Leadershi Effectivenes

s
p
FRED E. FIEDLER
of
Universi
ty Washington

To the layman, leadersh is something a person &dquoan

inherited or
ip traiwhich,
acquire therefor remains withthe
;has,
There t littlevidence
person. d is however, very e, to support
&dquo thi
widel held belief
, (Stogdil e
197. Outstandin general do snot
;
make
necessari outstandin l, 4)research
g directors
s chairmen of
or
y
community
ly clubsgand effectiv presiden of hospit boardsor
radio stations, do not e
necessari make outstandin sales
ts al man-
agers or tank divisioncommanders.
ly Whether
g a person issuccess-
ful ina leadership seems to depend as much on thesituatio
as on thepersonali and skills or her
she bring to then
jobFor
job
thisreason,tycontemporary leadership s asks . kind
&dquo;W
ofpeople are require forthe successful hat
theory leadership ,
ina particul dtype of situation?
&dquo;
performance
ar

EMERGENT LEADERSHIP

Earlyleadersh research
primaril concerned itselwithhow
one becomesa leader. Althoughthisarticl f
focuses factor
ip y on

that determineleadersh e
effectivene sorder
a few words are in
on theemergence
ip of leaders
ss,
As alread . leaders
mentioned who emergeininformal groups
or
through
y the
,process ofelection
cannot readil be distinguis
on the basisof their traitory attributes
from those
personali hed
s
ty

619

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

620 AMERICAN BEHAVIORAL SCIENTIST

of
who are the members Practically
their groups.
everyone performs leadership functions in
some
groups and organizations, for example,
chairing a committee, directing a work crew,
managing a
department or
office, or

presiding over a meeting.

different
And all of us are members of many groups
and organizations. By and large, our
personality does not determine whether we are
leaders or members of a particular group. More
often it simply depends on whether we want to be
in leadership position and happen to have
a
the
particular knowledge, skills, or
A
resources the group needs at that time. bowling
team
will, therefore, choose a good bowler as
its
captain, and a protest group will choose a
leader who can express the group’s feelings.
Some consistent results have been found in the
more than 200 studies on leadership traits.
These have shown that the leaders tend to be
somewhat somewhat
brighter than followers,
taller, somewhat socially adept, and more
more

talkative. In other words, people who fade

into the
wallpaper are less likely to become
crowd
leaders than those who stand out in the
(Hollander, 1978).
There are also the leaders who stir our
imagination, who have the
&dquo;charisma&dquo; or gift to kindle the
enthusiasm of their

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

that the effective leaders did not differ
greatly in traits and attributes from those
who were ineffective. The most that can be said is
that the effective leaders tend to be slightly
more intelligent than the members of their
and
groups slightly more
socially competent,
fndings which very similar to those obtained
are

in studies of emergent leadership. Even these

few differences between effective and ineffective
leaders are too small to be of practical
significance in most selection procedures.
A number of important theories hold,
nevertheless, that individuals with certain
a
personality, philosophy of
management, or
will
attitude generally be more effective tha
others, that is, that there is an ideal type of
leader. These theories are intuitively
appealing and have had a considerable
influence on modern managerial thinking.
Two of the best known theories of this type were

developed by
Rensis Likert and by Douglas McGregor.
Likert’s (1967) &dquo;System 4&dquo;
proposed that effective the
manager is
participative and has an approach to
management that is based on trust and open
and
communication with subordinates superiors.
In a somewhat similar vein, McGregor (1967)
of
postulated opposing two
theories
and
management, called &dquo;Theory X&dquo;
&dquo;Theory Y.&dquo; Theory X managers

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

a similar tale was told by Warren Bennis (1970),
one of McGregor’s disciples, suggesting
universal
that these theories do not have
applicability. Empirical tests have
generally not supported these theories.
But if neither personality traits nor
philosophical management approaches and
attitudes have identified the ideal method of
leadership, are there certain types of leader
successful
behaviors that assume leadership
It is
performance? obvious, of course, that
leaders must communicate with their followers
by word or gesture; and it seems only reasonable
to hypothesize, therefore, that certain
types of interpersonal behaviors are
in
effective while others are ineffective
getting the group to perform its task.
This position is best represented by the
massive research on leader behaviors conducted at
Ohio State University under the direction of
Carroll Shartle in the 1940s and 1950s
(Stogdill,
1974). This impressive series of studies
isolated two major behavior clusters by which
subordinates described their superiors. The
first of these, called
&dquo;structuring,&dquo; consists of such
behaviors as organizing the group
tasks to
interaction, assigning roles and
group members, setting and
maintaining
standards, and evaluating the subordinates’

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

performance. The second cluster, called
&dquo;consideration,&dquo; is defined as
showing concern for the welfare and well-being
of group members, involving them in the
decision-making process, listening to
and
their opinions and suggestions,
treating them as equals.
The
validity of these behavior clusters is
now
well-established, and similar
&dquo;task&dquo; and &dquo;group
found
maintenance&dquo; factors have been by
other of
investigators. However, none these
behaviors differentiates the successful from the
unsuccessful leaders (Korman, 1966). The
ineffective leaders are as likely to be
considerate and structuring as are the
effective leaders.
House’s (House and Mitchell, 1974) Path-
Goal Theory takes a more sophisticated
approach. It specifies the conditions
which require considerate behaviors and those
which
require structuring behaviors. The
theory states that the leader must guide the

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015
 Fiedler
/ LEADERSHIP EFFECTIVENESS 625

Figure1: Graphicrepresentation of the Contingency Model, indicating

the inter-
action of the leader’s measured by the Least
motivational structure,
PreferredCo-worker (LPC)scaleand situational control.
,

Figur1 implie that the effectivenesscan be

improve
either leader’s
leader’s
bye changin
s the personali to match
dsituatio
the
or else
by
g modifyin situationalcontrol to match thenleader’
ty s affec
in stuational
personali gA change controlactuall does
the performanc of task- and relationship- leaders t
ty. y
different
e Thishas now been shownin studiesof
motivated widely
organizations,
ly. school administrators polic
varyingsuperv
sors, post
including officemanagers, ,
company executives e and milita
i-
personn , ry
el.For example inone study ofarmy infantr squads (Bonsand
Fiedle , 1976, situational controlincreased
y for leaderswho
remained
r, ) with
thesame unitand under the same superior for
six
s

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

624

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015
 Fiedler
/ LEADERSHIP EFFECTIVENESS 627

2: Comparisonof task performance

Figure by relationship- and task-
motivated leaderswith and without task (from Chemers et al
training
motivated
1975). .,

Model predic thattask-motivatedleaders shouldperform better

in thets
low control(untraine conditionwhilerelations
motivated leaders should
d)perform better in trained
the (moderat
hip-
control condition. results
The this
supporte hypothesi e (se
Figur
) 2 . However, of importanc
even greater isthe
s findinthat
e
d
the
e task-motivated
) leaderswithoutetraininhad better
g perform
inggroupsthandidtask-motivated leaders
g withtrainin -We may
infer that
therefor leadersh trainin in case
g. increased the
performance
, e, ofrelationship- this
leadersbut decreased the
ip g
of task-motivated
performancmotivated leaders.
e The studies which demonstrate that a change in situationa
controlchanges leadership different ltask- and
for
performance ly

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

 is being performed. This is
essentially

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

628 AMERICAN BEHAVIORAL SCIENTIST

for relationship-motivated leaders have

led to a training method based on the
Contingency Model, called LEADER MATCH
(Fiedler et al., 1976), which capitalizes on
these
findings. The method assumes that it is
difficult if not impossible to change one’s
leadership style or leader behavior, but
that it is relatively
easy to change one’s leadership situation.
A leader can, for instance, modify leader-
member relations by becoming more or less
accessible subordinates, the leader can
to

change task by asking the boss for

structure

very detailed instructions

and
clearly defined
and
jobs or for broad challenging
problems. The leader can increase position
in the
power by becoming a expert group’s
the or decrease it
task, or
acting like one
of gang,
by participative management or relying on
subordinates for guidance on how the task is to be
performed.
A number of well-controlled studies in
civilian
military as well as organizations
have tested the effectiveness of the LEADER MATCH
and
training program (Fiedler Mahar,
1979b). Although the training requires
only from six to eight hours, supervisory
ratings obtained 2 to 12 months later showed

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

that thetrained leaders had performed
significantly and

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015
630 AMERICAN BEHAVIORAL SCIENTIST

with
intelligence and experience
performance under two contrasting conditions:
for groups in which the leader reported low
the
stress with the boss and for those in which
reported stress with boss was high. The results
were quite consistent over different studies:
leader’s
When stress with boss was low, the
performance correlated with his intelligence
with his
but not experience; when stress with
boss was high, performance correlated with
his experience but not with his
intelligence. This suggests that the use of
one’s creativity and problem-solving
abilities requires a nonthreatening,
stress-free
relaxed, and interpersonal
these
environment. Under conditions, also,
leaders
attempt to find new solutions rather than
relying on what they have learned in the
past. When interpersonal stress is high,
the situation is too threatening to solve
problems or seek new solutions. Rather, the
will
leader rely on the safe and proven behaviors
and actions learned from past experience. The
more experience he has, the more likely it is
that the leader will have a correct solution in his
repertoire.

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015
 is obvious from these findings
It that
intellectual abilities and experience do play
an
important part in
leadership
performance.
to
However, they contribute performance

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015

Downloaded from abs.sagepub.com at UNIV OF CALIFORNIA SANTA CRUZ on April 2, 2015