.18
Often incurs high personal cost for the good of the 0.19 0.78 0.08 0.12 0.09
organization
Sensitivity to the environment (SE)
Readily recognizes constraints in the physical environment 0.09 0.09 0.74 0.00 ÿ0.08
(technological limitations, lack of resources, etc.) that may
stand in the way of achieving organizational objectives
Readily recognizes constraints in the organization's social 0.07 0.14 0.71 0.14 ÿ0.03
and cultural environment (cultural norms, lack of grass
roots support, etc.) that may stand in the way of achieving
organizational objectives
Recognizes the limitations of other members of the 0.18 ÿ0.11 0.60 0.24 0.08
organization
Recognizes the abilities and skills of other members of the 0.35 0.01 0.50 0.22 ÿ0.01
organization
Sensitivity to member needs (SMN)
Shows sensitivity for the needs and feelings of the other 0.06 0.11 0.25 0.78 ÿ0.17
members in the organization
In¯uences others by developing mutual liking and respect 0.09 ÿ0.08 0.04 0.78 ÿ0.05
Often expresses personal concern for the needs and 0.02 0.19 0.26 0.76 ÿ0.09
feelings of other members in the organization
Unconventional behavior (UB)
Engages in unconventional behavior in order to achieve 0.02 0.06 ÿ0.11 ÿ0.04 0.80
organizational goals
Use non-traditional means to achieve organizational goals 0.25 0.17 0.05 ÿ0.10 0.79
Often exhibits very unique behavior that surprises other 0.12 0.30 0.03 ÿ0.16 0.61
members of the organization
Table 5. C±K sub-scales: descriptive statistics and intercorrelations

Variable Mean S.D. alpha 1 2 3 4 5
C±K Sub-scales
1. SVA 4.8 0.63 0.81
2. SE 4.76 0.64 0.64 0.42{
3. SMN 4.51 0.91 0.77 0.21{ 0.40{
4. PR 4.15 1.08 0.84 0.40{ 0.18{ 0.16*
5. UB 3.83 1.00 0.71 0.28{ ÿ0.04 ÿ0.20{ 0.39{
Outcome variables
REV 5.13 0.74 0.85 0.39{ 0.49{ 0.36{ 0.29{ 0.05
TRU 3.94 0.48 0.72 0.14* 0.21{ 0.12 0.08 0.07
SAT 5.02 0.75 0.81 0.39{ 0.44{ 0.44{ 0.16* ÿ0.02
CI 4.71 0.61 0.78 0.31{ 0.32{ 0.32{ 0.12 0.01
GP 4.83 0.61 0.85 0.31{ 0.38{ 0.21{ 0.13* 0.08
EMP 5.29 0.46 0.91 0.32{ 0.36{ 0.22{ 0.07 0.03
* p 5 0.05; { p 5 0.01; { p 5 0.001.
Note. SVA, strategic vision and articulation; SE, sensitivity to the environment; SMN, sensitivity to member needs; PR,
personal risk; UB, unconventional behavior; REV, reverence; TRU, trust; SAT, satisfaction; CI, collective identity; GP,
group performance; EMP, empowerment.
environment, and sensitivity to member needs. Collective identity is related strongly to both
strategic vision and articulation and sensitivity to member needs. Perceived group performance,
on the other hand, is most strongly related to sensitivity to the environment. Empowerment is
signi®cantly related to both strategic vision and articulation and sensitivity to the environment.
None of the dependent variables were related to the leader's unconventional behavior (UB). The
eects of unconventional behavior of superiors on subordinates need further exploration in
future studies. Perhaps the main eect of unconventional behavior lies in drawing follower's
attention to the leader's vision and implementation strategies.
Discussion
This study identi®ed six follower eects from the charismatic leadership literature and tested
whether the behavioral components of the Conger±Kanungo model of charismatic leadership
produced these predicted eects. The ®ndings indicate that three of the components have a strong
direct relationship with charismatic leadership: (1) leader reverence, (2) follower collective
identity, and (3) follower perceptions of group task performance. The remaining three
components all had mediated or indirect relationships with charismatic leadership.
One of the important ®ndings of the study was the fact that indeed the followers of charismatic
leaders develop a reverence for their leader and that this appears to be based most strongly upon
their perceptions of the leader's sensitivity to the environment. Secondarily, the leaders' abilities
at formulating and articulating an inspiring vision and their sensitivity to member needs play a
role. Presumably followers see the leader's environmental sensitivity and visioning abilities as
exemplary skills in their leaders which are deserving of admiration. This outcome may be a
byproduct of the managerial composition of the sample. Many managers have a strategic and/or
marketplace responsibility and in turn may be more attuned to the need for their leader to be
highly sensitive to the larger environment. Non-managers may be less knowledgeable about or
impressed by the leader's sensitivity, and therefore their reverence for the leader may be
Table 6. LISREL (gamma) estimates of relationships between charisma sub-scales and outcome variables
Independent variables Dependent variables
Reverence Trust Satisfaction Collective Group Empowerment

(REV) (TRU) (SAT) identity performance (EMP)
(CI) (GP)
Strategic vision and 0.21{ 0.03 0.31{ 0.22{ 0.17* 0.16{
articulation (SVA)
Sensitivity to the 0.38{ 0.15* 0.26{ 0.12 0.25{ 0.18{
environment (SE)
Sensitivity to member 0.13{ 0.02 0.26{ 0.15{ 0.06 0.04
needs (SMN)
Personal risk (PR) 0.10* ÿ0.00 ÿ0.02 ÿ0.03 ÿ0.03 ÿ0.04
Unconventional ÿ0.02 0.03 ÿ0.03 0.00 0.05 0.02
behavior (UB)
*p 5 0.05; { p 5 0.01; { p 5 0.001.
in¯uenced by other factors. On the other hand, the leader's sensitivity to member needs re¯ects a
caring concern and respect for followers which most likely engenders reciprocal respect from
followers. These ®ndings would appear to support assertions by Weber (1925/1968) that one of
the foundations of charismatic leadership is a `devotion to the speci®c and . . . exemplary
character of an individual person' (Weber, 1925/1968, p. 46)Ðperceptions of the èxemplary
character' being shaped by speci®c leadership behaviors and competencies. What is also
intriguing is the link to prophecy or foresight which is implied in the strong relation between
reverence and the subscale of environmental sensitivity. Early Biblical referencesÐspeci®cally
two letters of St. Paul in Romans, 12, and 1 Corinthians, 12Ðuse the term of charisma to
describe gifts from God speci®cally related to prophecy.
Leader reverence is generally an unexplored topic in the leadership research literature and is
deserving of signi®cant future attention. There is a need to develop a more comprehensive
conceptualization of the construct. Our own measure of reverence was built around only three
itemsÐrespect, esteem, and admiration. There may be other dimensions of reverence itself. In
addition, there may be other leadership behaviors which were not measured in this study that
induce follower feelings of reverence. Future research is needed to help us broaden our
understanding of how followers come to revere their leader and to determine whether this quality
of follower reverence is an attribute that eectively distinguishes charismatic leadership from
other forms.
A somewhat surprising ®nding was that trust and satisfaction with the leader were mediated by
reverence rather than being directly related to charismatic leadership. There has been little
emphasis in the leadership literature on variables that intervene between leader behaviors and
follower eects (Kirkpatrick and Locke, 1996). A mediating variable (Kenny, 1979) comes
between a dependent and an independent variable, where the independent variable in essence
causes the intervening variable which causes, in turn, the dependent variable ( p. 4). In this case,
the mediating relationship is a two-part causal sequence in which the independent variable
signi®cantly aects the intervening variable, which, in turn, aects signi®cantly the dependent
variable, but the independent variable does not directly aect the dependent variable (Baron and
Kenny, 1986). In this case, reverence as the mediating variable aects follower trust and
satisfaction in the leader under charismatic leadership. We might speculate that trust and
satisfaction are powerfully shaped by a cognitive (leaders held in high esteem) and an aective
state (aectionate respect) felt by followers in their relations with the charismatic leader rather
than simply by the behavioral components of charisma per se. One earlier study has shown that
the leader's vision was related to greater trust in the leader (Kirkpatrick and Locke, 1996).
Another study has also shown that the core components of vision, appropriate role-modelling,
the promotion of group cooperation, and the provision of individualized support signi®cantly
in¯uenced follower trust in the leader (Podsako et al., 1990). However, both the studies did not
investigate the leader reverence variable as a distinct follower eect. Hence these studies could not
identify the mediational role of reverence variable in leader behaviors and trust relationships. It is
natural to expect that followers would trust or have reliance on the leader only when they
recognize superior ability or quality in the leader which they adore. It is adoration not just liking
that creates trust in the leader.
Collective identity, as it turns out, is related to charismatic leadership especially through the
subscales of vision and its articulation and sensitivity to member needs. This ®nding con®rms
earlier theory which hypothesized that charismatic leaders employ their vision not only to set
goals but to shape a collective identity for followers (Meindl and Lerner, 1983; Shamir et al.,
1993). In addition, these leaders are modelling the importance of cooperative behavior through
their own concern for others and by direct actions promoting mutual liking and respect among
followers.
Perceived group task performance is also directly in¯uenced by charismatic leadership largely
through the leader's sensitivity to the environment and somewhat less so through their
formulation of a strategic vision and its articulation. We might speculate that followers perceive
that the leader's strong sensitivity to the external and internal environment ensures that more
realistic assessments are made of what tasks can actually be accomplished collectively by
organizational members. As a result, we would expect to see followers describing perceptions of
their collective ecacy, as re¯ected in assessments of past performance, to be positive.
Finally, we do not ®nd it surprising that feelings of empowerment are mediated through
collective identity and perceived group performance. While the strategic vision espoused by the
charismatic leader may be personally appealing to the subordinate, it is the acceptance and
endorsement of the leader's ideas by the reference group that provides social con®rmation of the
appropriateness and worthiness of the vision. The leader's empowering message is also typically
phrased in terms of achieving a dicult or lofty goal through collective eort. Goal internal-
ization is thus facilitated by the sense of shared values. Collective identity may also provide
mutual support which in turn provides a source of self-ecacy information that is empowering.
Speci®cally, we would propose that group cooperation oers support in the form of at least three
sources of self-ecacy information, as identi®ed by Bandura (1986)Ðvicarious experience,
verbal persuasion, and emotional arousal. In any group, there are varying levels of competence.
Those members who are more competent may serve as vicarious models for others. Less
competent co-workers may come to believe that if they act in a like manner they too will be able to
achieve improvements in their performance. Secondly, a highly cooperative group may, with
greater frequency, use words of encouragement with one another to empower co-workers. Finally,
a group with a strong collective identity has a higher probability of creating an emotionally
supportive and trusting group atmosphere which can be eective in strengthening self-ecacy
beliefs of its individual members (Neilsen, 1986). In addition, followers' sense of successful group
task performance strengthens a collective sense of task competence, control over the work
environment and commitment to task goals.
This investigation has several important limitations. In this study, unlike in a purely experi-
mental study, the respondents knew and/or had close organizational contact of an ongoing
nature with the managers whom they rated for charismatic leadership behavior. The strong
association between charismatic behavior and reverence might be partly due to this close
interaction. The present research design however cannot rule out the possibility of reverse/
reciprocal causality, i.e., it is the reverence that can partially account for the ratings on
charismatic behavior.
Another limitation of this study is the nature of the sample. All the respondents in the study
were managers. Charismatic leaders are typically thought to in¯uence personnel at all levels of
the organization, including the rank and ®le workers who are crucial for the actual imple-
mentation of the leader's vision. Thus, the results of this study need to be replicated with a more
representative cross-sectional organizational sample.
It should also be recognized that the grouping of the dependent variables into leader focus and
follower focus variables was primarily intended to facilitate the interpretation of emerging
relationships between charismatic behavior and the dependent variables; as can be recalled, there
were no speci®c hypotheses linking the various dependent variables to each other. It is quite
possible that a dierent grouping based on a dierent logic or no grouping at all would have
produced a dierent pattern of results. For instance, charismatic behavior could be linked to
empowerment (classi®ed as a follower focus variable in this study) through trust (classi®ed as a
leader focus variable).1 This issue can only be resolved by future research with a priori hypotheses
about direct and mediational links between charismatic behavior and dependent variables of
interest.
In this study, ratings of charismatic behavior as well as the follower outcome variables were
obtained from the same subordinate. This can lead to possible same source bias accounting for
the associations between variables. On the other hand, the charismatic behavior ratings and the
follower outcome ratings were obtained using two dierent questionnaires, with a 24-hour hiatus
between administrations. This procedure would help reduce possible same-source bias (Avolio
et al., 1991; Podsako and Organ, 1986). The data analysis results also provide some indication
about the extent of common method bias. In the structural model relating charisma with the
leader focus variables (see Table 2), the direct relationships between charisma and the dependent
variables trust and satisfaction were statistically non-signi®cant. Similarly, in the structural
model with follower focus variables, the direct relationship between charisma and empowerment
was statistically non-signi®cant (see Table 3). The statistical non-signi®cance of some of the
direct relationships between charismatic behavior and the dependent variables is not consistent
with extensive method bias eects. Nevertheless, as a possible test of method bias eects, a
statistical test was conducted using the procedure recently demonstrated by Markel and Frone
(1998). The two models in Figures 1 and 2 were re-estimated with all the indicator variables
loading on a general method factor. In the case of the leader focus variables model (Figure 1),
including the general method factor resulted in a chi-square (df 39) of 71.56 ( p 0.001), GFI
of 0.953 and rmsr of 0.026. As can be seen from Table 2, this represents an improved ®t.
However, none of the individual path coecients corresponding to relationships between the
indicators and the general method factor was signi®cant. The beta and gamma estimates (see
Figure 1) were similar to those obtained earlier. This pattern of results was also obtained for the
structural model involving follower focus variables. Thus, while the ®t of both models improved,
indicating the presence of a common latent factor, the parameter estimates reveal that the earlier
results and conclusions are still valid. This ®nding is consistent with other research on common
method bias which has concluded that while method bias may be present, it may not always
signi®cantly aect results or conclusions (Crampton and Wagner, 1994; Doty and Glick, 1998;
Spector and Brannick, 1995). Nevertheless, the presence of method bias, as suggested by the
1
The authors are grateful to an anonymous reviewer for raising this insight.
above statistical test as well as the magnitude of some of the correlations among the dependent
variables, is an inherent limitation of the research design used in this study. In future investiga-
tions, splitting the measures of independent and dependent variables while concurrently using
dierent raters of the same leader will better address this potential problem.
It is important that future research also examine additional follower eects. For instance,
theoretical literature on charismatic leadership suggests that often followers develop two types
of feeling under the in¯uence of the leader: (1) a sense of dependence on the leader, and (2)
a sense of empowerment. Only the latter type of feeling among subordinates was investigated
in this study. The conditions under which a dependence relationship with the leader
develops needs to be studied, and such conditions need to be contrasted with conditions
promoting feeling of empowerment among followers. Furthermore, other follower eects such as
organizational and task commitments as a result of charismatic in¯uence need to be explored.
Since the vision of a charismatic leader forms the core of the organizational mission, it is expected
that followers who internalize the vision would exhibit a higher degree of organizational and task
commitment.
Finally, one of the principal contributions of the Conger±Kanungo model of charismatic
leadership is the positioning of followers' attributions about the leader as mediating variables
between the leader's behaviors and other eects such as ecacy and trust. This focal aspect of the
model was not directly tested by the present study. Our results regarding the mediating eect of
reverence for the leader on certain follower eects may, however, be seen as a partial test of the
Conger±Kanungo attribution hypothesis if we assume that reverence for the leader is closely akin
to the attribution of charisma to the leader. Nonetheless, future investigations will need to
include explicit measurement of followers' attributions about the leadership unique qualities and
the leader's motivations to test this important dimension of the model properly.
References
Agle BR, Sonnenfeld JA. 1994. Charismatic chief executive ocers: are they more eective? An empirical
test of charismatic leadership theory. Academy of Management Proceedings, Dallas, 2±6.
Anderson JC, Gerbing DW. 1988. Structural equation modeling in practice: a review and recommended
two-step approach. Psychological Bulletin, 103: 411±423.
Avolio BJ, Yammarino FJ, Bass BM. 1991. Identifying common methods variance with data collected from
a single source: an unresolved sticky issue. Journal of Management, 17(3): 571±587.
Bandura A. 1977. Self-ecacy: toward a unifying theory of behavioral change. Psychological Review, 84:
191±215.
Bandura A. 1986. Social Foundations of Thought and Action: A Social-cognitive View. Prentice-Hall:
Englewood Clis, NJ.
Bass BM. 1985. Leadership and Performance Beyond Expectations. Free Press: New York.
Bass BM, Avolio B. 1993. Transformational leadership: a response to critiques In Leadership theory and
research: Perspectives and directions, Chemers MM, Ayman R (eds). Academic Press: New York; 49±80.
Bass BM, Waldman DA, Avolio BJ, Bebb M. 1987. Transformational leadership and the falling dominoes
eect. Group and Organization Studies, 12: 73±87.
Bass BM, Yammarino FJ. 1988. Leadership: Dispositional and situational. (ONR Tech Rep No 1) State
University of New York, Center for Leadership Studies: Binghamton, NY.
Bennis W, Nanus B. 1985. Leaders: The Strategies of Taking Charge. Harper & Row: New York.
Berlew DE. 1974. Leadership and Organizational excitement. California Management Review, 17: 21±30.
Burke WW. 1986. Leadership as empowering others. In Executive Power, Srivastva S (ed.). Jossey-Bass:
San Francisco; 51±77.
Burns JM. 1978. Leadership. Harper & Row: New York.
Butler JK. 1991. Toward understanding and measuring conditions of trust: evolution of a condition of trust
inventory. Journal of Management, 17(3): 643±663.
Conger JA. 1985. Charismatic leadership in business: A exploration study Unpublished doctoral
dissertation, School of Business Administration, Harvard University.
Conger JA. 1989. The Charismatic Leader. Jossey-Bass: San Francisco.
Conger JA, Kanungo RN. 1987. Toward a behavioral theory of charismatic leadership in organizational
settings. Academy of Management Review, 12: 637±647.
Conger JA, Kanungo RN. 1988. The empowerment process: integrating theory and practice. Academy of
Management Review, 13(3): 471±482.
Conger JA, Kanungo RN. 1992. Perceived behavioral attributes of charismatic leadership. Canadian
Journal of Behavioral Science, 24: 86±102.
Conger JA, Kanungo RN. 1994. Charismatic leadership in organizations: Perceived behavioral attributes
and their measurement. Journal of Organizational Behavior, 15: 439±452.
Conger JA, Kanungo RN. 1998. Charismatic Leadership in Organizations. Sage Publishers: Thousand
Oaks, CA.
Conger JA, Kanungo RN, Menon ST, Mathur P. 1997. Measuring charisma: dimensionality and validity of
the Conger±Kanungo scale of charismatic leadership. Canadian Journal of Administrative Sciences, 14(3):
290±302.
Crampton SM, Wagner JA III. 1994. Percept±percept in¯ation in micro-organizational research: an
investigation of prevalence and eect. Journal of Applied Psychology, 79(1): 67±76.
Doty DH, Glick WH. 1998. Common method bias: does common methods variance really bias results?
Organizational Research Methods, 1(4): 374±406.
Friedland WH. 1964. For a sociological concept of charisma Social Forces, 43: 18±26.
House RJ. 1977. A 1976 theory of charismatic leadership. In Leadership: The cutting edge, Hunt JG,
Larson LL (eds). Southern Illinois University Press: Carbondale, IL; 189±207.
House RJ, Spangler WD, Woycke J. 1991. Personality and charisma in the U.S. presidency: A
psychological theory of leader eectiveness. Administrative Science Quarterly, 36: 364±396.
House RJ, Dessler G. 1974. The path-goal theory of leadership: some post hoc and a priori tests In
Contingency approaches to leadership, Hunt JG, Larson LL (eds). Southern Illinois University Press:
Carbondale, IL; 29±55.
Howell JM, Frost PJ. 1989. A laboratory study of charismatic leadership. Organizational Behavior and
Human Decision Processes, 43: 243±269.
Jahoda M. 1981. Work employment and unemployment: values, theories and approaches in social research.
American Psychologists, 36: 184±191.
Joreskog KG, Sorbom D. 1989. LISREL 7: A Guide to the Program and Applications, 2nd edn. SPSS Inc.:
Chicago, IL.
Kanter RM. 1967. Commitment and social organization: A study of commitment mechanisms in utopian
communities. American Sociological Review, 33(4): 499±517.
Kanter RM. 1979. Power failures in management circuits. Harvard Business Review, 57: 65±75.
Kanungo RN, Mendonca M. 1996. Ethical Dimensions of Leadership. Sage: Thousand Oaks, CA.
Katz J, Kahn RL. 1978. The Social Psychology of Organizations. Wiley: New York.
Kirkpatrick SA, Locke EA. 1996. Direct and indirect eects of three core charismatic leadership
components on performance and attitudes. Journal of Applied Psychology, 81: 36±51.
Kouzes JM, Posner BZ. 1987. The Leadership Challenge. Jossey-Bassy: San Francisco.
Locke E, Kirkpatrick S, Wheeler JK, Schneider J, Niles K, Goldstein H, Welsh K, Chau D-Ok. 1991. The
Essence of Leadership. Lexington Books: New York.
Markel KS, Frone MR. 1998. Job characteristics, work±school con¯ict, and school outcomes among
adolescents: testing a structural model. Journal of Applied Psychology, 83(2): 277±287.
McClelland DC. 1985. Human Motivation. Scott Foresman: Glenview, IL.
Medsker GJ, Williams LJ, Holahan PJ. 1994. A review of current practices for evaluating causal models in
organizational behavior and human resources management research. Journal of Management, 20(2):
439±464.
Meindl J, Lerner MJ. 1983. The heroic motive: some experimental demonstrations. Journal of Experimental
Social Psychology, 19: 1±20.
Menon ST. 1999. Psychological empowerment: de®nition, measurement, and validation. Canadian Journal
of Behavioural Science, 31(3): 161±164
Menon ST, Borg I. 1995. Facets of subjective empowerment. In Facet Analysis and Design, Hox J,
Swanborn P (eds). Zeist Press: Zeist, The Netherlands; 129±140.
Neilson E. 1986. Empowerment strategies: Balancing authority and responsibility. In Executive Power,
Srivastrax S (ed.). Jossey-Bass: San Francisco; 78±110.
Podsako PM, Organ DW. 1986. Self-reports in organizational research: problems and prospects. Journal
of Management, 12: 531±544.
Podsako PM, MacKenzie SB, Moorman RH, Fetter R. 1990. Transformational leader behaviors and
their eects on followers' trust in leader, satisfaction, and organizational citizenship behaviors. The
Leadership Quarterly, 1(2): 107±142.
Puer SM. 1990. Attributions of charasmatic leadership: The impact of decision style, outcome
characteristics. Leadership Quarterly, 1: 177±192.
Salancik GR. 1977. Commitment and the control of organizational behavior and belief. In New Directions
in Organizational Behavior, Staw BM, Salancik GR (eds). St Clair: Chicago; 1±54.
Shamir B. 1992. Attribution of in¯uence and charisma to the leader: The romance of leadership revisited.
Journal of Applied Social Psychology, 22(5): 386±407.
Shamir B, House R, Arthur MB. 1993. The motivational eects of charismatic leadership: a self-concept
based theory. Organizational Science, 4(4): 577±594.
Shamir B, Zakay E, Breinin E, Popper M. 1998. Correlates of charismatic leader behaviors in military
units: subordinates' attitudes, unit characteristics, and superiors' appraisals of leader performance.
Academy of Management Journal, 41(4): 387±409.
Shore LF, Barksdale K, Shore TH. 1995. Managerial perceptions of employee commitment to the
organization. Academy of Management Journal, 38(6): 1593±1615.
Spector PE, Brannick MT. 1995. The nature and eects of method variance in organizational research.
International Review of Industrial and Organizational Psychology, 10: 249±274.
Spreitzer GM. 1995. Individual empowerment in the workplace: dimensions, measurement, validation.
Academy of Management Journal, 38(5): 1442±1465.
Spreitzer GM, Kizilos MA, Nason SW. 1997. A dimensional analysis of the relationship between
psychological empowerment and eectiveness, satisfaction, and strain. Journal of Management, 23(5):
679±704.
Thomas KW, Tymon WG. 1994. Does empowerment always work: understanding the role of intrinsic
motivation and personal interpretation. Journal of Management Systems, 6(2): 1±13.
Thomas KW, Velthouse BA. 1990. Cognitive elements of empowerment: an ìnterpretive' model of intrinsic
task motivation. Academy of Management Review, 15: 666±681.
Walster E, Aronson D, Abrahams D. 1966. On increasing the persuasiveness of a low prestige
communicator. Journal of Experimental Social Psychology, 2: 325±342.
Weber M. 1925/1968. Economy and Society, 3 vols, Guenther R, Wittich C (eds). Bedminster: New York.
Weber M. 1947. The Theory of Social and Economic Organizations, Henderson AM, Parsons T (eds). Free
Press: New York.
Yukl GA. 1989. Managerial leadership: a review of theory and research. Journal of Management, 15:
251±289.
Zaleznik A, Kets de Vries M. 1975. Power and the Corporate Mind. Houghton Miin: Boston.
Appendix 1: Measures
Reverence
1. I hold him/her in high respect.
2. I have great esteem for him/her.
3. I admire him/her as a leader.
Trust
1. I have complete faith in him/her (Bass, 1985).
2. Sometimes I cannot trust him/her (reverse scored; Butler, 1991).
3. I can count on him/her to be trustworthy (Butler, 1991).

Satisfaction with leader
1. I feel good to be around him/her.
2. I am satis®ed that his/her style of leadership is the right one for getting our group's job
done.
3. I am pleased (or satis®ed) with his/her leadership.
Collective identity
1. We see ourselves in the work group as a cohesive team.
2. In our work group, our con¯ict is out in the open and is constructively handled.
3. Members of our organizational unit share the same values about our task and purpose.
4. Among our work group, we are remarkably similar in our values about what has to be
done.
5. There is widely shared consensus about our goals and the approaches needed to achieve
them.
Group performance
1. We have high work performance.
2. Most of our tasks are accomplished quickly and eciently.
3. We always set a high standard of task accomplishment.
4. We always achieve a high standard of task accomplishment.
5. We almost always beat our targets.
Empowerment
1. I can in¯uence the way work is done in my department.
2. I can in¯uence decisions taken in my department.
3. I have the authority to make decisions at work.
4. I have the authority to work eectively.
5. Important responsibilities are part of my job.
6. I have the capabilities required to do my job well.
7. I have the skills and abilities to do my job well.
8. I have the competence to work eectively.
9. I can do my work eciently.
10. I can handle the challenges I face at work.
11. I am inspired by what we are trying to achieve as an organization.
12. I am inspired by the goals of the organization.
13. I am enthusiastic about working toward the organization's objectives.
14. I am keen on our doing well as an organization.
15. I am enthusiastic about the contribution my work makes to the organization.

Conger Et Al 2000

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Conger Et Al 2000

Uploaded by

Copyright:

Available Formats

Journal of Organizational Behavior

J. Organiz. Behav. 21, 747±767 (2000)

Charismatic leadership and follower eects

The Conger±Kanungo model of

Hypothesized relationships between charismatic

Reverence for the leader

Trust in the leader

Satisfaction with the leader

Work-group collective identity

Perceived group task performance

Leader focus variables

Follower focus variables

Table 1. Descriptive statistics and intercorrelations

Table 2. Results of hypothesis testing: leader focus variables*

Table 3. Results of hypothesis testing: follower focus variables*

Table 4. Factor loadings for C±K scale items

Table 5. C±K sub-scales: descriptive statistics and intercorrelations

Reverence Trust Satisfaction Collective Group Empowerment

3. I can count on him/her to be trustworthy (Butler, 1991).

You might also like

Conger Et Al 2000

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Conger Et Al 2000

Uploaded by

Copyright:

Available Formats

Journal of Organizational Behavior

J. Organiz. Behav. 21, 747±767 (2000)

Charismatic leadership and follower eects

The Conger±Kanungo model of

Hypothesized relationships between charismatic

Reverence for the leader

Trust in the leader

Satisfaction with the leader

Work-group collective identity

Perceived group task performance

Leader focus variables

Follower focus variables

Table 1. Descriptive statistics and intercorrelations

Table 2. Results of hypothesis testing: leader focus variables*

Table 3. Results of hypothesis testing: follower focus variables*

Table 4. Factor loadings for C±K scale items

Table 5. C±K sub-scales: descriptive statistics and intercorrelations

Reverence Trust Satisfaction Collective Group Empowerment

3. I can count on him/her to be trustworthy (Butler, 1991).

You might also like

Charismatic leadership and follower eects