Leadership styles ,cultural values ,managers ,subordinates,Soviet Union, Germany, the US

HRDI 5:1 (2002), pp. 99–117
Leadership styles and cultural values among
managers and subordinates: a comparative
study of four countries of the former Soviet Union,
Germany, and the US
Alexander Ardichvili and K. Peter Kuchinke
University of Illinois at Urbana-Champaign
Abstract: This cross-cultural study utilized the full range leadership framework
developed by Bass and Avolio and Hofstede’s model of culture, and compared leadership
styles and cultural values of over 4,000 managerial and non-managerial employees in
ten business organizations in Russia, Georgia, Kazakhstan, Kyrgyzstan, Germany, and
the US. Regarding socio-cultural dimensions, the study found that, compared to Germany
and the US, the four former USSR countries differed primarily by much lower levels of
Power Distance, higher levels of Masculinity and much longer planning horizons. The
results on leadership indicate that two dimensions – Contingent Reward and Inspirational
Motivation – produced the highest scores in all four countries of the former USSR.
Two less efficient leadership styles, Laissez-faire and Management by Exception, have
received signi cantly higher scores in the four former USSR countries, than in the US
and Germany. Finally, the study suggests that cross-cultural human resource develop-
ment issues cannot be described in terms of simpli ed dichotomies between the East and
West. For constructs measured in this study, signi cant differences were found not only
between the two groups of countries but also between individual countries within these
groups.
Keywords: leadership styles, national culture, international HRD, former USSR,
Russia
In the past ve years, research on leadership and management has evolved as a key area
of interest among HRD scholars (Jacobs 2000; Woodall 2000). Within this area, two
strands of scholarship can be discerned: the development of leaders and managers and
their behaviors, attitudes and attributes. The former has received much interest from
UK scholars, with notable recent UK publications by Lee et al. (1996) on management
education in the post-communist Central European States, Woodall and Winstanley
(1998) on the strategy and practice of management development, and Lessem (1998)
on management development through cultural diversity. Originating in the social
science tradition of US academia, studies on behavioral aspects of management and
leadership are regularly found in HRD research journals, for example, issues of managers
acting as facilitators of organizational learning (Ellinger et al. 1999), leadership styles
of Russian entrepreneurs (Ardichvili et al. 1998), and managers’ abilities to create
organizational visions (Thoms and Greenberger 1998).
Studies comparing two or more countries, however, are still rare in HRD research,
and this despite repeated calls for the comparative, cross-national research needed to
Human Resource Development International
ISSN 1367-8868 print/ISSN 1469-8374 online © 2002 Taylor & Francis Ltd
http://www.tandf.co.uk/journals
DOI: 10.1080/13678860110046225

---

Peer-Reviewed Articles
keep pace with the rapidly developing international and multi-cultural organizational
environments (for example, Hansen and Brooks 1994; Peterson 1997). Especially
lacking is scholarship addressing issues of leadership in post-communist Europe and
Asia where countries of the former USSR are facing the drastic challenges of political
and economic transformations and the adjustment to a market-driven, competitive
world economy. While some single-country research is available in the HRD literature
(for example, Lee et al. 1996), most of the countries that gained independence
after the break-up of the Soviet empire remain, from a scholarly perspective, terra
incognita.
The authors of this article, then, hope to contribute to the eld in three ways. By
using an established theoretical framework for measuring leadership and culture
(Kuchinke 1999), this study strengthens the international HRD knowledge base,
heeding the call for increased replication of research by Russ-Eft (1999). By including
four new countries in addition to the ones surveyed by Kuchinke, the study expands
existing knowledge and allows comparisons of the variables across six countries with a
sample of over 4,200 employees and managers in ten organizations. This is of particular
importance since the research design allows for comparisons not only of countries, but
also of geographic, political and economic regions that differ vastly from each other,
the established Western economies of the US and Germany and the newly emerging
economies of the former Soviet Union. Finally, the study begins to build an HRD
knowledge base in countries where it is virtually non-existent, namely Georgia,
Kazakhstan, Kyrgyzstan, and, to a lesser extent, Russia.
Related research, theoretical framework, and research questions
Socio-cultural dimensions
The concept of culture is widely used in international management, organization
behavior, and human resource development literature to measure effects that can
discriminate between countries and ethnic or occupational groups (Kuchinke 1999).
Culture constitutes the successful attempt to adapt to the external environment; it
presents a social group’s shared strategy for survival (Triandis 1995). In this study, we
used ve socio-cultural dimensions, identi ed by Hofstede (1984, 1997). The rst
dimension is called power distance (PDI), and is de ned as the degree of inequality
among the people which a group of people considers as normal. The second dimension,
individualism (IND), is the degree to which people prefer to act as individuals rather
than as members of groups. The third dimension, masculinity (MAS), is the degree to
which such ‘masculine’ values as assertiveness, competition, and success are emphasized
as opposed to such values as quality of life, warm personal relationships, and service.
Uncertainty avoidance (UAI) is the degree to which people in a country prefer
structured over unstructured situations. Finally, the fifth dimension, long-term
orientation (LTO), was intended to account for speci c traits of many Asian cultures,
which were not covered by the first four dimensions (Hofstede 1993). Long-term
orientation is de ned as the degree to which people’s actions are driven by long-term
goals and results, rather than the short-term results and the need for immediate
grati cation.
100

---

Ardichvili and Kuchinke: Leadership styles
According to Hofstede (1997), the US business culture is characterized by low PDI,
LTO, and UAI, and high IND and MAS. Furthermore, German employees display
low PDI and LTO, and high UAI, MAS, and IND. Regarding Russian managers,
Hofstede hypothesized that they would be characterized by high PDI, high UAI,
medium-range IND, and low MAS (Hofstede 1993). Bollinger (1994) and Naumov
(1996) tested Hofstede’s hypotheses in their studies of Russian managers, and found
support for these predictions on all four dimensions. Elenkov (1998), in his comparative
study utilizing Hofstede’s dimensions, found that US managers are more individualistic
than their Russian counterparts and the managerial culture in the United States is also
characterized by lower power distance and uncertainty avoidance than the Russian
managerial culture.
Prior research providing data on Hofstede’s socio-cultural value dimensions for
Georgia, Kazakhstan, or Kyrgyz Republic could not be found. Thus, due to the lack
of specific findings on these three countries, relational hypotheses regarding the
differences in socio-cultural dimensions between all six countries in this study could not
be developed, and our rst research question was formulated as follows:
R1: What are the socio-cultural value dimensions of managers and employees in selected
organizations in the six countries? Are there signi cant differences between these six
countries on any of the socio-cultural dimensions?
Leadership
Since the late 1980s, much of the leadership research has concentrated on characteristics
and speci c effects of charismatic and transformational leadership (Bass 1985; Kanungo
1990; Sashkin 1988; Tichy and Devanna 1990). In this study, we used a version of
transformational leadership theory formulated by Bass and his colleagues (Bass 1985,
1996; Avolio et al. 1995). According to Bass (1985), transformational leaders motivate
their followers by inspiring them, offering challenges, and encouraging individual
development. Transformational leadership stresses achievement of higher collective
purpose, of common mission and vision. The second leadership style is transactional
leadership. Transactional leaders stress speci c bene ts that their subordinates would
receive by accomplishing agreed-upon tasks. A transactional leadership style involves
negotiations between leaders and their subordinates, and exchange relationships
between them. Research shows that different behaviors are involved in transformational
and transactional leadership. The behaviors are measured with the Multifactor
Leadership Questionnaire (MLQ) (Avolio et al. 1995). Transformational leadership
includes individualized consideration (IC), intellectual stimulation (IS), charisma
(CHA), and inspirational motivation (IM). Transactional leadership includes contingent
reward (CR) behavior and management by exception (ME). A series of studies reviewed
by Bass (1996) support the distinction between transformational and transactional
leadership. There is also considerable evidence that transformational leadership
is effective, and is positively related to subordinate satisfaction, motivation, and
performance (Lowe et al. 1996).
Transactional and transformational leadership styles are contrasted with laissez-faire
leadership. Laissez-faire leaders abdicate their responsibility and avoid making decisions
(Bass 1990b). Subordinates working under this kind of supervisor basically are left to
101

---

Peer-Reviewed Articles
their own devices to execute their job responsibilities. Although laissez-faire leadership
is observed infrequently in the US businesses (Bass and Avolio 1989), managers still
exhibit it in varying amounts (Bass 1990a). Prior research has found that laissez-faire
leadership has an adverse effect on work-related outcomes of employees (Bass 1990a;
Yammarino and Bass 1990).
Most of the extant leadership research is based on data collected in the North
American context (Northhouse 1997). Regarding leadership styles of German
managers, Kuchinke (1999), in his comparison of US and German telecommunications
employees, has found that the US respondents ranked higher than Germans on two
dimensions of transformational leadership (charisma and inspirational motivation).
The majority of studies dealing with leadership styles of managers in Russia are based
either on consulting or teaching experience of the authors, or on a limited number of
case studies (e.g. Berger 1999; Clarke 1996). Attempts were made at developing lists
of leadership traits or management styles of business people in the transitional
economies. For example, a framework developed by Puffer (1996) suggests that
contemporary Russian managers tend to share power, are inclined to delegate decision
making (largely to avoid the responsibility for unforeseen consequences), prefer to
concentrate on strategic decision making, are tenacious and energetic, and have strong
collectivist attitudes. As in the case of socio-cultural dimensions, literature on leadership
styles of managers in the other three countries of the former USSR could not be found.
Thus, our second research question was also exploratory in nature:
R2: What are the leadership styles of managers in the six countries? Are there any
signi cant differences in leadership styles between the six countries?
The leadership and culture interface
Leadership exists in all societies and is essential to the functioning of organizations
within societies. However, the attributes that are seen as characteristic for leaders may
vary across cultures (Den Hartog et al. 1999). House (1995) noted that prevailing
theories of leadership are North American in character, and are based on the
assumptions of individualism as opposed to collectivism, rationality rather than ascetics,
hedonistic rather than altruistic motivation, centrality of work, and democratic value
orientation. Cross-cultural psychology and sociology research shows that many cultures
do not share these assumptions (Den Hartog et al. 1999). ‘As a result there is a growing
awareness of the need for a better understanding of the way the leadership is enacted
in various cultures’ (House 1995: 443).
Bass (1997) argued that there is universality in the transactional–transformational
leadership paradigm, and presented supporting evidence collected in organizations
in business, education, the military, the government, and the independent sector,
from several continents. Bass (1997) maintained that the same conception of
phenomena and relationships can be observed in a wide range of organizations and
cultures, and exceptions can be understood as a consequence of unusual attributes of
the organizations or cultures. Den Hartog et al. (1999), in their study in sixty-two
cultures, found that, although cross-cultural research emphasizes that different cultural
groups are likely to have different conceptions of what leadership should entail, certain
attributes associated with transformational leadership are universally endorsed as
102

---

Ardichvili and Kuchinke: Leadership styles
contributing to outstanding leadership, and some other leadership attributes are
universally seen as impediments to outstanding leadership. Jung et al. (1995) speculated
that transformational leadership is more effective in collectivist cultures than in
individualist cultures, being enhanced by the respect for authority and obedience
characteristics of collectivist cultures. Further, Jung et al. (1995) hypothesized that
high uncertainty avoidance cultures may require more transaction-based leadership,
while low uncertainly avoidance cultures will tolerate more innovative, transformational
behavior (Jung et al. 1995). Elenkov (1998) argued that, since Russian managerial
culture is characterized by high power distance and a strong collective mentality, Russian
employees expect an autocratic leadership style, which is offset by the support given to
subordinates’ families. In addition, Elenkov (1998) asserted that American concepts
of leadership that advocate participation in managers’ decisions by their subordinates
(small power distance) and that presuppose the con dence and ability to negotiate with
one’s boss (high individualism) are incompatible with the large power distance/low
individualism of Russian managerial culture. Due to the lack of previous empirical
ndings to guide hypotheses regarding speci c relationship between leadership and
the socio-cultural dimensions in the former Soviet Union, the following research
question was formulated:
R3: What is the relationship between culture and leadership in the six countries in this
study?
Methodology
This study was conducted using causal-comparative, one-shot survey design (Howell
1992). Twelve sites (including nine independent companies and three divisions of one
company) in six countries have been surveyed. Companies in the former USSR were
identi ed and contacted by the Center of Sociological Research (CSR) of the University
of Moscow (three companies in Russia, three in Kyrgyz Republic, two in Kazakhstan,
and one in Georgia). All ten rms were in the manufacturing sector of industry. Two
levels of employees were surveyed: middle-level managers and non-managerial
employees, including engineers and production employees. In all cases, managers were
immediate supervisors of relevant non-managerial employees. A survey was administered
to strati ed random samples of employees in each of the companies surveyed in the US,
Russia, and Kazakhstan, and to populations of all employees in Germany, Georgia, and
the Kyrgyz republic. Data collection procedures were similar in the US and Germany.
After a site visit by the second author, survey instruments were distributed by the
personnel department at each location with a cover letter explaining the purpose of the
study and the voluntary and anonymous nature of the study. Respondents were asked
to place the completed instruments in drop boxes that were placed throughout the
three sites (main entrance, cafeteria, and conference/meeting rooms), collected by the
site’s personnel staff, and forwarded to the researcher. Data collection in the four
countries of the former USSR was organized in a slightly different fashion, to re ect
the realities of organizational and work cultures in these countries, where mail surveys
managed at a distance do not yield signi cant response rates. Thus, in all four countries
representatives of CSR made presentations explaining the study goals to management
103

---

Peer-Reviewed Articles
and to employees, during regular employee meetings. Next, at the survey stage, the CSR
representatives collected completed surveys from participants, who were filling the
surveys out at or after general employee meetings.
The survey instrument was developed using a number of existing, field-tested
instruments. The five culture constructs were measured by using Hofstede’s 1994
version of the Values Survey Module, VSM 94 (Hofstede n.d.). The leadership
construct was measured utilizing Avolio et al.’s (1995) MLQ5x questionnaire. Both
the VSM 94 and the MLQ5x have been used extensively in the past and have known
psychometric properties (for VSM94, see Søndergaard 1994; for MLQ5x, see Avolio
et al. 1995). The instruments were obtained from the test authors in the English and
German versions and given to the US and German participants in their native language.
The survey was conducted in Russian in Russia, Kazakhstan, and the Kyrgyz Republic,
and in Georgian in the Georgian Republic. The instrument was translated into Russian
and Georgian by the representatives of the CSR. A back translation was made by
independent translators. The translations were compared to the original sources, which
helped to identify and correct a number of errors that have arisen from interpretation
differences. The co-author of MLQ5x, Bruce Avolio, checked the Russian back
translation of the leadership section of the survey instrument for accuracy. A pilot survey
was conducted with a sample of 100 employees of an experimental plant of the
University of Moscow to identify potential problems with the interpretation of terms
and concepts.
Since we were interested in country-level comparisons, the results reported in this
manuscript are based on pooled, country-level data, and not on the individual company
data. The country-level response rates ranged from 31 per cent in Russia to 75 per cent
in Kyrgyz Republic, with the total number of usable responses at 4,065. To address the
issue of possible response bias, comparisons (utilizing chi-squares) were made between
the demographic data on respondents and on all employees in a given company,
obtained from the personnel departments. These comparisons showed that there was
no signi cant difference between survey respondents and non-respondents in the two
categories (managers and employees). Information about the sample by country and
demographic data are reported in Table 1.
The overall percentage of male respondents was 68 per cent but higher than 70 per
cent in Georgia, Germany, and the US. The demographic variables of age, gender, and
education were statistically controlled in the following analyses to remove the possible
effects of the over-representation of certain gender categories in individual countries.
The two US sites had a much older population with a modal age of between 50 and
59 years, while the overall age for the six countries was 30 to 39 years. Managers at the
German and US sites had a modal level of education comparable to a master’s degree,
while non-managerial employees at the German site had a higher modal level of
education than their US counterparts. Overall, the educational levels of employees and
managers in all four countries of the former USSR were high, with the majority of
managers possessing a ve-year college degree or higher, and the majority of employees
having at least a two-year college degree.
104

---

)
)
)
6
8
3
2
1
4
,
5
2
0
9
,
9
5
5
3
0
–
3
9
v
e
r
a
l
l
8
5
9

(
2
1
%
O
4
,
0
6
5

(
4
1
%
3
,
2
0
6

(
7
9
%
2
-
y
e
a
r

c
o
l
e
g
e
)
)
)
a
n
y
8
0
2
0
1
,
7
0
0
1
,
6
9
5
a
s
t
e
r
’
s
e
r
m
5
3

(
5
%
3
0
–
3
9
M
G
1
,
0
4
9

(
6
2
%
9
9
6

(
9
5
%
)
)
)
7
1
2
9
3
,
7
5
0
1
,
8
4
2
5
0
–
5
9
S
6
2
5

(
3
4
%
1
1
1

(
1
8
%
5
1
4

(
8
2
%
U
2
-
y
e
a
r

c
o
l
e
g
e
)
)
)
5
2
0
5
2
0
5
2
4
8
a
s
t
e
r
’
s
y
r
g
y
z
s
t
a
n
4
0
–
4
9
K
3
9
1

(
7
5
%
8
8

(
2
3
%
3
0
3

(
7
7
%
M
)
)
)
6
0
4
0
a
z
a
k
h
s
t
a
n
1
,
8
0
0
1
,
1
8
8
4
0
–
4
9
a
s
t
e
r
’
s
K
3
8
5

(
3
2
%
1
1
8

(
3
1
%
2
6
7

(
6
9
%
M
)
)
)
=

4
,
0
6
5
)

8
5
1
5
e
o
r
g
i
a
7
5
0
7
5
0
3
0
–
3
9
a
s
t
e
r
’
s
G
3
9
9

(
5
3
%
1
2
8

(
3
2
%
2
7
1

(
6
8
%
M
)
)
)
5
6
4
4
6
,
0
0
0
3
,
9
6
0
3
0
–
3
9
u
s
s
i
a
3
6
1

(
3
0
%
8
5
5

(
7
0
%
R
1
,
2
1
6

(
3
1
%
2
-
y
e
a
r

c
o
l
e
g
e
o
g
r
a
p
h
i
c

c
h
a
r
a
c
t
e
r
i
s
t
i
c
s

(
N
o
d
e
)
p
l
o
y
e
e
s
p
l
e
R
e
s
p
o
n
s
e

r
a
t
e
s

a
n
d

d
e
m
a
n
a
g
e
r
s
a
l
e
a
l
e
o
d
e
)
l
e

1
o
t
a
l

e
m
a
l
i
d

r
e
s
p
o
n
s
e
s
a
n
a
g
e
r
s
o
n
-
m

M

F
e
m
I
n
i
t
i
a
l

s
a
m
V
M
N
%
e
s
p
o
n
s
e

r
a
t
e
T
e
n
d
e
r
%
g
e

(
m
i
g
h
e
s
t

d
e
g
r
e
e

e
a
r
n
e
d

(
m
T
a
b
R
G
A
H

---

Peer-Reviewed Articles
Results
The scores for the seven dimensions of leadership and ve dimensions of culture were
calculated based on the formulae provided by the test authors (Avolio et al. 1995;
Hofstede n.d.). The leadership scores for the four transformational styles of charisma,
inspirational motivation, intellectual stimulation, and individual consideration ranked
higher in most countries than the two transactional styles of management by exception
and contingent reward. Laissez-faire leadership behavior was present but ranked lowest
among the seven styles. All dimensions of leadership showed excellent internal
reliabilities, with the exception of management by exception which came very close to
the commonly accepted reliability coef cient of a = .7 (Nunally 1967). Leadership
mean scores, ranging from 0 (low) to 4 (high), their standard deviations, and reliability
coef cients are shown in Table 2.
The scores for the five dimensions of culture – power distance, individualism,
masculinity, uncertainty avoidance, and long-term orientation – were calculated using
weighted means of individual items and constants, as described by Hofstede (n.d.),
which result in a distribution ranging from 0 to 100, although above or lower scores
are possible (Hofstede 1994). This allows for comparisons with previously published
country scores. The internal reliability for the entire instrument was a = .88, but the
dimension of uncertainty avoidance fell short of the minimum required and was
therefore excluded from the analyses. This dimension had shown poor internal reliability
in previous studies (Kuchinke 1999). Masculinity also fell below the recommended
value but was included in the analyses because it approximated the cut-off score.
However, the low reliability coef cient of a = .53 for this dimension raises concerns
about the factor structure of this dimension, and this needs to be seen as a limitation
to the study. While most of the culture scores were within the range of 0–100, several
countries scored above or below. Power distance scores, in particular, were negative for
the four countries of the former Soviet Union, indicating very low levels of this
dimension.
To answer the questions about country-level differences, two separate multivariate
analyses of co-variance (MANCOVA) were conducted with country as the independent
variable and dimensions of leadership and culture respectively as the dependent vari-
ables. In both analyses, demographic information (age, gender, and education) as well
as job category, site, and company were statistically controlled in order to rule out
their in uence on the results. By statistically controlling these variables, the analyses
focused on the country-level effects alone, without having the results distorted by these
other factors. Valid MANCOVA results depend on the data meeting a number of
assumptions: random sampling, independence of observation, and normal distribution
and homogeneity of variance among the variables (Bray and Maxwell 1985). Where
these assumptions are violated, the statistical analysis will result in invalid ndings. The
rst two conditions were ful lled by the research design, the nal two, however, were
not met for all the constructs. To correct these shortcomings, the sample sizes for each
country were equalized by selecting, at random, a sample of 392 from each of the six
countries. Given equal sample sizes, as Bray and Maxwell point out, ‘all of the test
statistics tend to be robust, unless the sample sizes are small’ (1985: 34). Thus, the
comparisons of culture and leadership by country were based on a sample of 2,352
employees, which represents a strati ed random sample of the initial population. An
106

---

0
.
7
9
0
.
8
5
0
.
7
9
0
.
9
2
0
.
9
0
0
.
5
4
0
.
7
9
a
n
y
S
D
4
7
.
1
9
5
0
.
1
2
9
1
.
1
1
4
3
.
5
1

=

c
o
n
t
i
n
g
e
n
t
W
e
r
m
E
G
2
.
1
8
2
.
2
7
2
.
3
4
2
.
1
8
2
.
2
3
1
.
4
8
0
.
9
9
3
5
.
2
5
5
2
.
6
0
4
4
.
0
0
a
s
c
u
l
i
n
i
t
y
,

M
–
2
0
.
5
0
S

=

m
A
0
.
9
8
1
.
0
2
0
.
9
7
1
.
0
6
1
.
0
4
0
.
7
1
0
.
9
4
,

M
S
S
D
4
6
.
6
9
4
2
.
4
6
9
1
.
1
3
4
3
.
5
1
U
2
.
3
9
2
.
6
7
2
.
2
3
2
.
1
1
2
.
2
4
1
.
4
7
0
.
9
5
M
2
9
.
0
5
7
7
.
8
5
1
2
.
8
0
4
3
.
7
0

=

i
n
d
i
v
i
d
u
a
l

c
o
n
s
i
d
e
r
a
t
i
o
n
,

R

=

i
n
d
i
v
i
d
u
a
l
i
s
m
N D
0
.
7
5
0
.
8
4
0
.
9
1
1
.
0
2
0
.
8
6
0
.
6
1
0
.
8
2
O
4
6
.
3
3
5
3
.
2
2
1
0
8
.
5
6
4
5
.
3
3
=
4
,
0
6
5
)
S
D
y
r
g
y
z
s
t
a
n
K
2
.
3
9
2
.
7
3
2
.
6
2
2
.
4
6
2
.
6
7
2
.
2
2
1
.
6
5
u
l
a
t
i
o
n
,

C
5
9
.
4
5
6
0
.
4
0
M
–
2
5
.
9
0
1
0
8
.
9
0
e
r

d
i
s
t
a
n
c
e
,

I
N
e
n
s
i
o
n
s

(
N
I

=

p
o
w
0
.
6
9
0
.
7
9
0
.
7
8
0
.
9
1
0
.
7
8
0
.
6
2
0
.
8
6
4
3
.
2
0
5
6
.
9
3
4
6
.
4
9
S
D
1
0
2
.
3
3

=

i
n
t
e
l
l
e
c
t
u
a
l

s
t
i
m
a
z
a
k
h
s
t
a
n
M
K
2
.
6
4
2
.
7
4
2
.
8
5
2
.
7
9
2
.
8
8
2
.
4
2
1
.
6
8
5
8
.
2
0
9
9
.
0
0
6
1
.
1
5
e
n
s
i
o
n
s
:

P
D
M
–
2
3
.
2
5
e
r
s
h
i
p

a
n
d

c
u
l
t
u
r
e

d
i
m
0
.
6
3
0
.
7
2
0
.
7
9
0
.
8
7
0
.
7
7
0
.
6
2
0
.
8
8
o
t
i
v
a
t
i
o
n
,

S
T
4
9
.
7
8
5
7
.
2
8
6
1
.
2
7
S
D
1
1
9
.
7
1
e
o
r
g
i
a
G
2
.
8
0
3
.
1
2
2
.
8
6
2
.
9
1
3
.
0
2
2
.
3
5
1
.
5
5
4
3
.
7
0
7
0
.
6
5
M
–
3
6
.
7
0
1
4
9
.
8
0
e
v
i
a
t
i
o
n
s

o
f

l
e
a
d

=

l
a
i
s
s
e
z
-
f
a
i
r
e
;

c
u
l
t
u
r
a
l

d
i
m

d

=

i
n
s
p
i
r
a
t
i
o
n
a
l

m
T FE
O
0
.
6
9
0
.
8
3
0
.
8
6
0
.
9
7
0
.
6
9
0
.
4
4
0
.
8
1
S
D
4
9
.
0
4
6
3
.
3
4
9
5
.
5
5
5
5
.
9
0
a
,

M
u
s
s
i
a
R
2
.
3
0
2
.
5
6
2
.
4
2
2
.
2
8
2
.
7
8
2
.
2
3
1
.
6
9
6
2
.
2
5
8
3
.
5
5
M
–
3
3
.
7
0
1
0
1
.
3
0

=

c
h
a
r
i
s
m
e
a
n
s
,

a
n
d

s
t
a
n
d
a
r
d
A
H ent by exception, L
a
n
a
g
e
m
h
r
o
n
b
a
c
h

o
r
i
e
n
t
a
t
i
o
n
.
C a
.
8
0
.
7
8
.
7
8
.
7
9
.
7
2
.
6
7
.
7
2
.
7
7
.
8
2
.
5
3
.
8
1
e
n
s
i
o
n
s
:

C

=

m
E
R
e
l
i
a
b
i
l
i
t
i
e
s
,

m
B
l
e

2
e
n
s
i
o
n
A
T
N
E
I
S

=

l
o
n
g
-
t
e
r
m
a
r
d
,

M
i
m
W
H
O
M
O
O
E
B
F
E
D
O
D
A
T
e
y
T
T
a
b
D
C
M
S
T
C
R
M
L
P
I
N
M
L
K
l
e
a
d
e
r
s
h
i
p

d
i
m
r
e
w L

---

Peer-Reviewed Articles
omnibus MANCOVA for leadership styles by country with age, gender, education, job
category, and site as covariates showed signi cant differences, and so did the analysis
for the dimensions of culture (both p < .001). Follow-up univariate analyses of variance
(ANOVAs) using conservative Scheffe post hoc tests showed these differences in more
detail. When comparing data sets from potentially heterogeneous populations, as is the
case with most cross-cultural research, researchers are faced with the choice of using
normal scores or transforming respondents’ answers to normalized (z-) scores. Using
normal scores incurs the risk of nding spurious differences, while the use of z-scores
is likely to eliminate substantive ones. While either choice presents a trade-off, most
cross-cultural researchers warn against the use of standardized scores. For example, van
de Vijer and Leung (1997) recommend conducting a comparison of structures obtained
from standardized and non-standardized scores, and, if the difference is not signi cant,
using non-standardized scores. Although in this study several sources of extraneous
variation were controlled for through the design of the study, the authors conducted
country-level comparisons using both methods and found no differences. For example,
when using normal scores for charisma, Russian respondents differed signi cantly from
those in Georgia, and this was also found when using z-scores for this variable.
The remainder of this section presents our ndings on each of the three research
questions. To facilitate the description of the contrasts in leadership styles and culture
dimensions among the six countries, Table 3 shows homogeneous subsets for each
dimension, indicating which countries do and do not differ on each dimension (for
mean values on each dimension and the direction of differences, refer to Table 2).
Table 3 Homogeneous subsets (p > 0.5) for leadership and culture dimensions by
country (N=2,352)
Russia
Georgia
Kazakhstan
Kyrgyzstan
US
Germany
CHA
a
b
b
a
a
a
MOT
c, d
e
d
d
c, d
c
STM
f
g
g
f
h
f
CON
i
k
i
i
l
i
REW
m, o
n
n, o
m
p
q
MBE
r
r, s
s
r
t
t
LFE
u
u
u
u
v
v
PDI
A, B
A
B
B
C
D
IND
C
D
C
C
E
C, D
MAS
F
H
F
F
I
K
LTO
L
M
M
M
O
O
Notes
Like letters denote similar levels of the variable, different letters different levels (p< .05).
Leadership dimensions: CHA = charisma, MOT = inspirational motivation, STM = intellectual
stimulation, CON = individual consideration, REW = contingent reward, MBE = management by
exception, LFE = laissez-faire; cultural dimensions: PDI = power distance, IND = individualism,
MAS = masculinity, LTO = long-term orientation.
108

---