THESIS
Q
800}

 

 

llBRARY 3
i” Michigan State e
University

 

This is to certify that the

thesis entitled

DIMENSIONS OF NATIONAL CULTURE AS
MODERATORS OF THE RELATIONSHIPS BETWEEN
ORGANIZATIONAL ATTITUDE VARIABLES

presented by

Bradley J. West

has been accepted towards fulfillment
of the requirements for

M . A; degree in Psychology

Major pro/{gr

Date 5 May 2001

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c-JCIRCIDateDuapGG—nts

DIMENSIONS OF NATIONAL CULTURE AS MODERATORS OF THE
RELATIONSHIPS BETWEEN ORGANIZATIONAL ATTITUDE VARIABLES

By

Bradley J. West

AN ABSTRACT OF A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF ARTS
Department of Industrial/Organizational Psychology
2001

Dr. Ann Marie Ryan

 

Tali

 

job
sat:
He
aw
Hi
dll

Sll

ABSTRACT

DIMENSIONS OF NATIONAL CULTURE AS MODERATORS OF THE
RELATIONSHIPS BETWEEN ORGANIZATIONAL ATTITUDE VARIABLES

By

Bradley J. West

This study proposes a model that is used to investigate whether dimensions of
national culture moderate the relationships between certain organizational attitude
variables. Included in this model are the relationships between the outcome variables of
job satisfaction and workplace stress, as well as potential antecedents of those variables,
satisfaction with working in teams and attitudes about empowerment. Three of
Hofstede’s dimensions (individualism/collectivism, power distance, and uncertainty
avoidance) were employed as theoretical difl‘erences/similarities between cultures.
Hierarchical Linear Modeling was used to test for the interaction between cultural
dimensions and the relationships between the attitude variables. Results provide limited
support for the moderation model. Specifically, it was found that
individualism/collectivism moderates the relationship between empowerment and
satisfaction, and the relationship between satisfaction and stress, while power distance

moderated the relationship between empowerment and stress.

 

 

ME

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. v
LIST OF FIGURES ............................................................................... vi
INTRODUCTION ................................................................................. 1
Cultural Research .......................................................................... 2
Hofstede’s Dimensions........... ............................................... 3
Power distance ........................................................... 4
Individualism/Collectivism ............................................. 4
Uncertainty avoidance .................................................. 5
Other Theories of Culture ....................................................... 5
Boundaries of Culture ........................................................... 10
Organizational Attitude Dimensions .................................................... 1 1
Attitudes About Job Satisfaction .............................................. 12
Attitudes About Stress .......................................................... l6
Attitudes About Teams ......................................................... 19
Attitudes About Empowerment ................................................. 25
METHOD ........................................................................................... 31
Sample ...................................................................................... 31
Job Attitude Measure ..................................................................... 31
Cultural Dimension Measurement ...................................................... 36
Power Distance ................................................................... 36
Uncertainty Avoidance .......................................................... 37

iii

Individualism/Collectivism ..................................................... 37

RESULTS .......................................................................................... 37
Mean Difference Hypotheses ............................................................ 57
Moderation Hypotheses .................................................................. 60
Exploratory Results ....................................................................... 81

DISCUSSION ...................................................................................... 81
Supported Hypotheses ................................................................... 83
Other Interesting Findings ............................................................... 88
Discussion of Exploratory Findings ................................................... 90
Non-supported Hypotheses / Limitations ............................................. 92
Other Limitations ......................................................................... 95
Conclusions ................................................................................ 99

REFERENCES .................................................................................... 102

APPENDIX A: SCALE ITEMS ................................................................ 1 16

iv

 

 

T2

T2

Ta

LIST OF TABLES

Table 1. Included Countries and Respective Scores on Hofstede’s Dimensions ......... 32

Table 2. Means and Standard Deviations of Study Variables by Country for
1996 Data ................................................................................. 39

Table 3. Means and Standard Deviations of Study Variables by Country for
1998 Data ................................................................................ 42

Table 4. Means and Standard Deviations of Study Variables by Country for
1999 Data ................................................................................ 45

Table 5. Means, Standard Deviations, and Intercorrelations of Study
Variables for 1996 Data ............................................................... 48

Table 6. Means, Standard Deviations, and Intercorrelations of Study
Variables for 1998 Data ............................................................... 49

Table 7. Means, Standard Deviations, and Intercorrelations of Study
Variables for 1999 Data ............................................................... 50

Table 8. Results of the Null Model: Testing for Between Nation Variance in
Measures of Job Satisfaction, Stress, and Satisfaction with Teamwork. . . . . ....56

Table 9. Results of Testing Hypotheses 1, 2, and 5 for 1996, 1998, and 1999 data. . ....61
Table 10. Results of the Random Coefficient Regression Model ........................... 66

Table 11. Results of the Slopes as Outcomes Model for Moderation Hypotheses ....... 72

Flt:

Fig

 

LIST OF FIGURES

Figure 1. A Model of the suggested hypotheses ................................................ 13
Figure 2. An example of the HLM Null Model ................................................. 55
Figure 3. An example of the HLM model that was developed to test the mean

difference hypotheses ................................................................... 59
Figure 4. An example of the HLM Random Coefficient Regression Model. . .. . . . . . . . .....63
Figure 5. An example of the HLM Slopes-as-Outcomes Model ............................. 68
Figure 6. An example of the HLM Intercepts-as-Outcomes Model .......................... 71

Figure 7. Representative plot of the marginally significant interactions found for
Hypothesis 9 in both the 1996 and 1998 data sets .................................. 79

Figure 8. Plot of the significant interaction found for Hypothesis 10 in the
1998 data set ............................................................................. 80

Figure 9. Representative plot of the significant interactions found for the exploratory
Hypothesis for the years 1996, 1998, and 1999 .................................... 82

 

 

0t}

COT

Fa)

CFO

aIIC

OII

t0:

Strc

mc

Stu

19$

 

INTRODUCTION

Since the end of World War H, research interests regarding cross-cultural
organizational psychology have increased steadily in modern psychology. Works in
recent years by well-known scientists such as Hofstede (1991; 1993), Triandis (1980;
1993; 1994), Adler (1991), Schwartz (1992; 1994), and Erez and Barley (1993) among
others, represent a continued desire to understand culture and its relations with other
constructs. As the new century begins, continued globalization (Adler, 1997;
Fayerweather, 1986; Kobrin, 1984) and improved communication capabilities ensure that
cross-cultural organizational issues will continue to be on the minds of both researchers
and practitioners alike.

Two endeavors that are described as necessities for furthering our understanding
of cross-cultural organizational psychology are the analysis of cultural attributes’ relations
to organizational constructs (e. g., satisfaction, performance, contextual performance,
stress, leadership, team work, etc.) (Brett et a1, 1997), and the description of the
moderating influences that cultural variables may have on relationships between these
constructs (Barley & Erez, 1997). The present study investigates both of these questions
using data fi'om a large multinational corporation (U .8. based; collected in 1996, 1998,
and 1999) that contains information in the form of employee attitudes. While numerous
studies have discussed the way in which various cultures may hold differing attitudes on
certain constructs (e.g., a just world (Fumham, 1993), independence (Gudykunst et a1,

1994), self esteem (Crocker et a1, 1994), leadership (Schmidt & Yeh, 1992), motivation

(Holt & Keats, 1992), few have looked at how cultural tendencies may influence the
relationships between organizational attitudes. This study begins to address that question
by investigating the moderating effects of culture on various relationships between
attitudes on teamwork, supervision, empowerment, stress, and satisfaction. I will begin by
discussing culture, its definition and the cultural dimensions that may moderate the
relations between job attitudes. This will be followed by a discussion of each of the
attitude variables that are assessed in this study including their relevant relationships.
Specific hypotheses will be discussed throughout the following sections.
Cultural Research

Based on the last half-century of research, one issue that appears to be quite clear
and agreed upon is the fact that culture is a difficult construct to study. In fact, it is down
right messy. Definitions of culture vary from one study to the next (J ahoda & Krewer,
1993; Misra & Gergen 1993), primarily in terms of inclusiveness. For example,
Herskovits (1955) suggested an inclusive definition where culture is the human-made part
of the environment. Such a definition allows for the inclusion of topics ranging fi'om
shared tools or highway systems, to beliefs, attitudes and norms passed down among
generations. Others have chosen more focused definitions such as that by Shweder and
Levine (1984) who defined culture as a shared meaning system (Triandis, 1994). The
variation in definitions is problematic in that on one hand, there is a shared appreciation
for the importance of operationalizing and studying culture, while on the other hand the
lack of agreement makes it difficult to compare studies and accumulate knowledge on the

subject. Theoretical and research approaches also differ. From Pike’s (1966) emic/etic

distinctions, through Hui and Triandis’s (1985) pseudo etic, and Berry’s (1969, 1989)
three-step process, to the “one way” approach discussed by Brett et a1 (1997), the
theories, methods used, and opinions are diverse. The point is, while there appears to be a
general consensus regarding where the field of cross-cultural psychology needs to go,
there is more confusion about what the best way to get there is. Publications in this area
are beginning to attack this issue from a number of fronts. Among these are improved
instrument design (Graen et al., 1997), increased theoretically driven research (Triandis &
Bhawuk, 1997), attempts to understand causes for cultural differences/similarities (Brett
et al., 1997), and issues of mediation and moderation of organizational variables (Brett et
al., 1997).

As I have mentioned, there is no firm agreement on a definition of culture among
organizational psychologists. I tend to agree with House et a1. (1997) that culture refers
to “a set of parameters of collectives that differentiate the collectives from each other in
meaningful ways”. The desire to seek out and understand these differences between
cultures has led to a focus on the identification of cultural dimensions, and the similarities
and differences of values between cultures (Erez, 1994).

Hofstede’s Dimensions

The set of parameters that will be used in this study to define collectives are three
of Hofstede’s (1980) four well-known dimensions of power distance, individualism, and
uncertainty avoidance. Hofstede feels that culture programs the human mind with values
such that certain reactions are more likely in certain cultures. This suggests that

individuals from cultures that are formed around shared meanings and values will be

 

 

des
8V0

be'u.

 

 

follc

in or
dist;

dist;

IOCU

gTOI;

predisposed to react to stimuli or value certain stimuli such as working in teams in ways
that may differ from individuals from cultures with dissimilar shared values. Triandis
(1980) states that it is this predisposition to weight stimuli in a particular manner that is
the major contribution of culture in clarifying the relationships between situations and
behaviors. This aspect of cultural differences thus lends itself to the study of culture as a
moderator of the relationships between organizational attitude variables. For example, if
a collectivist values group harmony, won’t a negative group experience prove more
detrimental to overall job satisfaction than it would for an individualist who doesn’t value
group harmony?

Hofstede feels that the most important differences between cultures may be
described using the dimensions of power distance, individualism/collectivism, uncertainty
avoidance, and masculinity/femininity (Hofstede, 1991). Masculinity/femininity is not
being used in this study. A brief description of each of the three included dimensions
follows:

Power distance refers to the extent to which a society accepts the fact that power
in organizations is distributed unequally (Hofstede, 1980). Cultures high in power
distance prefer unequally distributed power structures whereas cultures low in power
distance prefer that power be distributed more equally among members.

Individualism/Collectivism: Individualism refers to the extent to which people are
focused on taking care of themselves (or immediate family) even when operating within a
group or societal structure. Collectivists give priority to group goals and distinguish

between in and out-groups, acting in ways that are both dependent on and for the benefit

 

”.3

It

OI]

be

C01

 

 

of the larger entity (T riandis et al., 1993; Hofstede, 1980).

Uncertaingg avoidance refers to the extent to which individuals within a collective
feel threatened by uncertain or ambiguous situations (Hofstede, 1.980). Cultures that are
high in uncertainty avoidance are those that are threatened by and attempt to avoid
uncertain situations. This is similar to what Triandis called “cultural tightness” which
suggested that the extent to which a culture dislikes uncertainty is at least partially a
function of the homogeneity of the culture itself.

While it is acknowledged that Hofstede’s research is over 20 years old, it remains
one of the most accepted, and well documented. As of 1991 , at least 61 replications had
been conducted concerning Hofstede’s dimensions. Of those that are even somewhat
comparable to Hofstede’s studies in terms of sample, size of study, type of data, etc., the
majority either fully or partially supported Hofstede’s dimensions including a large-scale
study by Hoppe (1990) and a follow-up study of Hofstede’s work using IBM employees
again by Lowe (1994).

Other Theories of Culture

Hofstede however, has not been the only researcher to work with issues regarding
cross-cultural psychology. For example, Schwartz and Bilsky (1987, 1990) and Schwartz
(1992, 1994) have conducted cross-cultural research with the intent of determining the
universal nature of both the content and the structure of values. Schwartz (1992)
conducted a study involving 30 nations and found that an initial list of 36 value variables
(e.g., a comfortable life, self-respect, freedom, forgiving, loving, responsible) could be

reduced to a grouping of 7 “motivational domains” (pro-social, security, maturity, self-

direction, enjoyment, achievement, & restrictive conformity) that both organized the
values, and that were common across the nations under study. He suggests that various
groupings of these domains may be mutually sought after due to similar interests being
served (e.g., pro-social, restrictive conformity, & security), while other groupings are
contradictory as they would seem to serve conflicting interests (e. g., self-direction &
restrictive conformity). By providing a foundation of commonalties among cultures and
suggesting connections between values and behaviors, Schwartz points out that it should
be easier to determine actual variation in value priorities and in the relationships between
values and behaviors across cultures (Schwartz, 1994).

Schwartz is somewhat critical of Hofstede’s work. Schwartz questions how
exhaustive Hofstede’s dimensions are, the adequacy of his sample, the fact that his study
has become dated, and the equivalence of the value items across cultures. However,
despite Schwartz’s apparent dislike of Hofstede’s work, both Hofstede and Schwartz
operationalize cultural membership in terms of national origin, and both define culture in
terms of values: Furthermore, both theories of culture assume that values transcend
situations, and may guide the selection or evaluation of behaviors and events. While the
research conducted by Schwartz and colleagues is an effective attempt at theory
development, the scope of their findings and suggestions for further study are too broad to
inform the research questions of interest in the current study.

Another attempt at thinking about cultural differences from a somewhat different
perspective comes from Erez (1994). She feels that while cultural typologies such as that

developed by Hofstede are useful in terms of understanding differences between cultures,

by themselves they fail to aid in an attempt to understand culture as a moderator of the
relationship between management practices and employee behavior (Erez, 1994). Erez
proposes a new theory of cross-cultural psychology with the primary intent being to
facilitate an understanding of the relationships between those variables. Her theory
suggests that cultural values act as criteria in the evaluation of managerial practices and
their effect on one’s well being. Erez suggests that the evaluations affect goal setting that
in turn regulates behavior. The primary addition that is offered by her model is that
aspects of the self-concept such as enhancement (i.e., maintaining a positive affective
state about the self), self-efficacy (i.e., the desire to perceive oneself as competent), and
consistency (i.e., the desire to sense and experience coherence and continuity) are treated
as the link between culttu'al values and organizational stimuli. The model would suggest
that organizational stimuli (e. g., an empowerment program) are “evaluated by the self in
line with the cultural values as they are represented in the self, and with respect to the
fulfillment of the self-derived motives, which are driven by the independent and
interdependent facets of the self (Erez, 1994, pp 578)”. However, the theory of culture
put forth by Erez (1994) does not necessarily contradict Hofstede’s work. Like Hofstede,
. Erez defines culture from a cognitive standpoint, discussing cultural distinctions in terms
of shared meanings and sets of mental programs. In fact, she even employs Hofstede’s
dimensions of cultural differences to serve as her culture variables in the model. I
interpret this to mean that she feels that the dimensions themselves are not a problem, but
only that the way that the dimensions have been employed in the past has not been ideally

useful. Given this fact, I feel that the current study takes something from Erez’s model in

that the primary hypotheses of this study deal with the potential moderation effects of
culture on organizational variables. That being said however, Erez’s theory on work
behaviors as outcome variables (rather than attitudes about work), combined with the
reliance on obtaining individual level measures of the “self” in terms of cultural
dimensions, limits the applicability of this theory to the present study. This study was
focused on attitude variables and did not involve an opportunity to gather the suggested
individual level standings on cultural dimensions.

There have also been major theoretical attempts at defining more specific cultural
dimensions such as that of Triandis et al. (1988) and Triandis (1993) with
individualism/collectivism. Unlike Hofstede who suggests that individualism and
collectivism are the opposite ends on the same spectrum of dependence, Triandis claims
that individualism and collectivism are two separate dimensions (Triandis et al., 1988).
He also refers to allocentric and idiocentric people as those individuals who have
tendencies towards collectivism and individualism respectively. This thinking allows for
the fact that there exist allocentric individuals within individualistic cultures and
idiocentric people in collectivist cultures, a fact that is brought up but not dealt with by
Hofstede. However, Triandis acknowledges the fact that the multidimensional approach
to individualism and collectivism is only important when focusing on self-in-group
relationships (within culture). When in the context of examining a broad range of values
(as Hofstede did), thinking of individualism/collectivism as one dimension is appropriate
since the level of detail in the data is not as important (Triandis, et al., 1988). It should be

noted that Triandis’s thinking on individualism and collectivism is the currently accepted

model. However, while Triandis has advanced the individualism/collectivism construct
and improved on Hofstede’s scales, the resulting dimensions mirror closely the dimension
produced by Hofstede’s work, even if interpreted differently. For example, the “most
importan ” factor for differentiating between cultures (in terms of
individualism/collectivism) in the study by Triandis et al. (1988) was Family Integrity,
which was significantly rank-order correlated with the individualism/collectivism scores
obtained by Hofstede (1980). Despite some differences in approach, both Hofstede and
Triandis define the essential nature of the individualism/collectivism construct similarly
in terms of reliance on in-groups, desires for achievement, support, security, etc.
Hofstede’s set of dimensions not only include individualism and collectivism but also
offer a more comprehensive view of cultural dimensions rather than the limited theory of
Triandis and colleagues.

This is not to say that Hofstede’s work goes without contest. There have been
questions raised regarding how exhaustive the dimensions are, how generalizable the
results of one company are to entire nations, etc (a review of the critiques of Hofstede
may be found in Sondergaard, 1994; and Schwartz, 1990 & 1994). However, the same
review by Sondergaard (1994) suggested that Hofstede’s findings regarding cultural
differences are generally supported by replications. Recently, a study by House et a1
(1995) covering multiple nations and another by Peterson and Smith (1997) have found
supportive evidence for the use of Hofstede’s dimensions, suggesting that they are still a
valid measure even in more recent decades.

Having said all of that, I am not implying that Hofstede’s measures are the end-all

 

Si

lex
bOI
ma.
not

mot

10p

 

199(
that
and l

PS}’c'

c0113i

Inch“

of cross-cultural dimensions. However, twenty years and countless studies later, modern
social science has failed to arrive at a definitively more appropriate method for describing
the differences between cultures that have been shown to exist. Given this fact,
Hofstede’s dimensions are not only adequate for the purposes of this study, but also
afford the best foundation for comparison to other studies.
Boundaries of Culture

A topic of many debates in cross-cultural research has been the issue of defining
the boundaries of culture (Earley & Erez, 1997). The study will make use of country
level assumptions about the nature of culture. While I admit that both taking national
boundaries as definitions of culture and assuming cultural homogeneity within nations
may be a flawed practice (Sego, Hui, & Law, 1997), there is reason to believe that this is
not a fatal one. Peterson and Smith (1997) state that, despite their initial intentions to
move away from Hofstede’s work, all evidence, both theoretical and empirical continued
to point towards Hofstede’s dimensions as being both a stable and meaningful concept of
culture (Frank, Hofstede, & Bong, 1988; H0ppe, 1990; Van de Vliert & Van Yperen,
1996). Overall, Peterson and Smith (1997) come to two important conclusions including
that fact that they feel that county is a reasonable, if not the only real surrogate for culture
and that Hofstede’s country level data still holds meaning in the study of cross-cultural
psychology.

From a logical perspective, thinking of country as culture makes sense when one
considers the shared experience of those within the country. These shared experiences

include but are not limited to geographic location, national social economic status,

10

I I I
.i.
v .
.
I
I
I
In
I

199.

for 9.

their

bfihax

educational systems, political systems, language, typical family structure (i.e., nuclear or
extended) and predominant religions. While these shared experiences do not guarantee
that all individuals within a culture will be the same, these aspects of a country are likely
mechanisms for the development and dissemination of shared values within the borders
of a country.

Despite the discussions regarding the problems associated with defining culture
based on nation, an agreed upon alternative method has not been developed. Some
researchers such as Ronen and Shenkar (1985) have attempted to cluster cultures based
on similar country level values. While this clustering technique avoids the practice of
studying culture as it pertains to individual nations, it does not solve the problem of
differences of culture within nations, which is the primary thrust of most argMents
against the use of country as culture. Oddly, amidst all of the talk of within country
differences, it has been stated that despite increased interconnectedness of nations and
organizations around the globe, differences between national cultures have remained
relatively stable and in fact may be increasing (Ali, 1988). As has already been discussed,
Hofstede’s dimensions have held up to replication and continue to be validated in the
1990s. Based on this information, this study assumes that Hofstede’s dimensions provide
for a general framework of cultural similarities and differences across nations.

Organizational Attitude Dimensions
There are a number of job attitudes that are of importance in organizations due to
their potential relationships with employee behaviors such as performance, citizenship

behaviors, withdrawal, and turnover. Two key outcome variables that will be studied in

11

the current project are job satisfaction and perceived stress. Potential antecedents of these
outcome variables that will also be studied in this project include satisfaction with
teamwork, and attitudes about empowerment. This study proposes a model that
incorporates these four variables as well as the cultural dimensions that were previously
discussed (see Figurel). Specifically, the model depicts the suggested relationships
between both empowerment and working in teams with both the job satisfaction and
stress variables. Also, as noted, cultural dimensions (individualism/collectivism, power
distance and uncertainty) are shown as moderators of these relationships. Each attitude
variable that is included in the model is described in the following section along with
specific hypotheses regarding culture as a moderator of the relationships among the
variables.
Attitudes About Job Satisfaction

The importance of investigating antecedents of job satisfaction becomes evident
when one considers the potential outcomes of low or high job satisfaction among
employees. Relationships have been found between job satisfaction and turnover
(Crarnpton & Wagner, 1994; Hulin, Roznowski, & Hachiya, 1985), citizenship behaviors
(Organ & Ryan, 1995), absenteeism (Farrell & Starnm, 1988; Hackett & Guion, 1985),
and employee health (Begley & Czajka, 1993; O’Driscoll & Beehr, 1994), although the
evidence for the health relationships has not been convincing (Brief, Burke, George,
Robinson & Webster, 1988). Further evidence for the continued interest in job
satisfaction comes from recent organizational research aimed at gaining a better

understanding of the construct and searching for improved relationships with outcomes

12

 

 

 

Figure I

 

 

 

mmobm

 

 

3:8me 626;

 

 

 

 

 

 

 

 

 

2L+4

 

 

 

 

 

 

 

 

 

 

eosoflmmfim sop.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a: EoEoBeaEm
NE,
E
E 2:
a / i 250... E wee—83
. a \
a:
/ E \ m:
:z \ N:
8§mo>< Emerge—.00
E585 \aaaésas

 

 

 

 

 

 

3350 m 360 am «5 me .352 <

_ page

13

 

Sl

 

 

 

tha

tha

glc

c0}
jot
PK

Sat

WC

Iti

such as performance (Weiss, Nicholas, & Daus, 1999; Weiss & Cropanzano, 1996;
Simon & Carey, 1998; Brief, 1998; Brief, Butcher & Roberson, 1995; Cranny Smith, &
Stone 1992; Judge & Hulin, 1993). Despite difficulty in establishing the link between
satisfaction and performance, there is reason to believe that satisfaction may be related to
improved performance. Two meta-analyses (Iaffaldano & Muchinsky, 1985; Petty,
McGee, & Cavender, 1984) found that the correlation between general job satisfaction
and performance is somewhere in the high teens to mid-twenties. Thus, it may prove
beneficial for organizations to understand potential influences on employee job
satisfaction.

Job satisfaction in this study is operationalized by a global satisfaction measure
that attempts to tap into the attitude one holds about one’s job. While it is acknowledged
that many researchers are calling for a component approach to job satisfaction (e.g.,
global, affective, beliefs, etc.), this study assumes the position of Brief (1998) and Brief
and Roberson (1989) which contends that the global operationalization of job satisfaction
contains the aforementioned sub-components. Additional support for this definition of
job satisfaction stems from positions taken by Hulin (1991), and Locke (1976) who both
propose that general attitudes about satisfaction encompass affective components of job
satisfaction.

A point of interest that has emerged from the continued investigation of job
satisfaction is the suggestion that job satisfaction may be influenced by both the job and
work environment, as well as dispositional factors (Gerhart, 1987; Staw & Ross, 1985).

It is further suggested that dispositional factors with potential to influence job satisfaction

14

p...

 

in:

an:

indi
Or I:

abor

 

 

 

Satis
auto:
influ
Sails;
Ame
Ame.

may develop as a function of genetic inheritance or by way of social learning (Staw, Bell,
& Clausen, 1986). Based on these ideas, is it not possible that culture may play a role in
the development of dispositional factors? For example, it has been suggested that
national culture may influence the development of values (Schwartz, 1994). That makes
sense when one considers that the definition of culture assumes shared values among
individuals within a given culture. Culture has also been studied in terms of its
influences on self-concept (Gudykunst et a1, 1994), emotion (Frijda & Mesquita, 1994),
attributions (Shweder & Boume, 1982), and personality (Ip & Bond, 1995). The idea that
dispositional factors may predispose people to react to environmental factors in certain
ways, thereby influencing the resultant experienced job satisfaction is a position which
has gained support in the literature (Arvey, Bouchard, Segal, & Abraham, 1989; Gerhart,
1987; Levin & Stokes, 1989; Pulakos & Schmitt, 1983; Staw & Ross, 1985; Staw et a1,
1986; Weitz, 1952). Could these shared culturally developed dispositions predispose
individuals to react more stongly to positive or negative attitudes about working in teams
or levels of empowerment? That is, as the model may suggest, will positive attitudes
about working in teams or being empowered be more highly related to overall job
satisfaction for individuals from cultures that are predisposed to value teamwork or
autonomy? The literature suggests that there is reason to believe that culture may indeed
influence experienced job satisfaction. For example, differences in overall job
satisfaction between cultures have been reported for comparisons of Mexican to
American workers, (Slocum & Topichak, 1972), as well as comparisons of Japanese to

Americans (Lincoln, Hanada, & Olsen, 1981; Smith & Misunri, 1989). Spector and

15

Wimalasiri (1986) have even shown that cultures with similar overall job satisfaction
levels may vary with regard to satisfaction with sub-components of the job (e.g.,
supervision, pay, promotion opportunities). Studies such as these show promise for the
continued investigation of differential influences of culture on satisfaction. The current
study builds on the literature by attempting to ascertain whether cultural differences
influence satisfaction by acting as a moderator between antecedents of satisfaction, and
reported job satisfaction. While some of the antecedents of satisfaction will be discussed
shortly, I will next cover the topic of workplace stress.
Attitudes about Stress

While there still exists some disagreement over what the exact definition of stress
in the workplace is (Kahn & Byosiere, 1992), the potential effects of stress have been
fairly well documented, as will be discussed next. Of obvious interest for organizations is
the fact that stress has been found to have influence on performance and employee health
(Appley & Trumbull, 1986). Some examples of the potential psychological responses to
stress have been shown to be anxiety (Caplan & Jones, 1975), depression (LaRocco,
House, & French, 1980; Caplan & Jones, 1975) job dissatisfaction (Ganster, Fusilier, &
Mayes, 1986; Ivancevich, Matteson, & Preston, 1982), and decreased health (Cooper &
Roden, 1985; Williams & Stout, 1985; Rhodewalt & Agustodottir, 1984). While these
stress responses may create costs for organizations (Jones, 1984), they primarily affect the
individual. Of perhaps even more importance to organizations is the fact that stress
responses have been shown to relate to behavioral responses as well. Some examples of

these behavioral responses include increased absenteeism (Jackson, 1983), increased

16

accident rates (Jones, 1984; Colquhoren, 1976), counterproductive behaviors (e.g.,
damaging property, stealing, slacking off) (Mangione & Quinn, 1975), and perhaps most
importantly, influence on job performance (although results have been mixed) (Weiss &
Cropanzano, 1996; Lazarus, 1991; Kaufman & Beehr, 1986; Mossholder, Bedeian, &
Armenakis, 1982; Buck, 1972).

While it is evident that workplace stress and the potential responses to stress may
be of importance to organizations and employees, what is not so obvious is when
potential stressors are realized by a person or group of people as actual stress. J ex and
Beehr (1991) make an important distinction between stressors and strain. They suggest
that stressors are conditions at work that require an adaptive response while strains are
aversive reactions to a stressor. This separation of concepts allows for the scenario that
not all individuals will react to potentially stressful events in the same way, individuals
must perceive a stressor as a negative challenge or threat in order for actual stress (or
strain) to be experienced (J ex & Beehr, 1991; Frese & Zapf, 1988). Based on this idea, it
has been suggested that stress research should more adequately address potential
moderators (e. g., personal characteristics, social development, and genetic characteristics)
of the relationships between stressors and strains. (Kahn & Byosiere, 1992). The
proposed model suggests that one such potential moderator of those relationships may be
national culture.

Research findings show some support for the idea that responses to stressors may
vary by culture. For example, using Hofstede’s cultural dimensions, Peterson and Smith

(1995) found that role stress varied more by culture than by either demographics or

17

organizational factors. Other studies have also shown that cultures may differ with regard
to the extent to which a potential stressor is perceived as stress. Cooper (1984) found
variance between managers in ten countries on their “pressure profiles” or primary stress
stimulus. For example, Swedish managers reported home/work conflicts to be
particularly troubling, US managers named power issues as a primary stressor, and
Japanese managers were most bothered by work overload (Cooper, 1984). Bhagat, et
al.(1994) conducted a 7 nation study and also found that the stressor to strain relationship
depended at least partially on culture, specifically on collectivist nations’ tendencies to
experience less perceived stress. This finding is explained by the nature of individualist
and collectivist cultures. Triandis et a1. (1988) discuss the tendencies of individuals
within collectivist cultures to have more social cohesion, more social support, more in-
group harmony, and less insecurity than individualists. Individuals fiom Individualist
cultures not only lack those tendencies of people in collectivist nations, but also tend to
emphasize self-reliance and competition which increases insecurities about seeking social
support as a means of coping (T riandis et al., 1988). Based on this information, the
following hypothesis is proposed:

H1: Individuals fiom cultures high on collectivism will report less perceived

stress than individuals from cultures high on individualism.

There have been other large and small-scale studies that further support the
concept of cultural differences in perceived stress, most of which involved two nation
comparisons. For example, a comparison of police in Germany to those in Ireland found

differences regarding external versus internal sources of stress, as well as differing styles

l8

of coping between the two nations (Kirkcaldy, Brown, & Cooper, 1994). Other examples
of dual nation comparisons have shown that types of prominent stressors and coping
styles differ between British and German managers (Kirkcaldy & Cooper, 1992), and that
interdependency in an independence oriented culture leads to increased stress (Cross,
1995). The primary thrust of these studies is that aspects of culture such as the amount of
social support one can expect and values such as self reliance appear to moderate the
relationships between potential stressors and experienced stress or strain. This study
suggests that attitudes about aspects of the job including working in teams and levels of
empowerment are potential stressors and therefore may be related to reported experienced
stress as I will discuss shortly. The proposed model incorporates those relationships and
further suggests that dimensions of culture moderate those relationships.
Attitudes about Teams

Teams have become embedded so strongly in organizations that the very nature of
how work is performed in an organization is dependent on team structures and how
people behave within those structures. Many researchers agree that workgroups and
teams have become a pervasive and important aspect of organizations (Guzzo & Shea,
1992; Magjuka & Baldwin, 1991; Hackrnan, 1990). Given the importance of teams in the
workplace, researchers have sought to understand both the nature of, and the effectiveness
of work teams (Hackrnan, 1992). However, despite the increased attention given to
research on teams, there has been a lack of focus on how the team-based nature of work is
perceived by different cultures. Most researchers agree that as organizations continue to

grow into global corporations, an understanding of cultural influences on teams will

19

become increasingly important (Guzzo & Salas, 1995). The model presented in this study
proposes that there are relationships between attitudes about working in teams and both
overall job satisfaction and stress. It further suggests that dimensions of culture will
moderate these relationships.

One outcome of the research that has been conducted on teams is the finding that
aside from just performance, working in teams may influence attitudes about satisfaction -
or perceived stress on the job (Hackrnan, 1987; Bass, 1982). Support for a relationship
between teamwork and satisfaction has been suggested by research defining satisfaction
as a component of team effectiveness which may vary depending on the nature of the
team (Sundstrom, De Meuse, & Futrell, 1990; Hackrnan, 1987; Wall, Kemp, Jackson, &
Clegg, 1986; Gladstein, 1984). For example, a team that fulfills all of its production
requirements may still lead to dissatisfaction among certain team members if they feel
that others in the group were not doing their share of the work. Research has found that
working in teams may have both positive and negative relationships with job satisfaction
depending on a number of factors ranging from contextual (e. g., group composition,
group task) to dispositional (Lindermen, 1998). Of more specific interest for the purposes
of this study are examples that show that experiences of working in teams relate
differently to levels of satisfaction as a function of affiliation with certain cultures. For
example, Hui and Yee (1999) found that collectivist tendencies in Hong Kong were
related to increased teamwork/job-satisfaction relationships. Studies such as this suggest
that the cultural dimension of individualism/collectivism may moderate the relationship

between teamwork and satisfaction.

20

According to Hofstede, cultures high on individualism will tend to focus on
individual self-interest, even when operating in a group. Individual rewards, status, and
self-reliance are important within individualistic cultures. Collectivist cultures on the
other hand will tend to orient more towards group benefit and support structures. In these
cultures, self-interests are subordinate to group success (Hofstede, 1980). That being
said, researchers have pointed out that interdependency (Salas, Dickinson, Converse, &
Tannenbaum, 1992; Tannenbaum, Beard, & Salas, 1992), responsibility for other’s work
(Kiggundu, 1983), and group level rewards (Shea & Guzzo, 1987) are all key aspects of
working in teams. These commonalties of group work fit more closely with Hofstede’s
definition of collectivism than with individualism. Furthermore, a preference for group
work has been shown to be related to higher satisfaction derived fi'om working in groups
(Fried & Ferris, 1987; Cummings, 1981; Hackrnan & Oldham, 1980).

Taken as a whole, these findings suggest that individuals from collectivist cultures
should find working in teams to be a more rewarding and satisfying process than people
from individualistic cultures who may find teams to be a stressful distraction from
individual pursuits of recognition and reward. Furthermore, increased satisfaction with
the experience of working in teams (i.e. workers that are satisfied with sharing
responsibility, and interdependency) is likely to have more of an influence on overall job
satisfaction for individuals from cultures that value successful collective efforts.
Therefore, the following hypotheses are proposed:

H2: Individuals fiom cultures high on collectivism will report higher levels of

satisfaction with teamwork than individuals fiom cultures high on individualism.

21

H3: The relationship between satisfaction with teamwork and job satisfaction will
be stronger and more positive for individuals from cultures that are high on
collectivism.

There is also evidence that working in teams may produce potential stressors that
employees must deal with (see Bass 1982; Morgan & Lassiter, 1992; for reviews of this
literature). I previously mentioned research that pointed out that interdependence,
responsibility for others’ work, and group rewards are commonly associated with working
in teams. It may be that these common aspects of working in teams are likely to be more
appealing and satisfying for individuals fiom collectivist cultures who orient more
towards group benefit and support. However, what might be the effect on the relationship
between team satisfaction and stress for individuals from cultures that are high on
collectivism if the cohesiveness, performance, interpersonal interactions, etc. break down
within a team? In a summary of empirical evidence, Ivancevich and Matteson (1980)
suggest that group dissatisfaction is a potential antecedent of stress. The lack of
satisfaction with team membership may be a more significant factor in the amount of
reported stress for cultures that highly value group success, support, and a sense of
belonging. Therefore, individuals from cultures high on collectivism that are dissatisfied
with team work will also report high levels of stress, while those satisfied with teamwork
will report moderate to low levels of stress.

H4: The relationship between satisfaction with teamwork and stress will be

stronger and more negative for cultures that are high on collectivism.

It is acknowledged that this hypothesis predicts a more negative relationship

22

heme“ ’
while H3
satisfactic
h_xpothesi
satisfactic
Al
dimensior
variables.
stable and
while ind
variability
bCN’een 1
avoidancr
may prod
Producing
(Hacknm
POint our
highly inI
0031111011
1997)_ 1
31110an (

interdepe

between satisfaction with teamwork and stress for individuals from collectivist cultures
while H3 predicts a positive relationship between satisfaction with teamwork and job
satisfaction for the same individuals. The opposite direction of the relationships
hypothesized in H3 and H4 does not present a problem due to the finding that stress and
satisfaction are generally uncorrelated (Brief, 1998).

Along with individualism/collectivism, this study also proposes that the cultural
dimension of uncertainty avoidance moderates relationships between work attitude
variables. Individuals fi'om cultures that are high in uncertainty avoidance tend to prefer
stable and predictable situations, clearly defined roles, and straightforward directions,
while individuals from cultures low on uncertainty avoidance tend to be less bothered by
variability and ambiguity (T riandis, 1990). Might the previously discussed relationships
between working in teams and stress or satisfaction also be moderated by the uncertainty
avoidance dimension? It has been shown that working in teams provides situations which
may produce uncertainty for employees. For example, teamwork is sometimes prone to
producing interpersonal conflicts and desires to discontinue working in the group
(Hackrnan, 1980; Alderfer, 1977). Also, Cannon-Bowers, Oser, and Flanagan (1992)
point out how frequently work teams are at the same time both highly autonomous and
highly interdependent. The potential conflict that arises from these dual characteristics of
common work teams has not gone unnoticed in the literature (J anz, Colquitt, & Noe,
1997). This conflict may place strain on employees who have difficulty balancing the
autonomous nature of their team (or their role in the team) with the demand for aspects of

interdependency such as sharing information or being responsible for other’s work.

23

 

 

 

There is 2115‘
overload. «1'
The
situations v
due to conf
These are a
uncenainn'
H5:
5311
um:
Dis
overall job
conflict, to
10 dissatisf
indixiduals

suggested I

 

km“Work
closely rel
proposed:
H6
be

3V

There is also evidence that teamwork produces multiple potential stressors such as role
overload, ambiguity, and interpersonal contact (Morgan & Bowers, 1995).

These combined aspects of teamwork suggest that working in teams may create
situations where roles are less well-defined, interpersonal interactions are less predictable
due to conflict, and tensions between autonomy and interdependency create confusion.
These are all scenarios which people fiom cultures with tendencies towards avoiding
uncertainty may not find particularly enjoyable. Therefore:

H5: Individuals from cultures high on uncertainty avoidance will report less

satisfaction with teamwork than individuals from cultures that are low on

uncertainty avoidance.

Dissatisfaction with working in one’s team is likely to influence an employee’s
overall job satisfaction (Hackrnan, 1987). However, dealing with the interpersonal
conflict, role ambiguities, overload, desires to quit a team, and other factors that may lead
to dissatisfaction with teamwork are likely to have a larger impact on job satisfaction for
individuals from cultures that prefer stability and clearly defined roles. It will be
suggested that cultures high on uncertainty avoidance are more apt to find coping with
teamwork and dealing with teams that are unsatisfying to be more stressful, and more
closely related to decreased job satisfaction. Hence, the following hypotheses are
proposed:

H6: The relationship between satisfaction with teamwork and job satisfaction will

be more strong for individuals from cultures that are high on uncertainty

avoidance than for individuals from cultures that are low on uncertainty

24

Worker en
(Thomas 6
meaningfi
monitor, 2
employee
1994; Sp:
Perfonna
e1"idence
more 33‘:
recently
higher It
emPOVH
some ex
implem
about “

debate ‘

avoidance.

H7: The relationship between satisfaction with teamwork and stress will be more

strong and negative for individuals from cultures high on uncertainty avoidance

that for individuals from cultures low on uncertainty avoidance.
Attitudes about Empowerment

The final piece of the model that has yet to be discussed is empowerment.
Worker empowerment remains a topic of interest, and one of importance in practice
(Thomas & Velthouse, 1990; Kanter, 1989). Allowing employees to oversee a
meaningful task while also increasing their responsibility, autonomy, ability to self-
monitor, and to make decisions has been purported to have a positive influence on
employee proactivity, satisfaction and productivity (Spreitzer, 1995; Thomas & Tymon,
1994; Spreitzer, Kizilos, & Nason, 1997; Hyatt & Ruddy, 1997). While links to
performance have been difficult to establish (Griffen, Welsh, & Moorehead 1981),
evidence of relationships between empowerment with job satisfaction and stress has been
more easily discovered. Kirkrnan and Rosen (1999), and Koberg et a1. (1999) have
recently found evidence that empowered employees and empowered work teams report
higher levels of overall job satisfaction. It has also been shown that a lack of
empowerment was related to increased dissatisfaction (Caplan et al., 1984). There is also
some evidence that certain aspects of empowerment (e. g., decisions about how to
implement change) may be more highly related to satisfaction than others (e.g., decisions
about whether or not to implement a change) (Sagie & Koslowsky, 1994). Despite some

debate over the components of empowerment that are related to worker satisfaction, even

25

workers

mp0“

increas

M6811

increa:

Suther

work:

ernpo

researchers who disagree with each other in the literature have agreed that empowerment
and satisfaction are somewhat related (Locke, Schweiger, & Latham, 1984; Sashkin,
1984)

While it appears to be the case that empowered workers tend to be more satisfied
workers, there is some debate in the literature regarding the relationship between
empowerment and the perceived stress levels of employees. Suggesting that the
increased workloads, decision latitude, responsibility, and other aspects of empowerment
are a negative influence on employees’ attitudes, some recent findings conclude that
increased empowerment leads to increases in perceived stress (Mishra & Spreitzer, 1998;
Sutherland, Fogerty & Pithers, 1995). However, others have shown that empowered
workers report lower levels of stress (Jackson, 1983), even when compared to their less-
ernpowered co-workers of equal status in the same organization (French & Caplan, 1972).

Researchers and practitioners may find it useful to know what types of
environments and people will react positively to empowerment programs. It has been
suggested that attitudes regarding participation are prominently influenced by developed
norms and experience with participation programs (French, Kay, & Meyer, 1966).
Culture may play a role in developing norms and dictating the likelihood that members of
the culture have had contact with empowerment programs. Thus, the current study
suggests that three of Hofstede’s dimensions of culture (power distance,
individualism/collectivism, and uncertainty avoidance) may moderate the relationships
between empowerment and satisfaction, and empowerment and stress. A discussion of

each of these dimensions and rationale for hypotheses involving the influence of cultural

26

dimensions as moderators follows.

Hofstede’s power distance dimension suggests that cultures that are high in power
distance will tend to have more autocratic and authoritative styles of leadership.
Employees are expected to follow established patterns of communication and are heavily
dependent on supervision and the organization for instruction. Individuals from low
power distance nations on the other hand are less dependent on supervision and a chain of
command. Employees are treated more as equals and are more involved in decision
making processes, and determining the exact nature of their own work (Hofstede, 1980).
Research has shown that cultural groups that vary on the power distance dimension
perceive employee participation in decision-making differently. The majority of research
on this topic concludes that individuals from cultures that are low in power distance are
likely to prefer, or benefit from increased participation due to flattened organizational
structures and a low tolerance for inequalities in power (Erez & Barley, 1987; Erez, 1986;
Erez & Arad, 1986; Rodrigues, 1990). It is suggested that individuals from high power
distance cultures will not benefit as much, or will not prefer increased participation due to
high tolerances for inequality, a tendency to rely on superiors for direction, and a
tendency to accept directions or goals, even when opposed to them (Erez & Barley, 1987;
Rodrigues, 1990). In fact, a recent study by Robert, Probst, Martocchio, Drasgow, and
Lawler (2000) involving four countries (U .S., Mexico, India, and Poland) found that the
relationship between empowerment and satisfaction differed between countries. Without
specifically incorporating cultural dimension scores in the analysis of their model (they

simply looked at the country standings on the power distance dimension), they assert that

27

the differences may be due to different levels of power distance. Their findings provide
further support that higher power distance may be related to decreased or negative
relationships between empowerment and satisfaction.

Given that the empowerment items used in this study measure the presence of or
lack of empowerment rather than actual satisfaction with empowerment, the following
hypothesis is suggested:

H8: The relationship between empowerment and job satisfaction will be more

positive for individuals from cultures that are low on power distance than for

individuals fiom cultures that are high on power distance.

Research has also shown that power distance is closely linked to role stresses.
Peterson and Smith (1995) found that those in high power distance nations reported
higher levels of role overload than low power distance nations. They state that this is
likely a result of overload due to work events or role structure stresses. This finding is in
line with Goffrnan (1961) who suggested that organizations within high power distance
nations are likely to incorporate dysfunctional role structures that are too rigid and serve
to compound role stresses. Empowerment typically is considered to involve most of the
following: increased autonomy, power to make decisions, access to information used to
make decisions, goal setting, problem solving, planning for change, and change
implementation (Spreitzer, 1995; Sashkin, 1984). By the very nature of its general
definition, empowerment is likely to increase the demands placed on the roles of the
workers. Following from findings such as Peterson and Smith(1995) and Goffman

(1961), nations high on power distance are likely to have organizational structures and

28

employee roles that are inadequately designed for coping with the increased workload.
Therefore, the increased overload of responsibility would create more potential stressors
for those used to Operating in a more controlled, passive environment.

H9: The relationship between empowerment and stress will be higher and more

positive for individuals from cultures that are high on power distance.

There is also reason to believe that perceptions of empowerment may differ as a
function of the previously discussed cultural dimension of individualism/collectivism.
People from individualistic cultures emphasize the benefits of autonomy, personal
freedom by way of participation in decision making, decentralized decision-making, and
employee centered incentive systems (Earley & Gibson, 1998). It has been suggested that
individuals fiom individualistic cultures should prefer a more participative work
environment as it increases their opportunities to promote self-interests. This idea makes
sense when comparing the basic components of empowerment (autonomy, fieedom to set
goals, implementing changes, decision making power, etc) with the basic desires of
individualists (autonomy, freedom to succeed, ability to manage one’s own work life,
etc.). The opportunities allotted by empowerment appear to coincide with many aspects of
individualistic preferences. Offering some support, Near (1986) found that freedom on
the job was related to higher levels of satisfaction for US workers high on individualism
(Hofstede, 1980), but was not related to satisfaction for Japanese workers high on
collectivism (Hofstede, 1980). Lawler (1986) has also pointed out that individual level
rewards are a key component of empowerment in organizations. Following from this

discussion, this study proposes that the increased autonomy, power in decision making,

29

opportunities to seek and receive rewards, etc. would be more pleasing to those in
cultures with individualistic tendencies.

H10: The relationship between empowerment and satisfaction will be more

positive for individuals from cultures that are high on individualism.

Attitudes about empowerment may also vary across cultures depending on cultural
tendencies towards uncertainty avoidance. As stated earlier, empowerment in
organizations typically involves a situation where employees are given more
responsibility for making and implementing changes, power to make decisions, freedom
to set goals, and in general a freedom to manage their own role within the organization. It
has also been discussed that reported stressful reactions to increased empowerment vary
from study to study. Might culture explain some of this variance? Peterson and Smith,
(1995) suggest that work events carry with them a certain level of ambiguity. Work can
also create role conflicts by introducing previously unattended to stimuli to an employee
that must be dealt with. The general definition of empowerment changes forms slightly
fiom article to article. However, the general thrust of each definition makes it clear that
empowerment in the work place is likely to provide more responsibility for employees to
manage and new challenges to be dealt with. According to Hofstede (1980), individuals
from cultures high on uncertainty avoidance are likely to react more negatively to, and
possibly perceive more stress as a result of increased and sometimes vague role
responsibilities.

H11: The relationship between empowerment and job satisfaction will be lower

for individuals from cultures that are high on uncertainty avoidance than for those

30

low on uncertainty avoidance.

H12: The relationship between empowerment and stress will be stronger and more

positive for individuals from cultures that are high on uncertainty avoidance.

Method

SAanls

Participants were salaried employees of a global manufacturing organization.
Data were collected anonymously from employees during internal census surveys in
1996, 1998, and 1999. It should be noted that the 1996 data were collected using paper
and pencil questionnaires while the 1998 and 1999 data were collected using computers.
Responses to the 1996 questionnaire were obtained from 53,298 individual respondents,
from a total of 28 countries. Responses to thel998 questionnaire totaled 63,277
employees from 28 different countries. Responses to the 1998 questionnaire were
obtained fi'om 62,345 employees fiom 28 different countries. For a list of the included
nations, see Table 1.
Job Attitude Measure

All questionnaires were developed internally by the organization. Item response
format for this survey was a Likert-type scale, consisting of a five-option answer format
ranging from “Strongly disagree” to “Strongly Agree”. The organization developed the
response option scales in the opposite direction than is normally found such that a
response of “I” meant, “Strongly agree”, and “5” meant, “Strongly disagree.” These
items were reverse coded in order to fit with traditional methods, and to ease the

interpretation of analyses. The 1996, 1998, and 1999 questionnaires consisted of 34, 40,

31

Om
ev
on
we
m m
be
K
me
E.
ow
mm
mm
mm
om
ev

88m Emm>eoo=oo
\EmzuaeEeE

_w
we
am
it
oe
mm
we
mm
M:
an
me
me
:
em
av

Boom 3585 SEE

Nw
Ne
mm
ow
N_ _
me
em
on
mm
3
en
we
on
3
ex

28m

8529a bear—8:3

5 _.N
Nem
ow
Sum

emw€
mme
34
ae
wmm.m
nee.—
hmm
be
_m_ _
Due

.32 com Z . wag 5m 2 ema hoe Z

Rmfi
ewv
mm

mm

3”
mm~€
men
me
me
befim
meo.N
2m
3
New;
ewe

9» _.v
vmm
em

mm
omm.m
vow
_N

NV
mvo.~
bee;
vow
ea

c 5.—
new

852 .2
a3 .3
use: .2
as: .2
885 .:
gene .2
85m a
Baez .w
E889 s
ego 6
Sam .m
8338 .v
253 .m
2353. .N
83.32 ._

3580

 

 

 

$56585 9036.3: co mouoom 28 $5550 woes—05

_ 033.

32

 

Q
_a
aw
om
C
we
K
G
we
em
ae
an
ow

28w Emm>eoe=o0
Ham—«33%,:

_w
ov
mm
ve
wm
vm
mm
mm
av
me
2“
mm
wm

Boom 8ng 330a

ex.
ev
m m
ve
ae
wm
am
ew
av
vo—
om
av
mm

28w
85222 bawdy—8:3

_vm
N_a.mw
:oK
own.—
v3
vb

vv
omw
wN
amm
mm
o:
as

aaa_ 8e Z

avm
vwa.mm
mgfi
mae
wmm
ve

_e
_mw
em
hwm
av
m:
::

waa_ .8.“ Z

evN
mwo.mm
oem.v
m

amm

hm

mm

vow

whm
mm
N2
am

eaa— cow Z

Seq—Noe“; .wm

<wD SN

MD .em

ages... .2
asap 3
efifiafim .8
883m .8.
saw .a

at? 58m .8
awaaa .2
.835 Z .w _
egg 302 .:
$820502 .2

.9550

 

8:528 _ 2.3

33

and 40 core items respectively, and were originally designed to tap into nine different
core dimensions (job and company (job satisfaction), stress, workgroup and team,
empowerment, reward and recognition, training & development, supervision, workload,
and quality). The total number of items differs due to new items being added to the same
scales between the 1996 and 1998 administrations of the questionnaire. However, factor
analytic techniques revealed that the nine dimensions did not hold up across cultures
(Horvath, Ryan, & Ployhart, 1998). In a report to the organization that administered the
questionnaire, Horvath, Ryan, and Ployhart (1998) discuss a confirmatory factor analysis
of the 1998 questionnaire across 17 different nations that revealed that certain items
loaded consistently on four dimensions across cultures.

The four factors that appeared were satisfaction/ feeling valued, stress, satisfaction
with workgroup and team, and supervision/empowerment. Ten items did not load
consistently on any factors, most likely due to differences in meaning of these items
across cultures (Horvath, Ryan, & Ployhart, 1998). This study used the clusters of items
that formed the four factors as measures of the organizational attitude variables that have
been discussed in this paper. The benefit of using these factor consistent scales is that it
allows for discussion and interpretation of questionnaire items and their relationships
across cultures. It should be noted that this study used only the empowerment items from
the supervision/empowerment scale. These items tap into attitudes about empowerment
rather than addressing attitudes about supervisors.

Items for the four dimensions (for 1996, 1998, and 1999) may be found in

Appendix A. Each of the scales will be discussed next. In the following section, please

34

note that all references to the construct validity refer to analysis conducted for a report to
the organization by Horvath, Ployhart, and Ryan (1998). The analyses that they
conducted made use of a sample of 177 college business students at a large mid-westem
university. They could not use the actual employees of an organization due to the
necessity of administering a battery of additional tests, combined with the fact that the
original organizational questionnaire was anonymous.

The job satisfaction scale consists of five items that measure overall satisfaction
with the organization and satisfaction with feeling valued by the organization.

The stress scale measures the extent to which employees report experiencing

 

excessive amounts of workplace stress. The 1996 stress scale consisted of four items that
focused on the level of stress and workload being experienced. The 1998 and 1999 stress
scales included three of those same four items but also added six more items that were
intended to assess how well experienced workplace stress was being coped with, as well
as additional workload items. The stress scale items in Appendix A are denoted as
belonging to only the 1996 scale, the 1998 and 1999 scale, and to all scales. The 1996
stress scale was correlated (r = .55) with the stress in general scale that was developed by
Smith et a1. (1992).

The satisfaction with teamwork scale in 1996 consisted of seven items that
attempt to measure levels of satisfaction with team cooperation and effectiveness. Two
of these items were dropped for the 1998 and 1999 scale. Evidence of convergent
validity for a subset of these items is indicated by correlations with measures of team

related constructs such as the Team Orientation Questionnaire (r = .45) (Isabella &

35

Waddock, 1994), sub-factors of the Work Group Characteristics scale (Ruddy & Hyatt,
1997) including work group support (r = .65), work group confidence (r = .65), and
interpersonal work group processes (r = .53).

The empgwerment scale consists of three items. These items were designed to
measure the extent to which employees felt that they had autonomy, decision-making
power, and participation in setting objectives, and the opportunity to self-evaluate.
Cultural Dimension Measurement

In order to assess differences between cultures, this study used three of the four
cultural dimensions developed by Geert Hofstede: individualism/collectivism, power
distance, and uncertainty avoidance. The development of these dimensions arose fiom
the analysis of survey data collected twice within a large multinational organization. The
data was collected over the span of 1967 to 1969, as well as from 1971 to 1973, with a
total of approximately 117,000 responses to the questionnaire. The cross-cultural
analysis of this was conducted on a total of 40 different countries using only those with at
least fifty responses. Drawing from early works describing the dimensions of power
distance (e.g, Mulder, Ritserna Van Eek, & De Jong, 1971), uncertainty avoidance (e.g.,
Cyert & March, 1963), and individualism/collectivism (e.g., Etzioni, 1975; Kluckhohn &
Strodtbeck, 1961; Parsons & Shils, 1951) Hofstede selected items fiom the questionnaire
that had content similar to the proposed dimensions of culture.

Power Distance: The PDI ranges between 0 and 100 with high scores meaning
more power distance. Country scores in Hofstede’s study ranged fiom a low score of 11

(for Austria) to 94 (for Philippines) with a mean country score of 52 (n = 40) (Hofstede,

36

1980). In the current study, the included countries’ PDI scores ranged fi‘om 11 to 81 with
a mean of 47.4 and a standard deviation of 19.3. PDI scores for countries included in this
study are located in Table 1.

Uncertaim/ Avoiding: The UAI has a range of values from 8, the least amount of
uncertainty avoidance (Singapore), to 112, the most uncertainty avoidance (Greece). The
mean score on the UAI in Hofstede’s sample was 64 (n = 40). The range of scores on the
UAI for countries included in this study is 23 to 112, with a mean of 63.8 and a standard
deviation of 23. 1. UAI scores for countries included in this study are located in Table l.

mividualism/Collectivism: scores on the IDV index fall within a range of 0 to
100 where higher scores mean more individualism. In Hofstede’s sample, an actual range
of values from 12 (Venezuela) to 91 (United States) was obtained with a mean
individualism/collectivism score of 51. In the current study, country scores on the IDV
ranged from 12 to 91, with a mean of 58.1 and a standard deviation of 22.9. IDV scores
for countries included in this study are located in Table 1.

Results

It should once again be noted that due to the anonymous nature of the data
obtained from the questionnaire, important links between administrations of the survey
could not be made. This fact made it impossible to track responses over time, resulting in
an inability to analyze the data across the three administration periods. Therefore, means
and standard deviations of job satisfaction, satisfaction with working in teams,
empowerment, and stress are provided for each country for the years 1996, 1998, and

1999 in Tables 2, 3, and 4 respectively. Individual level correlations of those same

37

variables across countries, as well as the reliabilities of the scales may be found in Table
5 for 1996 data, Table 6 for 1998 data, and Table 7 for 1999 data.

The following section will discuss the outcome of the analyses that tested both the
direct effect and moderation hypotheses that are under investigation in this study.
However, preceding the results of this study is an overview of Hierarchical Linear
Modeling (HLM) and the models used to test the actual data.

As stated earlier, the data in this study were analyzed using HLM techniques. The
hierarchical nature of HLM implies that individuals may be studied not only as
individuals, but also as collective groups ranging across many levels (i.e., teams,
departments, organizations, and countries). For the purposes of discussing the analyses,
the data collected at the individual level will also be referred to as level-l data later in this
section. This data includes each employee’s individual ratings of job satisfaction,
satisfaction with teamwork, empowerment, and stress. The data that has been aggregated
to a higher group level (in this study the mean country level scores on power distance,
uncertainty avoidance, and individualism/collectivism based on Hofstede’s research) will
be referred to as level-2 data.

HLM is designed to investigate models that involve relationships between both
individual and group level data. While ordinary-least-squares (OLS) regression analyses
could also be employed, HLM offers two primary advantages. The first of these
advantages is highlighted by the fact that HLM techniques model both the individual and
group level variance. This allows for the inclusion of potentially meaningful information

regarding the levels of independence/interdependence of individuals’ responses to the

38

ew. EN
ov. vs."
_a. wmd
aw. me
.a. and
6. ma.—
we. va
ww. v_.m
ma. afim
ew. _m.m
en. vmw
me. EN
oa. haw
Qw M
EoEoBanm

we. o_.m
ae. Sum
on. mm.~
_e. eN.~
ve. ZNN
em. aa._
mm. wm.m
he. m3...
on. vv.~
Ne. o_.m
oe. ZN
we. 2d
2.. mm.~
Ow war
€958on

fies coeoflmemw

be.
me.
me.
em.
we.
we.
mm.

me.

am.

ee.
we.

Ow

mmobw

_w.N
_w.N
vw.N
ee.N
mew
vv.N
vw.N
ew.N
wa.N
med
end
aw.N
_w.N

oa.
ae.
an.
en.
we.
wm.
mm.
ma.
oa.
an.
be.
ow.
mw.

Ow

med
mm.~
ov.~
new
mv.N
ca;
and
hvw
new
hvw
end
NON
med

X

:oeomwmeaw e2.

am

mm
ommfi
vav
a

Nv
mvo.N
nee;
vow
an

o 8;
new

case .2
£5 .2
8020 .:
€an .2
88am .a
23:5 .m
gage .

«350 .

Seam .

h
e
mufim .m
v
«525. .w

N

e353 .

@3232 ._

.3560

 

 

was 82 ac 958 a 85§> “am a 2325a Eegm e5 was:

 

N 2.5

39

.a. v_d
ww. _vd
ww. ee._
dw. edd
ww. de
ea; dwd
wa. owd
ad; we;
oa. emd
wo._ w md
ma. vd
we. :d
oa. w_d
aw. med
Ow M
EoEoBanm

ae. a_d
we. Zvd
am. _e._
de. _a._
ae. ew._
mw. omd
ee. odd
ww. aeg
ee. _wd
ee. wed
ee. vmd
dw. aod
ve. aod
we. dvd
Ow M.
x8388...

as, 835%

mmobw

dwd
—ad
dad
ewd
wvd
_ad
wed
_wd
dad
wed
_ad
dmd
vwd
dad

da.
ew.
oe.
we.
_e.
da.
ae.
mv.
ca.
ww.
va.
ve.
dw.
we.

Ow

ewd
wwd
ee._
_md

_a._

ved
owd
av.—
med
ewd
wed
vdd
wdd
wed

X

ceeoawmemw e2.

mwodm
oewv

aww
em
dm
voe

wed
wd

dm _
am
mv_.v
vwd

<3
. v5
225:

533. .

ecu—555w
:3on
Exam

moE< 530w
Ema—om
ean Z
BEEN 3oz
menu—6502
03qu

5%..

3:80

 

3:588 d 035.

40

deem/V. bwcobw n w 23>» oohwema bwcobw u _ .83 £5 E 85 80: ammo—a Aw I Z ”gouge 838$
338 .anw :o momcoamou no women 0.8 EoEoBanm e5 J—uoaeaoe fies :oeomwmemw .mmobw .eoeoflmemw eon how 333m .802

 

ww. eod ae. Z Zd

Qw M Ow M
x8383...

EoEoBOQEm 5m? :oeoflmemw

ve.

Qw

mmobw

dwd

_a. evd evd m_o=No=o> .wd

Ow X Z

coeoawmeaw e2. .3580

 

325:8 d 033.

41

ew. ovd
ew. 2d
vo; _wd
aw. _dd
vw. e_d
ae. eod
ve. dvd
wa. d_d
ew. 2d
ew. vdd
me. wdd
ew. vmd
ma. mmd
Ow M.
EQEBBOQEm

dw. w _ d
ce. 3d
ae. wmd
de. e_d
we. w _d
ae. w_d
mw. wvd
oe. 2d
oe. ovd
we. a_d
av. w_d
oe. edd
ae. w md
Ow M
{03:83.
55 coeomwmemw

ew.

va.
we.
dw.
ww.
ew.
5;
ww.
ow.
_w.

ww.

Ow

«mobw

Odd
wvd
wow
eed
ved
oed
wow
wwd
vfim
Zed
wed
ead
ofim

we. awd
we. ddd
ea. wwd
_w. vvd
ae. w md
ae. a_d
ee. wmd
ea. ovd
ew. vvd
ew. vvd
ve. ddd
da. awd
vw. eed
Ow WM

:oeoflmeaw e2.

md

dd

vm
wwd.v
dew
we—
we
ewwd
weed
o—w
_w
dew.—
ewe

eca—ob .m—
amefi d_
0880 .2
.8250 .2
35cm .a
BREE .w
#8859 .e
3230 .e
.fiflm .w
Edam .2.
«Bang .m
«28.23. d
25ng .—

.9560

 

 

 

man 32 ac 958 3 33%? ram co gag;

m 2.5

42

da. eod
Zw. edd
oa. end
we. d_d
ew. a_.d
wa. w_d
wa. d_.d
oa. _v.d
ww. Sud
ee. wfid
ww. wvd
ee. wdd
aw. wdd
ma. oed
Ow M

EoEoBanm

oe. v_.d
ve. _m.d
me. w _.d
aw. ea.—
de. ma.—
ee. wdd
me. vwd
ww. ewd
we. vwd
we. emd
we. ovd
_e. wdd
ae. w _.d
ae. dwd
Ow M
ioBEMoH

a? Seesaw

ma.
ow.
_e.
ee.
we.
mw.
ww.
ve.

dw.

5.

de.

ee.

Ow

mmobw

aed
dod
wwd
_wd
awd
owd
ewd
ewd
dwd
dad
ead
eed
evd
ed.m

|><

ea.
aw.
_e.
de.
ae.
3.—
ew.
ow.
aw.
ww.
va.
ee.
ow.
ww.

Ow

vvd
wed
de
wdd
aa._
wwd
med
ddd
aed
wwd
wed
dmd
edd
vwd

M

:eeomememw e2.

vwa.mm
dee.w
wae
wmw
ve

_e
_mw
ed
ewd
av

m _ _
S—
eew.w
ewv

<wD

v5
ens—mag.
533E
ecu—uofiesw
coeocsw
Emaw

mommy. 550w
awaken
eeEoZ
ecu—mod >>02
mesa—conga Z
92on

=25.

.ed
.ed
.wd
.vd
.Md
.dd
. _d
dd
.a_
.w_
.e~
e—
.w_
.v_

.9550

 

Essence m 035.

43

68w< bweobw u w 233 oouwmma bwcobw u _ See 25 E 35 80: SEE Aw I _ "meoeao 6333
8—8.... Emoaw co momeoamou no woman 08 EoEoBanm Ea £83.33. fies geomememw .mmohw .eoeoflmeew 92. 8a 2.83— .BoZ

 

aw. cod we. dod

Dw M. Ow M
2338on

EoEoBon—Em 5m? coeoflmemw

da.

Ow

mmobw

eed

|><

vw. add avm £03228, .wd
Ow Wm Z
eoeoflmeww see .9560

 

8:228 m 2.3

wa. wvd
mo. vM.~
Nb.— VwN
mm. m fim
S. ved
mo. ww."
mm. afim
no. 8N
3. EN.
om. _N.N
Eu 3.—
ow. 3:6.
wa. mud
Om M
“Egokoqfim

3. i
we. 2a
2. «E
S. m:
8. :.~
3 92
mm. wwd
N». 03..
K. 2%
3. m2
3. :2
3. N3
ac. wwm
mm M
MBBanH

as, cocoemfim

mmobm

wad
Gd
cud
cofi
wwd
mwN
owN
vwd
afim
Ema
mad
cod
wo.m

wa. ewd
mm. NYN
we; Nod
ww. 3d
3. and
ac. wo.m
3.. wfim
mo. wvd
va. mwd
ow. Sum
ww. ma;
wa. ww.m
mm. and
Om M
:ocoflmtam £2.

ow
Sum

w m
9meq
wa
9...

ac
wmm.m
moo.—
5w w
w
~w~ ~
5%

cam—o: .2
32: .S
885 .2
3:50 .2
85cm 6
wag—Em .w
fine—SQ H
«350 .
mnfim .
aawsm .
a5§< .
£863. .
«assume ._

NMVIOO

.3550

 

was 33 E .9550 E 8§E> 3m .8 82338 Eugm Ea 882

 

 

v 05¢.

45

vw. wwd
5. SN
3. cozo.
ww. mad
ww. mud
wa. mm.m
3. RN
ww. 8N
mw. w fim
mo; amN
on. vfim
ww. wed
ww. mod
Om M.
Eon—.5338”;

S. RN
8. m3
9. Ed
mm. 8;
mm. 3;
2. am
on 8d
S. v2
on. and
m. N2.
R. m2
3. SN
S. :5
mm M
€03.50...
5? Suémfim

mmobm

wow
9mm
wwd
8N
cod
Rum
omN
mod
ewd
wwd
3N
an
awd

aw.
mm.
co;

5.

Gm

mod
end
ww.~
cod
cod
ww.N
w_.m
wwd
3am
mum
3N
8N
vwd

X

cowoflmwam 92.

>23.
owm._
3w
3.

own
3
amm
ww
c:
on
Sfim
mom

v5
van—RE.
538,—.
tau—523m
533m
53m

85 530m
Ema—om
.3382
cam—aoN >52
mug—“2:02
cons—z

59:.

.cm
.wN
.vm
.mm
.mm
AN
.cm
.3
.w—
.2
.3
.w _
.3

.9980

 

3:558 v 03$.

46

623.. bwcobm n w 22>» oouwmma bwcobm n _ :83 £5 E 85 08: 383 Aw I _ “98:9. 8323
8—8.0. Emoqw co 8&8qu :o woman 03 waogoaomEm 98 £836on 53» :ocoflmcmm .mmobm .couoflmuam £2. .80 3.33— .802

 

cm. 3;
3. mod
Om M
“anchoBanm

mo. ooN
Nb. 2N

Om X

xuoBEmoH
as, nouoemfim

ca.
3.

Om

mmobm

and
3:0.

mm. ad
8.— wwd
Om X

cozoflmunm :2.

3m m_osno=o> .wm
m_a.mm <mD SN

2
.9580

 

8:528 v 23.

47

.038 :80 00 A223 3:30:00 05 0:000:60 “WE—Swan
.0003. bwcobm u w 0.2? 00%85 bwaobm n _ .83 £5 E «05 08: 00003 .Aw I _ Sconce 00.583 8.08 “flaw :0 0030900..
no woman 03 30::03008m .05 £00th 5?» 00300033 .mmobm .couoflmuam pop. .80 £30m .wc. v a... .ocw.Nw u 2 .802

 

5.0 LS. :0- Sm. 8. SN .Eabsaam .v
35 Sn- .3. S. 12 083889 55 8:20me .m

G5 :2... S. 92 ”8% .N

9.3 3. SN 808033 no. ._

v m N _ am :82 03%;

 

 

 

«an 82 .8 83.5., 25 .6 30:20:83.5 93 223300 S§§m 88:

w 0309.

48

.0300 £000 .00 A293 3:50:00 05 000000000 0300905
.000w< bmcobm u m 0—23 009005 bwcobm u _ .800 0:: E 0.05 0000 00003 .G I _ “000300 003003 00300 05006 00 00000000..
00 0003 0.8 0000:03008m 000 0.003800% 55 00muofi0m00m .000bm dozoamgam 90—. .8.“ 3.000% .3. v 0... 03.3 n Z .0002

 

85 .00. .3..- an om. :0 08082080 .v
GS .00.- .8. «.0. 03 0.83.590 5? 80.0.00 .0

30.0 .00.- 3. a: .800 .N

00.0 8. 30 80800000 no. ._

v 0 N _ 00 :82 20.25

 

 

 

$0 052 .0 83.5.» 030 0c 0802008505 0:. 080.300 28:50 882

0 2.3

49

.2000 £000 .00 ASE—0V 5:30:00 05 0000000000 30000905
.000w< bwcobm n m 0—23 000w005 bw0000m u _ .0000 00.: E 0000 000: 00005 An I _ ”000000 003003 00—000 0500.0 00 000000000
00 00000 00 0000000300000.“ 000 £00th 505 000000000m .00000m .000000000m 00.. 00.0 00303— .mo. v a... ._~o.~0 n Z .0002

 

80.0 .00. .00.- .00. 00. 00.0 08802080 .0.
A05 .00.- .00. E. 20 08250.0 503 80800000 .0

000.0 .00.- 00. 00.0 0800 .0

000.0 00. 00.0 80203.0 02. .0

v m N _ Gm 0002 0_n0t0>

 

 

 

.000 300 80 8500.5 30 0o 0802208005 0.0 08003.0 .5230 0:82

h 030.0.

50

survey items within a country (Hofmann, 1997). OLS techniques disregard this
information and focus only on the between group variance. The second advantage
offered by HLM is that it allows for the study of both individual level and group level
variance in individual level outcome variables while maintaining the appropriate level of
analysis for the predictor variables (i.e., group level country predictors and individual
level attitude predictors) (Hofmann, 1997).

OLS techniques on the other hand would require one of two approaches to
handling a multi-level model such as the one under investigation in this study. One
option would be to disaggregate the country level data. This would be accomplished by
assigning the mean country score on individualism/collectivism, power distance, or
uncertainty avoidance for each nation, to each individual within each country, thereby
creating individual level “country” scores for each person. However, by disaggregating
the country level data to the individual level, the potential variance in the individual level
outcome variables that may be explained by nationality becomes based on the number of
employees rather than the number of countries. The effect of this is that standard errors
may be affected due to the increased sample size and, as a result, the inferences made on
the data may also be influenced (Bryk & Raudenbush, 1992; Tate & Wongbundit, 1983).
The other option is to aggregate the individual level data to the country level by taking the
mean of individuals’ responses within each country, thereby creating mean country scores
on all of the attitude variables. All of the data would then be analyzed at the country
level. However, this approach fails to address individual level variance in the outcome

variablas and therefore treats everyone within a country as equals. Given that one would

51

expect a great deal of within country variance in responses to attitude measures due to
individual differences, aggregation would not appear to be the most adequate approach
for analyzing the organizational attitude data.

Taken as a whole, HLM appears to be a positive alternative to traditional
regression techniques for analyzing and interpreting the data that are involved in the
present study. While there are other modeling techniques available that are similar to
HLM, HLM is considered to be the most appropriate technique currently available (Klein
& Kozlowski, 2000).

In order to test the hypotheses that involve a level-2 (cultural dimension)
moderator of the relationship between two level-1 (organizational attitude) variables,
HLM simultaneously accomplishes two tasks. One of these tasks involves the
computation of the relationship between the two level-1 variables in terms of intercept
and slope within each country. The other task involves the computation of the
relationship between a level-2 variable and the slopes calculated in the level-1
relationships for each country. HLM accomplishes this test for moderation using a series
of regression models that build on each other and eventually result in a somewhat large
multi-step regression equation. The four basic models are referred to as the null-model,
the random coefficients regression-model, the intercepts-as-outcomes model, and the
slopes-as-outcomes model. While only the final slopes-as-outcomms model is necessary
for testing for the hypothesized interactions, the previous sub-component models provide
useful information on issuw pertaining to variance (e. g., percentage of within group

versus between group, or amount of variance explained), and also allow for tests on direct

52

effect hypotheses. The sub-component models also supply the information necessary to
determine if certain requirements deemed necessary before moving on towards a test for
an interaction are actually met. Perhaps most importantly, a version of a sub-component
model will be used to test the direct effect hypotheses (H1, H2, and H5). Descriptions of
each of the sub-component models, the requirements they fulfill, and other specific
information is included in the following section that reports the actual results of the
hypothesis tests.

As stated earlier, in order to test the multi-level hypotheses suggested in this
study, certain conditions must first be met. The first of these conditions states that there
should be systematic between group variance. Given that each hypothesis suggests that
an outcome variable (e.g., job satisfaction, stress, or satisfaction with teamwork) will be
predicted differently for individuals from different countries, there must first be evidence
that variance actually exists between countries that rate differently on the cultural
dimension scales. HLM allows for a test of this condition in the null model.

As can be seen in Figure 2, the null model includes only a level-1 outcome
variable and an intercept estimate in the level-l regression equation (e. g., job
satisfaction). The level-l equation is the regression equation that in HLM that may
include both level-l outcome and predictor variables. Similar to the level-l equation, '
there are no predictors entered into the level-2 equation. This is the equation that will
eventually include a level-2 variable as a predictor of either the intercept or slope
calculated between two level-l variables. However, the lack of inclusion of level-2

predictors does not preclude the inclusion of the level-2 equation in this model as will be

53

explained shortly. Because there are no predictors in the level-l equation, the variance in
the level-1 outcome variable is regressed onto a unit vector. Keeping in mind that the
level-l equation calculates this regression intercept for each country, the regression onto a
unit vector results in an intercept that is equal to the mean value of the level-1 outcome
variable for arch country. Furthermore, by regressing the level-1 outcome variable onto a
constant unit vector, any within-country variance associated with that variable is thus
forced into the residual term (rij). Following the level-1 regressions, the level-2 equation
takes the intercept term (e. g., the mean value of job satisfaction) for each country and
regresses that value onto another unit vector. This level-2 equation yields both a gamma
coefficient (700) that is equal to the grand mean of job satisfaction across all countries, and
a residual term (UOj) that contains the between country variance. To summarize, while
the null model provides the foundation from which progressive HLM models will be built
off of, the result of primary interest at this stage is that the variance in the outcome
variable is partitioned into both between and within country variance. This model also
provides for a significance test of the between country variance in the form of a chi-
square test.

The null model was run to test for satisfactory between group variance in the three
outcome variables addressed in this study: job satisfaction, stress, and satisfaction with
teamwork. The results for job satisfaction, stress, and satisfaction with teamwork suggest
that there is significant between group variance for each of these outcome variables and,
therefore, that the first condition is met. Table 8 summarizes the results of the HLM null

model analyses for the three outcome variables for 1996, 1998, and 1999. Providing both

54

Figure 2

An Example of the HLM Null Model

 

 

Level-l Model:
Job Satisfaction ij = Bo; + fij

Level-2 Model:

[301' = 700 + U01
Where:

Job Satisfaction: example of level one outcome variable
[30,- = mean job satisfaction for group (country) j

700 = the grand mean of job satisfaction across all countries
rij = 0'2 = the within country variance in job satisfaction

Uoj = 100 = the between country variance in job satisfaction

 

55

 

00300000, 080800 0.0 00 0000000>

000000 0003000 .00 000000000 05 0000000000 00003 000000.000 0000000000 .1. 000 .0000000> .9008 0003000 0000000w00 .00 000000000
00.0 0000 000000-000 u 0.00 0000300000 0003 0000000000 000 .000000 .0000000000 000 00 000000.» 0000000 0003000 000 u 80 .3. v 0.. .0002

 

 

 

 

000. .3300 08. 002
03. 30.0000 000. 002
000. .8002 000. 0000
0000300000. .003 000000000m .m
03. 0.0.00: 000. 0000
000. 300:0 000. 0000
30. 30.800 000. 002
000% .N
08. .0000: 08. 002
$0. .00. :0 08. 000:
000. .0302 08. 002
80200000 000. .0
02 00 g. 0:00;
0000300000.

 

 

0003 000000000m 000+00000m .000000Q0m 000. .00 0000.002 00 0000.0I0|0> 0000 2 00030000 000 w00000. ”000002 :02 000 .00 0:0me

0 .500

56

the between and within variance, the null model also supplies the necessary information
required to calculate intraclass correlations, or the proportion of between group variance
in the outcome variables. The ICC values for job satisfaction, stress, and satisfaction
with teamwork for the years 1996, 1998, and 1999 are located in Table 8. The ICC
values, ranging from 3.7% to 6%, may be interpreted as the amount of variance in the
outcome variables that may be explained by country level data. The purpose of this study
then, is to determine if country rankings on the cultural dimensions of
individualism/collectivism, power distance, and uncertainty avoidance explain a
significant portion of the variance that may be explained by country level differences.
The small amount of country-level variance is not unexpected. The individual and
organizational variety within countries (e.g., language, class, education, geographic
location, organizational size, supervision style, etc.), combined with the individuality of
human beings, is likely to result in very large amounts of within country variance. It is
also true that all subjects are part of the same company. This company may have a
pervasive culture that acts to limit country level culture effects between nations. So,
while this study contends that individuals within a culture do to an extent share common
values and environments, it would be improper to assume that national membership
would explain large amounts of variance over and above individual differences.
Mean Difference Hymtheses

Given that the condition stipulating that there must be between culture variance in
the level-1 outcome variables was met, Hypoth0ses l, 2, and 5 may now be tested. Those

hypotheses suggwted that individuals who are fi'om countriw that differ on the cultural

57

dimension scales would report higher or lower levels of stress or satisfaction with
working in teams. The further conditions that will be discussed shortly, and that must be
met in order to test more complicated hypotheses, do not apply to H1, H2, or H5. These
hypotheses need only for there to be systematic variance in specific level-l outcome
variables that may be explained by culture. These hypotheses suggest a model to test for
direct level-2 predictor effects on level-l outcome variables (without the usual level-l
predictor also in the equation). In order to test these hypotheses, a somewhat unique
HLM model was created that included only the level-1 outcome variable in the level-1
equation, and a level-2 variable in the level-2 equation as a predictor of the level-l
intercept (see Figure 3). By setting up the model this way, the resulting analysis reveals
whether or not variance in individuals’ reported satisfaction with teamwork (H2 and H5)
or stress (HI) is explained by culture in terms of national ratings on the cultural
dimension scales.

Hypothesis 1 suggested that individuals from cultures high on collectivism would
report less perceived stress than individuals from cultures high on individualism. Table 9
summarizes the findings associated with testing this hypothesis. The Gamma
Coefficients in the table represent the relationship between individualism/collectivism
and stress. The t-test for this coefficient provides a significance test for the stated
hypothesis that in this case yielded non-significant findings for data collected in 1996,
1998, and 1999. While the satisfying of the first condition (based on the results of the
null model test) means that there are in fact differences in reported stress between

countries, this finding suggests that an individuals’ belonging to a culture that ranks

58

Figure 3

An example of the HLM model that was develmped to test the mean difference
hmtheses

 

 

Level-1 Model:
Job Satisfaction ij = Boj + fij
Level-2 Model:

Boj = 700 + 701 (Power Distance) + Uoj

Where:

Job Satisfaction: example of level-l outcome variable
Power Distance: example of level-2 predictor variable
[301' = mean job satisfaction for group (country) j

700 = the level-2 intercept

701= the level-2 slope

rij = 0'2 = the within country variance in job satisfaction
UOj = 100 = the residual intercept variance

 

 

59

higher or lower on individualism/collectivism does not significantly predict those
differences.

Hypothesis 2 predicted that individuals from cultures high on collectivism would
report higher levels of satisfaction with teamwork than individuals fi'om cultures high on
individualism. This hypothesis was also not supported for any of the years that data was
collected (see Table 9 for a summary of results). Differences in reported satisfaction with
teamwork between countries could not be explained by the country rankings on
individualism/collectivism.

Hypothesis 5 suggested that individuals from cultures high on uncertainty
avoidance would report less satisfaction with teamwork than individuals fiom cultures
that are low on uncertainty avoidance. Once again, cultural differences as defined by
Hofstede’s dimensions failed to predict differences in reported satisfaction with
teamwork in all three data sets (1996, 1998, and 1999) (see Table 9 for a summary of
results).

Moderation Hmtheses

The nine remaining hypotheses test for the moderating effects of culture on the
relationships between organizational attitude variables (i.e., satisfaction with teamwork,
empowerment, stress, and job satisfaction). Once again, in each of these hypotheses, the
organizational attitude variables represent individual level data (level-1) while culture is
represented by country level (level-2) rankings on power distance, uncertainty avoidance,
and individualism/collectivism. However, in order to test for potential moderation effects

of the level-2 variables on the relationships between the level-1 variables, two more

60

61

E m. mom. 0 So. 300. 0000
mam. 0mm. 0.. So. Soc. 0000
mwm. «mm. 08¢. 88. 0000
900300000 0003 0000000000

032 00 0000000 00030000003400 .N
mmm. Noe. Coo. N08. 0000
020. $0. cmoo. 008. 0000
mm». 0mm. Soc. 88. 0000
0000000 00300000

000— 00 0000000 E00>000=00v|00l0 .0

ABC
00_0>-0 0000-0. 0000 0000000m 00000000000 000000 00000004me

 

 

000 0000 0:380 .002 50 0 0:0 .0: 88580: 000000. 0o 03.0%

0. 0300

0000000w00 00 0000000000000 0.0 .00 0000000000 00 0000 n 0000

-H 030000> 00000000 0-00.60 000 000 0300000, m.._0>0_ 000 000300 0000000200 000 00 0000000000 00000 030 00>0_ 000 100. .3. v 0 0 .0002

 

new.
00”.
:00.

020>-0

00m.
oi.
000..

208-0.

0 So.
zoo.
omoo.

00000 0000005

vooo. 0000
0000. 0000
c000. 0000

300300000

0003 0000000000 0000 00
0000000 000000000 000000000000fl .m

as:
000000000000 08000 0000000002

 

00000000 0 030.0.

62

Figure 4

An example of the HLM Random Coefficient Regession Model

 

 

Level-1 Model:
Job Satisfaction ij = BOj + Br j (Empowerment ij) + rij
0014210001:

301': 700 + U01
Bu = 710 + Uu

Where:

Job Satisfaction: examaple of level-l outcome variable
Empowerment: example of level-l predictor of job satisfaction
700 = the mean of the intercepts across countries

710 = the mean of the slopes across countries

rij = o“2 = the level-1 residual variance

UOJ' = too = the variance in intercepts

U1,- = r” = the variance in slopes

 

63

 

conditions must first be met.

To begin with, in order for a level-2 variable to moderate the relationship between
two level-1 variables, there must first be a relationship between the two level-1 variables.
The random coefficients regression model provides a test of this condition. This model

builds on the null model by including both a level-1 outcome variable (e.g., job
satisfaction) and a level-1 predictor variable (e. g., empowerment) in the level-1 equation
while still leaving out any specific level-2 predictors in the level-2 equation (see Figure
4). In the random coefficient model, the level-1 equation simply regresses the level-1
outcome variable on the level-1 predictor and intercept. Then, as there are no level-2
predictors included in the level-2 equation, this model regresses the intercepts and slopes
created in the level-1 equation on an intercept term and a residual (similar to the null
model). The resulting gamma coefficients represent the mean of intercepts across
countries (700) and the mean of slopes across countries (Y10)- Of primary interest for the
satisfaction of the second condition is the t-test of the 710 coefficient which tests whether,
on average, the slope, or relationship, between the level-l predictor and level-1 outcome
differs significantly fiom zero.
The random coefficient model was used to test the four level-l relationships that
are suggested in the hypotheses. The level-l relationships that were tested included
satisfaction-with-teamwork with job satisfaction, satisfaction-with-teamwork with stress,
empowerment with job satisfaction, and empowerment with stress. Rmults of the t-tests
on the 710 are located in the first two columns of Table 10. All four level-1 relationships

were found to differ significantly fi'om zero in all three years that data was collected.

This model also provides R2 values that represent the percentage of variance in the level-1
outcome variable that is accounted for by the level-l predictor. The resulting R2 values
for data in 1996, 1998, and 1999 are contained in Table 10 and range from 9.7% to 42.8%
with a mean of 28. l %. Combined, these results provide evidence that the suggested
level-l relationships do exist, and therefore satisfy the second necessary condition for
testing the actual moderation hypotheses.

The random coefficients regression model also provides a test for the third and
final necessary condition. In order to test hypotheses that suggest that a level-2 variable
moderates the relationship between two level-l variables, there must also be evidence
that the slopes generated by the regression of the level-l outcome variable on the level-l
predictor differ between groups. Without such evidence, there would be no reason to
expect that being from different countries would differentially influence the relationships
between the level-1 variables. The random coefficient model in HLM tests for this
condition using a chi-square test for the residual variance component of the level-2
equation (I. 1). Much like in the null model, this residual component in the level-2
equation contains the between-country variance, this time the variance in slopes.
Therefore, a significant chi-square test is evidence of significant variance in slopes
between countries.

Results of these analyses showed that for the level-1 relationships involving
satisfaction-with-teamwork with job satisfaction, empowerment with job, satisfaction-
with-teamwork with stress, and empowerment with stress, the regression slopes did vary

significantly between groups. Thus, the final condition for testing the suggested

.65

mmm.
000m.
0mm.

mum.
mmm.
mN0.

wmv.
80.
men.

0.00va
0.2.090
09.0.3

.300»
*mon0
000.000

000.0“N0
000.000
0 0N.N 0 v

x

moo.
moo.
voo.

N0o.
o—o.
moo.

woo.
woo.
woo.

05

00005 00
0000000>

30.02%
000202
0mm.0~om

Kwfimom
03.33
50.32

03630
002.030
0NNSMNN

x

wmo.
ovo.
mvo.

ovo.
mmo.
0mo.

omo.
ovo.
mvo.

0800

:0

0000000>

3.

mo.

mo.
No.
No.

mo.

No.

000m

0000000000 0000000m

00m.
00m.
*Nm.

*om.
*wm.
«NM.

LE.
02..
Kb.

A0 0 3
0000000000
00000000

0000
300
000 0

000000000m

 

 

000. 0003 00000003000fl

000 0
woa 0
000 0

0..0||000m

£§fi03800h 0003 00000.0000m

0000
woo 0
0000

 

 

000000000m 000.
00003 0000300000400 03 00000.00.0|00m|

.0

00000000000000

 

 

00002 0000000000 0000000000 800$ 0.0 .00 00003.

o0 030.0.

66

.0733.» 8868a :05— ofi .3 Eu 3:583 mm 35 03%"? 0803.8 :26— 05 5 8E?» mo owficoobq M Na demo—m 8 $3285
5 count? 2: .«o oofiommcwfi 05 356 8m 38 9833-30 n NR .3358 $83 mono—m 05 mo :88 05 n 2» 6N u .3 .3. v a... .082

 

mom. Pfimmh

com. *mmfimc

3o. *woémm
Ny— Nx

coco.
:8.
35¢.

A. _ Pv
mono—m E
853$

*modhw—
$038
1.5de

x

o3.
mmo.
5o.

A8:

3328.:

E
oo§§>

No.
No.
No.

Sum
235%

..&m.
..&m.
LN.

G5
856580
«8:80

mam—
M32
coo—

 

mmobm 5m? BogoBomam .v

qumcomE—om

 

32:38 3 2an

67

Figure 5

An example of the HLM Slopes- -Outcomes Model

 

 

Level—l Model:
Job Satisfaction ij = Bo,- + B1 j (Empowerment ij) + rij
Level-2 Model:

[30,- = 700 + 701 (Power Distance j) + UOj
(31,- = 710 + 71. (Power Distance j) + U1,-

Where:

Job Satisfaction: example of level-1 outcome variable

Empowerment: example of level-l predictor

Power Distance (in BOj equation): example of level-2 predictor of level-1 intercept
Power Distance (in Blj equation): example of level-2 predictor of level-1 slopes
700 = the level-2 intercept

701= the level-2 slope

710 = the level-2 intercept

y“ = the level-2 slope (actual test of moderation hypothesis)

rij = 0'2 = the level-1 residual variance

UOj = too = the residual intercept variance

U1j= 1:” = the residual s10pe variance

 

68

 

moderation hypotheses has been satisfied. Results of these analyses are summarized in
the right-most portion of Table 10 (chi-square tests for variance in intercepts between
countries (I; 1) are also included in the table for reference).

The moderation hypotheses were tested using a slopes-as-outcomes model. This
model maintains the characteristics of the previous model (i.e., one level-1 predictor and
one level-1 outcome variable) but adds aleve1-2 predictor into the equation (see Figure
5). The term “slopes-as-outcomes” is used to name this model because the level-2
variables, in this case either power distance, uncertainty avoidance, or
individualism/collectivism, are brought into the equation as predictors of the level-1
regression slopes. It should be noted that the level-2 variables in this model are also
included as predictors of the level-1 regression intercepts. Intercepts-as-outcomes is a
separate model used to test for differences in intercepts depending on group membership
(see Figure 6). While that particular question is not of direct interest for answering any of
the current hypotheses, the characteristics of the intercepts-as—outcomes model are
maintained in the slopes-as-outcomes model as it is necessary to partial out the main
effects of culture before testing for moderation effects in the slopes-as-outcomes model.
As may be seen in Figure 5, the slopes as outcomes model takes the level-1 relationship
slope coefficients and regresses them in an equation that includes a level-2 variable (e.g.,
power distance). In this way, the model tests for the existence of a moderating
relationship between the level-2 variable and the slope of the level-l relationship. In
terms of this study, a significant t-test of the gamma coefficient (71 1) in the level-2 slope

equation is an indication that the relationship between the organizational attitude

69

variables (level-l) is moderated by culture (level-2). Associated R2 values represent the
amount of variance in the level-1 slopes across countries that is explained by the level-2
predictor. The slopes as outcomes model also provides a chi-square test on the residual
variance in slopes between countries in order to determine if significant amounts of this
variance remain to be explained by other unidentified level-2 predictors. This will be
discussed further after a summary of the results of the hypothesis tests.

Hypothesis 3 stated that the relationship between satisfaction with teamwork and
job satisfaction would be stronger and more positive for individuals from cultures that are
high on collectivism. Results of this analysis may be found in first column of Table 11.
Analysis of this hypothesis resulted in non-significant results for all three years that that
data was collected. These findings suggest that the relationship between the level-1
variables that were reported by the individual respondents are not moderated by the
individualism/collectivism scores for the countries that they are from.

Hypothesis 4 proposed that the relationship between satisfaction with teamwork
and stress would be stronger and more negative for cultures that are high on collectivism.

This hypothesis was also not supported by the data for 1996, 1998, or 1999 (see Table
11). Once again, there is lack of evidence to support a moderation of the level-1
relationship.

Hypothesis 6 suggested that the relationship between satisfaction with teamwork
and job satisfaction would be stronger for individuals from cultures that are high on
uncertainty avoidance than for individuals fiom cultures that are low on uncertainty

avoidance. This hypothesis was not supported by the data in 1996, 1998, or 1999 (see

70

Figure 6

An example of the HLM Intercepts-as-Outcomes Model

 

 

Level-1 Model:
Job Satisfaction ij = 301' + Bl j (Empowerment ij) + rij
Level-2 Model:

[30,- = 700 + 701 (Power Distance) + U0,-
flu‘ = 710 + U11

Where:

Job Satisfaction: example of level-1 outcome variable

Empowerment: example of level-1 predictor

Power Distance (in B0,- equation): example of level-2 predictor of level-1 intercepts
700 = the level-2 intercept

701= the level-2 slope

710 = the mean of the slopes across countries

rij = 62 = the level-1 residual variance

UOj = too = the residual intercept variance

U],- = m = the variance in the slopes

 

71

 

K52:
*2.ow
$0.th

*3.me
*mndmm
Lwfiom

C S.
35.
ego.

mmoo.
once.
goo.

88:83
285 8.23:

3. Eco. 28.- 33
3o. Soc. o 8?. @8—
mo. Soc. 88.. coa—

A.=oo\.§ 3 38088 .38: 2
8822 x8383: 5m? :oaomwmuwmvflm .N

:8. 88. 88. 32
of. 88. :8. «9:
E. 28. 38.- 32

@8965

3 888288 838833 ace 8
@322 x8388: £3» :ouoflmuamvfl .—

NM :85 BEESm A. 5 momofiogm
356580
«880

 

 

83509.3 89:20:02 8.: .352 38880 ma mono—m 05 me 338%

2 2an

72

*meoS
*mmdoh
*NodE

*NNd—v
.30wa
embdmm

3o.
o_ S.
omoo.

mvoo.
mooo.
whoo.

88:83
255 3:28:

S.
omo.
moo.

SN.
woo.
woo.

NM

Coo.
Soo.
mooo.

wooo.
oooo.
o _oo.

8:: 88%

M So.
hooo.
moooo.

floor
ofioor
mooor

c 5
953880
«88:0

ooo _
woo_
woo:

308208
b88308 3 3:80:08 .393 8
@838 x8383: 5m? 8:088:8on .v

moo ~
woo ~
woo _

328298 58880::
3 38588 828838 new 3

38—2 x8383: 53> :ouoflmwemvuolm .m

momofiogm

 

8:538 : 2%:

73

*oh.mvo
*ooaoh
*oo.oom

*om.mvm
*vamm
*mo.mom

coo.
ovoo.
Vmoo.

mmoo.
mvoo.
goo.

00§3>
0%: 8.28:

hmo.
coo.
to.

So.

m9.

Nm

o_oo.
ofioo.
oooo.

wooo.
oooo.
o _ oo.

8.00 03890

*Eoo.
ehfioo.
m 8o.

Nooo...
oooor
oooo.

o 8 3
08005000
08800

oooo
woo_
woo

00:88: 030: 3 308038
.303 9 300—0.: 808003383 mm. .0

ooo _
woo_
ooo_

00:80.6
0038 .3 308038 8030.330
com 8 380.: 80833383 3 .m

0000809»:

 

3:888 2 030B

74

0.363
{3.2.0
0.2.2N

emodom
03.9:
.52.?

Vmoo.
wvoo.
omoo.

omoo.
mmoo.
goo.

00:88;
2.20 3:28:

mmo.
omo.
mo.

N3.
mom.
2.

NM

hooo.
wooo.
«ooo.

oooo.
oooo.
hooo.

8:: caegm

woooor
noooo.
mooov

hooo.
0N _ oo.
Nooor

30
08805000
08800

ooo —
woo _
coo ~

0080896
3808008. 3 308038 88008880
08. 3 38.00 808003883 film .w

ooo ~
woo ~
ooo g

3.00208 .3 38038 88008880
new 8 300—00 80803883 3 .0

000052;:

 

fiofiflmufloo ~— 2th

75

800$:me 00 0080.83 ooo—m 332000 05 .00 088000000 3 0000 00033.30 u 005 00000008 m-_0>0_ 05 3 0080—900
00 005 005850 000000 000020 T_0>0_ 05 8 00000009 .00 08580 N Na .0030_00> 205— 030 05 0000303 932020200 05 000000008
05800> N-_0>00 05 00:..— 00000E>0 000 080030.008 008800» 80000::me 8205000 ooo—m 030 _0>0_ H F» 6m .I. .3 .3. v P. .0002

 

.30de
000.8:
*cwdw

*mm.:w
*mmdmn
*ovdmm

mooo.
whoo.
omoo.

moco.
omoo.
smoo.

00800000»

2020 03280

owm.
#3.
m3 .

who.
moo.
ooo.

Nd

wooo.
wooo.
Boo.

ocoo.
o 0 oo.
goo.

008m 0002000m

0w Soc
0.025..
0.300..

88.
o So.
825..

as

mU—QDWOEOOU

08800

ooo~
woo _
coo _

A8030000=o€80=0=23©8
03 000000008 £00000 9

 

0000000 000300.033 no.0 bSEBQBi .3

ooo~
moo—
ooofi

30802020 380000080 03 600000008
£00000 8 000200 808003383 Na .o

0000509»:

 

825:8 : 2.3

76

Table 11 for a summary of results).

Hypothesis 7, that the relationship between satisfaction with teamwork and stress
will be more strong and negative for individuals from cultures high on uncertainty
avoidance was also unsupported by the data for any of the three years (see Table 11 for a
summary of results).

Hypothesis 8, that the relationship between empowerment and job satisfaction
will be more positive for individuals from cultures that are low on power distance than
for individuals from cultures that are high on power distance was not supported by the
data (see Table 11 for a summary of results).

Hypothesis 9, that the relationship between empowerment and stress will be
higher and more positive for individuals from cultures that are high on power distance
was marginally supported in both the 1998 data (yu = .0017, se = .001, t = 1.718, p =
.097) and the 1999 data (711 = .0016, se = .001, t = 1.517, p = .141). A summary ofthe
results may be found in Table 1 1. Plotting the marginally supported interactions for both
years (see Figure 7 for the plot of the interaction found in the 1998 and 1999 data) reveals
that empowerment and stress are actually more negatively related for individuals from
cultures that are accustomed to unequal power distributions. A test of the R2 value for
this result shows that power distance accounts for .6% and 3.7% of the variance in
empowerment-stress slope across groups for 1996 and 1998 respectively.

Hypothesis 10, that the relationship between empowerment and satisfaction would
be more positive for individuals from cultures that are high on individualism was

supported by the 1998 data (711 = .0012, se = .001, t = 2.09, p = .046) but not by the 1996

77

or 1999 data (see Table 11 for a summary of results). Plotting this result shows that
individuals from individualistic cultures do indeed show a more positive relationship
between empowerment and satisfaction (see Figure 8). A test of the R2 value for this
result shows that individualism/collectivism accounts for 28.5% of the variance in the
empowerment-satisfaction slope across groups.

Hypothesis 11 suggested that the relationship between empowerment and job
satisfaction would be lower for individuals from cultures that are high on uncertainty
avoidance than for those low on uncertainty avoidance. The hypothesis was not
supported by the data fi'om any of the three years (see Table 11 for a summary of results).

Hypothesis 12, that the relationship between empowerment and stress would be
stronger and more positive for individuals from cultures that are high on uncertainty
avoidance was also not supported by any of the data (see Table 11 for a summary of
results).

It was mentioned earlier that the slopes-as-outcomes model also provides a chi-
square test to determine if the residual slope variance across countries is significant. As
can be seen in Table 11, the residual slope variance parameter (U 1) was significant for
every hypothesis for every year. These findings suggest that there remain other country
level variables, not considered by this study, which may account for unique variance in
the slopes of the level-1 relationships across groups. For example, country level variables
including religion, language, socio-economic status, geography, weather climate, and
political systems may all potentially account for some part of the variance in slopes that

does exist between countries.

78

,u Ln; '_ 0 ,I‘ I!.=_r_l
mm1228gd1999dataߤ

   

 

+ Low Power Distance
+ High Power Distance

 

 

Low High

Empowerment

79

Job Satisfaction
N 0)

#010)

 

80

 

 

 

+Lo~|ndvidrdism
(“gm collectivism)

+ th Indvidrdism
(lawodledivism)

 

 

Exploratog Results

After looking over the data it was determined that there was a relationship
between job satisfaction and stress. After reviewing the literature on this relationship
(which will be covered in the discussion section) I decided to test an exploratory
hypothesis to investigate whether that relationship was moderated by
individualism/collectivism. The suggested hypothesis is that the relationship between job
satisfaction and stress will be stronger and more negative for individuals from cultures
high on individualism. This was based on the idea that individuals fiom collectivist
cultures have more support structures in place and are therefore less likely to report stress,
even when less satisfied with their jobs. Analysis of this hypothesis resulted in
significant findings for 1996 data (711 = .001, se = .001, t = -2.184, p = .038), 1998 data
(m = .002 se = .001, t = -2.145, p = .041), and the 1999 data (m = .002, se = .001, t = -
2.375, p = .025). A summary of results may be found in Table 11. Plotting the
interactions for those three years reveals that the relationship between job satisfaction and
stress is stronger and more negative for individuals fiom cultures that are high on
individualism (see Figure 9 for a plot that represents the interaction found for all three
years). A test of the R2 values for these results show that power distance accounts for
14.5%, 12.4%, and 28% of the variance in the job satisfaction-stress slope across groups
for 1996, 1998, and 1999 respectively.

Discussion
The purpose of this study was to examine a model that pr0posed that the

relationships between certain organizational attitude variables (e. g., satisfaction with

81

 

 

+ Low hdividualism
(tigh collectivism)

+ High hdividualism
(low collectivism)

 

 

 

 

Low High
Job Satisfaction

82

teamwork, empowerment, job satisfaction, and stress) were moderated by dimensions of
culture. Three of Hofstede’s well known dimensions of national culture (e. g.,
individualism, power distance, and uncertainty avoidance) were employed as delirniters
of culture, as they proved to be one of the more well grounded attempts at defining
differences between cultures. Questionnaire responses fi'om salaried employees of a
global manufacturing organization provided data from 28 countries that had also been
included in the Hofstede studies.

The results of this study provided only minimal support for the suggested model.
Specifically, there was no evidence of support for the direct effect hypotheses and only
minimal support (specifically for H9 and H10) for the hypotheses that proposed culture as
a moderator of the relationships between organizational attitude variables. Overall, the
results point towards a less broad application of cultural variables and their influence on
organizational attitudes than was suggested by the model.

Supp_orted Hymtheses

Hypothesis 9 suggested that the relationship between empowerment and stress
would be higher and more positive for individuals from cultures that are high on power
distance. Analysis of the data revealed marginally significant results for the 1998 and
1999 data sets hinting that power distance did moderate this relationship. However,
contrary to the stated hypothesis, the relationship between empowerment and stress
turned out to be negative. As mentioned earlier, there is some debate in the literature as to
the relationship between empowerment and stress. While some suggest that increased

responsibilities associated with empowerment lead to an increases in perceived stress

83

(Mishra & Spreitzer, 1998; Sutherland, Fogerty & Pithers, 1995), others purport that
empowered workers, who have more choices and control in their work lives, experience
lower levels of stress (Jackson, 1983; French & Caplan, 1972). Therefore, while the
direction of the relationship was not suggested, it is not a surprising finding. Congruent
with Hypothesis 9 is that fact that the relationship between empowerment and perceived
stress was stronger for individuals from countries that are high on power distance.

In order to best understand the marginally significant findings, the interaction
must be discussed. The first point of interest is that the individuals from high power
distance nations, who operate at what they see as low levels of empowerment, report
more perceived stress than individuals from low power distance nations who also report
low levels of empowerment. This finding is in line with the thinking of Goffman (1961)
who suggested that organizations in high power distance nations tend to have rigid role
structures that compound role stresses. While both groups of individuals (high and low
power distance) are experiencing low levels of empowerment, those individuals from
countries high in power distance are more likely to operate in a system that includes
unequal power distributions (Hofstede, 1980) and dysfunctional role structures (Goffinan,
1961). These conditions, that are associated with high power distance nations, appear to
create an environment where workloads and other work place stimuli are perceived as
more stressful when compared with reports from individuals operating in low power
distance environments. Perhaps despite low levels of reported empowerment, individuals
from nations low in power distance have enough role autonomy and shared power built

into their every day organizational life that “limited empowerment” carries a different

84

meaning for them. In other words, the low levels of empowerment reported by
individuals from nations low in power distance may be qualitatively higher than the low
levels of empowerment reported by individuals fiom countries high in power distance. In
turn, individuals fi'om low power distance nations may have freedoms built into their
work roles that allow them to better deal with work place stressors, and hence report less
perceived stress.

As levels of reported empowerment increase, individuals from both low and high
power distance nations report less perceived stress. This finding, that empowered workers
report less stress, supports the contentions of researchers such as Jackson (1983) and
French and Caplan (1972). However, at one standard deviation above the mean for
empowerment, individuals from countries that are high on power distance report levels of
stress that are very similar to individuals from countries low on power distance.
Therefore, it appears that autonomy, decision-making power, and other characteristics
associated with empowerment have a more powerful influence on the reported stress
levels of those individuals from countries where such freedoms in the workplace are not
the norm.

Taken as a whole, these findings suggest that organizations that are concerned
about the health of their employees and the costs associated with stress related problems
should be aware of the worker’s perceptions regarding levels of empowerment. This is
especially true in high power distance cultures where individuals do not have the general
work role freedoms that are more common in low power distance cultures. It appears that

employees who perceive themselves as being empowered feel that they have been given

85

the necessary tools and fieedoms to handle workloads and workplace stressors more
effectively than those who are not empowered.

It should be noted that these marginally significant results were only found in the
1998 and 1999 data sets and could not be established in the earlier 1996 data set. While
this presents some cause for concern regarding the reliability of these findings, the
differences in the results may be due in part to the fact that the sample sizes significantly
increased between the 1996 and 1998 data collection periods. Overall, the fact that the
results were obtained across two years of data collection gives good reason to believe that
the results are not due to chance.

The only other significant finding for the moderation hypotheses was in support of
Hypothesis 10. This hypothesis suggested that the relationship between empowerment
and job satisfaction would be more strong and positive for individuals from cultures that
are high on individualism than for people from cultures that are high on collectivism (low
on individualism). As predicted, the nature of this interaction is such that people from
individualistic cultures displayed a stronger relationship between increased empowerment
and increased job satisfaction than did individuals from collectivist cultures. This result
is in line with the thinking of Barley and Gibson (1998) who suggested that people fiom
individualistic cultures tend to enjoy autonomy, decision-making latitude, and employee-
centered incentive systems, many of the same terms that are used to describe
empowerment. This finding is also in support of previous research done by Near (1986)
who found that freedom on the job was related to satisfaction for workers from

individualistic cultures, but not for those from collectivist cultures.

86

As was the case with empowerment and perceived stress, it appears that levels of
empowerment are importantly related to employee’s formation of opinions regarding their
job satisfaction. Also, much like the relationship of empowerment with perceived stress,
increased levels of empowerment appear to be beneficial to employees across culture.
This is evidenced by the fact that both individualists and collectivists show a strong
positive relationship between empowerment and satisfaction. Therefore, if an
organization is concerned with their employee’s general job satisfaction, it would seem to
be important to monitor how employees perceive the level of empowerment that they are
experiencing. This is especially true for organizations operating in cultures such as the
United States, Canada, Ireland, Australia, and the United Kingdom which are extremely
high on individualism and where people place more emphasis on the individual fieedoms,
and reward opportunities that are offered by empowerment programs. This fact is
emphasized by the finding that 28.5% of the variance in the empowerment-job
satisfaction slopes across countries is accounted for by a nation’s standing on the cultural
dimension of individualism/collectivism.

While this finding does support Hypothesis 10, and explains a large amount of the
variance in slopes across countries, it should be viewed with hesitation. The interaction
that was found in the 1998 data could not be replicated in either the 1996 or the 1999 data
sets. As has been mentioned, these data sets vary in size, and there is no way to know if
the same people filled out the surveys each year, or if it is an entirely different sample.
Nonetheless, the fact that this hypothesis was not even marginally supported in either of

the other two data sets provides reason to believe that to some extent the finding is just a

87

chance occurrence within the 1998 data set.
Other Interesting Findings

Despite the lack of significant findings for both the direct effect and moderation
hypotheses involving cultural dimensions, analysis of the data from all three years did
provide some interesting information regarding differences and similarities between
countries. First of all, it was found that there is variance between countries in the
outcome variables. This finding shows that levels of satisfaction with teamwork (in the
direct effect hypotheses), perceived stress, and job satisfaction vary by country in a
meaningful way. Specifically, 3.7% to 6% of the variance in the outcome variables is
attributable to country level differences. While perhaps not unexpected given the many
differences between the countries that were included in this study, this finding
nonetheless provides evidence that even at the country level, there do exist differences in
attitudes about organizations.

A second finding of interest is that across countries, the hypothesized individual
level relationships between organizational attitude variables did exist and were
significant. Specifically, the relationship between satisfaction with teamwork and job
satisfaction, as well as the relationship between empowerment and job satisfaction were
found to be positive and significant. Also as predicted, the relationship between
satisfaction with teamwork and stress was found to be negative and significant such that
higher levels of satisfaction with teamwork were related to lower levels of perceived
stress. Finally, empowerment was related to stress but in the opposite direction from

what was hypothesized. However, this negative relationship was not completely

88

surprising given the earlier discussion on empowerment and stress. In fact, these finding
proved further support for the relationships between these variables as have been
suggested in some of the literature (i.e., the relationship between working in teams and
job satisfaction (Hackrnan, 1987; Sundstrom, De Meuse, & Futrell, 1990), working in
teams and stress (Bass, 1982), empowerment and job satisfaction (Kirkrnan & Rosen
1999; Koberg et al., 1999, Locke, Schweiger, & Latham, 1984), and empowerment with
stress (Jackson, 1983; French & Caplan, 1972)).

A third finding was that the slopes of the relationships that were just described
were found to vary significantly between countries. This finding provides evidence that
the strength of the relationships between the organizational attitude variables do indeed
differ between countries. This finding alone is evidence that there may be some unique
quality to certain countries that cause the relationships between the attitude variables to
be either more or less strong. Given this finding, an argument can be made for the
usefulness of cultural dimensions. As there is evidence that as of yet unexplained
systematic between nation variance exists in the slopes of the individual level
relationships, there is reason to search for common factors, cultural dimensions among
them, that may explain that variance. While this particular study was mostly unsuccessful
in that endeavor (and these lack of findings will be discussed later), the evidence that
between country variance in slopes does exist is a small victory that should not be
overlooked.

A fourth finding was that no support was found for Hypothesis 8, that the

relationship between empowerment and satisfaction would be higher and more positive

89

for individuals from cultures low on power distance. These lack of results fail to provide
evidence in support of the findings of Robert et. a1 (2000) which was one of the only
studies to directly test a hypothesis involving culture as a moderator of the relationship
between two organizational attitude variables. However, unlike in the study by Robert et.
a1 (2000), this study actually included power distance scores in the analysis and therefore
may provide better insight into the actual lack of cultural (in terms of power distance)
moderation between empowerment and satisfaction.

Discussion of Exploratory Findings

One interesting finding did result from the exploratory analyses. After examining
the literature, it was determined that it was worthwhile to investigate a relationship
between the two variables that were primarily used as outcome variables in this study, job
satisfaction and stress. While there is some support for a negative relationship between
job satisfaction and stress (Bateman & Strasser, 1983; Cavanaugh, Boswell, Roehling, &
Boudreau, 2000; Tetrick & LaRocco, 1987), there was no mention of an attempt to look
for cultural moderators of this relationship in any of the literature that I reviewed.

The relationship between job satisfaction and stress was more strongly negative
for individuals from cultures that are high on individualism than for individuals from
cultures that are high on collectivism. This finding is similar to that of Bhagat, et
al.(l994) who found that individuals fiom cultures high on collectivisim tended to show
less of a relationship between stressors (e. g., job dissatisfaction) and perceived strain
(e.g., perceived stress). This finding may be explained by the contention that people in

collectivist cultures tend to display more social cohesion, social support, and less

90

insecurity than people in individualistic cultures (Triandis, et al., 1988). It is likely the
supportive in-group structure of a collectivist culture that provides a coping mechanism
for dealing with potential stressors that is lacking for a higher percentage of individuals in
an individualistic culture. Of special importance in the case of this finding is that the
moderation effect of individualism/collectivism existed for all three years of data
collection.

Researchers such as Jones (1984), Jackson (1983), Coquhoren (1976), Cooper &
Roden (1985), Weiss & Cropanzano (1986) and Lazarus (1991) among others have
shown that stress and its associated health outcomes in employees have not only a
monetary cost for organizations (e. g., absenteeism, turnover, health care, etc.), but also
performance decrements (e.g., accidents, counterproductive behaviors, and sometimes
decreased job satisfaction). Therefore, it would seem to behoove organizations that are
wonied about the aforementioned issues to also be concerned about the amount of stress
that employees are experiencing. Given the relationship between satisfaction and stress
that is suggested in the literature discussed above, and which is supported by the data in
this study, one potential way to help reduce the likelihood of perceived stress affecting
individuals in the workplace may be to either ensure that employees are satisfied. As it is
often difficult to “ensure” that employees are satisfied with work, a more informed
conclusion to take from this finding is that all organizations should learn fi'om collectivist
culture techniques for coping with potential stress. If organizations in individualistic
cultures could find a way to include group support, foster the in-group cohesion amongst

working teams, and build social support mechanisms, employees may learn to cope with

91

potential stressors such as job dissatisfaction before they become perceived as adding to
experienced levels of stress.
Non-summed Hypotheses / Limitations

Also important, albeit more frustrating, was the lack of support for both the direct
effect and moderation hypotheses. The suggested direct effect relationships between
Hofstede’s cultural dimensions and the attitude variables (Hl, H2, and H5) simply did not
exist. Even when the cultural dimensions not specified in the hypotheses were tested for
direct effects in exploratory analysis, no consistent findings emerged. While the
satisficing of the HLM condition that there exists variance between countries in the
attitude variables does offer some support for a cross-level relationship between the
attitude variables and some country level variable, use of Hofstede’s cultural dimension
variables proved to be unable to explain that variance. A similar lack of findings was
discovered in the majority of the moderation hypotheses that were under investigation
(H3, H4, H5, H6, H7, H8, H11, and H12). It appears that there may be two primary
reasons why there was lack of support for both the direct effect hypotheses and the
moderation of the attitude variable relationships: the lack of variance and the nature of
Hofstede’s cultural dimensions.

During the analysis of the data, it was discovered that significant variance in the
attitude variables and the slopes of the attitude variable relationships across countries did
indeed exist. While this seemed promising at first, it should be remembered that this
variance is based on differences between countries and not cultures. Keeping this in

mind, one major factor that may have influenced the lack of findings was that across all

92

these hypotheses, across all three years of data collection, the amount of this variance was
consistently very small (.006 on average). This meant that that there was limited variance
that could be explained by any country level variable, not to mention a specific aspect of
each country such as the national score on one of Hofstede’s cultural dimensions. This
finding provides evidence that opposes what seems to be a prevalent view in the applied
psychology literature, that there are important differences between cultures with regard to
employee attitudes. This may hold particularly true within a specific organization,
especially if that organization has a strong internal culture and norms that serve to limit
the influence of national culture. Numerous researchers have discussed the potential
strength of the influence of the organization on an individual’s values and attitudes
(Badawy, 1980; Griffeth, Horn, DeNisi, & Kirchner, 1980; Hull, 1987; Ronen & Kraut,
197 7; Redding, 1976). Findings of particular interest include those that have shown that
founders of organizations and top management have long-term effects on organizational
culture and that people within organizations tend to be somewhat homogeneous as a .
result of said culture (Schneider, Goldstein, & Smith, 1995). Ronen and Shenkar (1985)
point out that this organizational cultural influence is too often overlooked in cross-
cultural psychology. Indeed, when one considers that there were 28 different nations
from various parts of the world that were included in this study, the fact that there is such
limited variance in both the direct effects and SIOpes of the relationships between attitude
variables is interesting, if not amazing. Given these findings, one way to look at this
would be to infer that upper management within a given global organization need worry

less about cultural differences with regard to the organizational attitude variables that

93

were under investigation in this study. Instead, their time would be better spent insuring
that their own organizational culture is well developed, consistent, and throughout all
branches of their organization. However, another potential inference may be that the
nature of the questionnaire and the items for each scale resulted in a context where
employees, regardless of country or culture, responded in certain socially desirable ways.

Such actions by employees, while not testable in this case, may have masked potential
variance in the attitudes held by employees from different cultures.

While it is true that the variance that could potentially be explained by country
level variables was very small, it was nonetheless significant. Furthermore, despite the
fact that the significance of the variance values were likely influence by the extremely
large sample sizes, there remained cross-country variance to be explained which may
have been of interest from an academic standpoint, even if not fi'om an applied one. This
point of view then begs the following question: why did culture fail to explain the
variance that did exist between countries? The most likely scapegoat for the explanation
of the non-results would be to turn to the inadequacy of Hofstede’s cultural dimensions.
Researchers such as Schwartz (1994) and Sondergaard (1994) have discussed possible
shortcomings of Hofstede’s dimensions including how exhaustive the dimensions are of
the culture construct, the adequacy of the sample, the age of the study, and how
generalizable the dimensions are from the results discovered in one organization.
However, I am not prepared to declare that Hofstede’s dimensions are an inadequate
means of delineating culture. First of all, despite potential limitations the adequacy and

benefits of Hofstede’s dimensions, as well as the lack of obviously superior alternative

94

cultural dimension, have already been discussed in this paper and I feel that those
arguments still hold true. Second, as evidenced by the significant and marginally
significant moderation findings that were established in this study, Hofstede’s dimensions
are capable of providing a meaningful measure of culture from which hypotheses
regarding cultural differences may be developed and tested. Finally, it may be that the
limited amount of variance to be explained by country level variables was too small or
narrow to address with differences in cultural dimensions.

As mentioned earlier, individuals’ responses to the questionnaire items displayed
a fairly common trend across countries. Furthermore, it is likely that factors which
countries may differ on (e. g., language, religion, geographic location/climate, education,
gender/race diversity) account for some parts of that variance. Once those factors are
partialled out of the equation, there would appear to be a very small piece of the pie left to
be claimed by differences in culture. It may be the case that Hofstede’s theory of cultural
dimensions is not well suited for teasing apart extremely small differences between
cultures. Instead, power distance, individualism/collectivism, and uncertainty avoidance
may be more functional for use in studies when country or culture differences are likely to
be a larger source of variance. It would be interesting to know if this fact holds true for
alternative theories of culture as well.
Other Limitations

As has just been discussed, there are always potential problems with the use of
any theory of cultural differences. The choice to employ the dimensions of culture that

were developed by Hofstede was one that should have been made on an “objectively the

95

best measure of cultural differences” basis. However, the fact is that there currently does
not exist an objectively ideal way of delineating the differences between cultures. As a
result, the benefits and costs of using Hofstede’s dimensions had to be weighed and a
decision to use them or not be made. Confounded with this choice is the fact that no
information could be gathered from the current sample with respect to their actual
standings on these dimensions. As a result, there must be some concern about the
usefulness of using Hofstede’s cultural dimensions that were developed at a different
time, in a different organization, within the context of a vastly difference world order
given all the changes that have taken place in the last 20 years. Furthermore, is it
reasonable to assume that all of the differences in culture are explainable in terms of the
four (including masculinity/femininity) dimensions that Hofstede developed? Not really.
i There are likely many more discrete differences between cultures, and perhaps even more
broad differences that are yet to be effectively measured. While the ability of researchers
to address this issue is understandably limited, more improved theories of the differences
between cultures would be welcomed if the field of cross-cultural psychology is to
continue to deve10p.

The other obvious limitations to this study are in regard to the actual data
collection process and the sample. First of all, while the constructs addressed by the
questionnaire are not inadequate for use in academic studies, they could be greatly
improved by the addition of relevant questions in each of the construct areas.
Furthermore, item wording could be improved in order to limit the socially desirable

answer effect and hopefully increase variance in responses. The socially desirable

96

response effect may even be increased in the current administration of the questionnaire
by the use of computer administration that may cause concerns regarding anonymity. It
should be kept in mind that this is an organizational survey and positive attitude
responses by employees are not things that the company is hoping to avoid. To sum up
this first point, it would be more ideal if future research in this area incorporated more
control or input into the development of the questionnaire from which data is to be
gathered. This lack of control over construct and item development was a major
limitation to the current study.

A second limitation with regard to data collection was the sample itself. Everyone
in the sample was an employee in the same American-based organization. As was
discussed earlier, this fact may have important implications for the ability of national
culture to influence the responses to the survey or the relationships between attitude
variables. Instead, it may have been the case that the culture of the organization was
strong enough that it overrode any country based cultural differences that may have
otherwise affected the results of this study. Given that everyone was fiom this same
company, the lack of significant results in the study may not be so surprising. This is not
to say that the cultural differences suggested by Hofstede do not exist, even within global
companies. In fact, Hofstede developed his cultural dimensions based on only one
company and still found the differences between countries that allowed him to create his
cultural dimensions. However, when attempting to use those cultural dimensions to
suggest that culture will affect attitude variables that are developed within the

organizational context, the effects of organizational culture in nullifying differences due

97

to national culture should not be underestimated. Future studies would be improved if
samples from various companies, perhaps companies based in different nations or
companies known to have limited organizational culture, could be included in the sample.

Another sample issue deals with the fact that all of the employees whose
responses were used in this study were salaried employees. There may be something
about salaried employees that limits the effects of any influence that national culture may '
have on them. For instance, increased education levels, higher salaries that allow them
certain freedoms, and increased social status may serve to free these individuals from
identifying too strongly with the cultural “constraints” that are prevalent in their own
nation. It may be the case that hourly-wage workers who tend to be relegated to operating
within the system of a national culture may identify differently with that culture and may
in turn respond differently to the questionnaire items. It would be interesting to
investigate the cultural influences on that level of employees within organizations rather
than focusing only on the salaried employees.

A final limitation with regard to the sample was the inability to track employee
responses across time. This hindered the ability to investigate the stability of the
individual level relationships across time, and ruined any opportunity to ask any questions
regarding the influence of culture over time. For example, do individuals fi'om cultures
high on individualism who also report increasing empowerment over time, report
increasing amounts of job satisfaction more so than individuals from cultures low on
individualism? The tracking of attitudes across time and the tracking of responses to

cultural dimension items across time would benefit both the development of cultural

98

theory, and studies attempting to apply those theories to other research questions. That
being said, it remains difficult to conduct employee attitude surveying without the
guarantee to employees that their responses will be anonymous and confidential.
Conclusions

From a hypothesis standpoint, this study showed that there is reason to believe

that empowerment is related to stress and job satisfaction in different way across cultures.

Of specific interest is that fact that this study showed (in Hypotheses 10, and the
exploratory hypothesis) that collectivism moderated the relationships between both job
satisfaction and empowerment with levels of perceived stress (such that less stress was
perceived) suggesting that perhaps social support and group-connectedness may be ways
to alleviate at least some of the negative effects of stress on attitudes about the job.
However, the majority of the hypotheses were not supported, suggesting perhaps that
culture does not play nearly as big of a role in influencing attitudes about the job as some
would have us all believe. Taken as a whole, I walk away from this study with a few
conclusions in my mind that have little to do with the specific hypotheses.

To begin With, there is evidence that, despite using an arguably outdated set of
dimensions, culture does moderate the relationships between some attitude variables.
While the findings may be of little significance to the applied setting, from the standpoint
of understanding, it is hard to argue with a significant interaction effect. Those
significant findings, while not in the majority in this study, combine with the findings of
many of the researchers mentioned in this paper to provide some reason to believe that

cultural research is worth conducting. The existence of differences between cultures that

99

influence individual responses means that the book on cross-cultural psychology should
not be shut; the work is not done yet.

That being said, another conclusion to be drawn from this study is the difficulty in
conducting good cross-cultural research. As evidenced by the problems that arose during
this study with regard to the questionnaire and the sample, firturc researchers in this area
should make every effort to gain control of as much of the research process as possible.
To the extent that one can select an appropriate sample (e. g., many countries, diverse
organizations), control the development of the questionnaires (e.g., develop the items)
and assess the concurrent cultural attitudes at the same time, the study of cross-cultural
issues is sure to improve.

A third and final point about cross-cultural research is in regard to the status of
this sub-field itself. While researchers addressing questions such as those in this study
are of value to the overall compilation of knowledge about cross-cultural issues, there has
been too much of a tendency to run with the theories of culture that have already been
developed. Given the disagreements that exist in the literature about the best theories of
culture, the best ways to define the boundaries of culture, and even the actual definition of
culture, it seems like the best move forward would be to take a step back. I would like to
see researchers with the necessary connections, time, and resources who are conducting,
or would like to conduct cross-cultural research, return to a focus on the development of
cultural theory/dimensions and the testing of the usefulness of the those theories of
culture that already exist. Some amount of consensus in the literature would be an

improvement over the current status of cultural theory that seems to be based on personal

100

preference and whatever is the newest idea to get published. Furthermore, important
questions about culture do not seem to be currently addressed in the literature. One
example of this is the lack of evidence for the consistency of culture over time. What are
the effects of political, economic, or social change on the culture of a nation? Another
point of interest is the different experiences of culture depending on class/ status within a
society that was discussed earlier. Researchers have tended to group everyone within a
nation together in terms of culture without consideration of position within that nation’s
society. One last suggestion is that researchers attend to the question of national culture
versus organizational culture. To what extent do theories of culture based on social
interaction and general life questions apply to attitudes about the work place that can
sometimes seem like an entirely different world? Are employees working for companies
with weaker organizational cultures more influenced by country-based culture? These
and other important questions about cultural differences between groups of people need
to be more properly addressed before the field of cross-cultural psychology can move

forward in a meaningful way.

101

REFERENCES

Adler, N. J. (1984). Understanding the ways of understanding: Cross-cultural
management methodology reviewed. In R. N. Farmer (Ed.), Advances in international
comparative management, (V 01. 1, pp. 31-67). Greenwich, CT: J AI Press.

Alderfer, C. P. (1977). Group and intergroup relations. In J. R. Hackrnan & J. L. Suttle
(Eds), Improving life at work: Behavioral science approaches to organizational change.
(pp. 227-246). Santa Monica, CA: Goodyear.

Ali, A. (1988). A cross-national perspective of managerial work value systems. In R. N.
Farmer & McGoun, E. G. (Eds), Advances in International Comparative Management,
(Vol. 3, pp. 151-169). JAI Press.

Apply, M. H., & Trumbull, R. (1986). Dynamics of stress and its control. In M. H.
Appley, & R. Trumbull (Eds. ), Dynamics of stress: Physiological, psychological and
social perspectives (pp. 309-327). New York, NY: Plenum Press.

Arvey, R. D., Bouchard, t. 1., Sega], N. L., & Abraham, L. M. (1989). Job satisfaction:
Environmental and genetic components. Journal of Applied Psychology, 74, 187-192.

Bass, B. M. (1982). Individual capability, team performance, and team productivity. In
E. A. Fleishman & M. D. Dunnette (Eds.), Human performance and productivity: Human
capability assessment (pp. 179-232). Hillsdale, NJ: Erlbaurn.

Bateman, T. S., & Strasser, S. (1983). A cross-lagged regression test of the relationships
between job tension and employee satisfaction. Journal of Applied Psychology, 68(3),
439-445.

Begley, T. M., & Czajka, J. M. (1993). Panel analysis of the moderating effects of
commitment on job satisfaction, intent to quit, and health following organizational
change. Journal of Applied Psychology, 78, 552-556.

Berry, J. W. (1969). On cross-cultural comparability. International Journal of
Psychology, 4, 119-128.

Berry, J. W. (1989). Imposed etics-etics-derived etics: The operationalization of a
compelling idea. International Journal of Psychology, 24, 721-735.

Bhagat, R. S., O’Driscoll, M. P., Babakus, E., Frey, L., Chokkar, J ., Ninokumar, B. H.,
Pate, L. E., Ryder, P. A., Fernandez, M. J. 6., Ford, D. L., Jr., Mahanyele, M. (1994).
Organizational stress and coping in seven national contexts: A cross-cultural
investigation. In G.P. Keita, & Hurrell J. (Eds), Job stress in a changing workforce:

102

Investigating gender, diversity, and family issues. (pp. 93-105). Washington, DC:
American Psychological Association.

Brett, J. M., Tinsley, C. H., J anssens, M., Barsness, Z. I., & Lytle, A. L. (1997). New
approaches to the study of culture in industrial/organizational psychology. In P. C. Barley
& M. Erez (Eds), New perspectives on international industrial/organizational
psychology. San Francisco, CA: The New Lexington Press.

Brief, A. P. (1998). Attitudes in and Around Organizations. Thousand Oaks, London:
Sage.

Brief, A. P., Burke, M. 1., George, J. M., Robinson, 8., & Webster, J. (1988). Should
negative affectivity remain an unmeasured variable in the study of job stress? Journal of
Applied Psychology, 73, 193-198.

Brief, A. P., Butcher, A. B., & Roberson, L. (1995). Cookies, disposition, and job
attitudes: The effects of positive mood inducing events and negative affectivity on job

satisfaction in a field experiment. Organizational Behavior and Human Decision
Processes, 62, 55-62.

Brief, A. P., & Roberson, L. (1989). Job attitude organization: An exploratory study.
Journal of Applied Social Psychology, 19, 717-727.

Bryk, A. S. & Raudenbush, S. W. (1992). Hierarchical linear models. Newbury Park,
CA: Sage.

Buck, V. E. (1972). Working under pressure. London: Staples Press.

Cannon-Bowers, J. A., Oser, R., & Flanagan, D. L. (1992). Work teams in industry: A
selected review and proposed framework. In R. W. Swezey, & E. Salas (Eds), Teams:
Their training and performance (pp. 355-377). Norwood, NJ: Ablex Publishing.

Caplan, R. D., & Jones, K. W. (1975). Effects of work load, role ambiguity, and Type A
personality on anxiety, depression, and heart rate. Journal of Applied Psychology, 60(6),
713-719. -

Caplan, R. D., Naidu, P. K., & Tripathi, R. C. (1984). C0ping and defense:
Constellations vs. components. Journal of Health and Social Behavior, 25(3), 303-320.

Cavanaugh, M. A., Boswell, W. R., Roehling, M. V., & Boudreau, J. W. (2000). An
empirical examination of self-reported work stress among US. managers. Journal of
Applied Psychology, 85(1), 65-74.

Colquhoren, W. P. (1976). Accidents, injuries, and shift work. In P. G. Rentos & R. D.

Shepard (Eds), Shift work and health. Washington, DC: US. Government Printing
Office.

103

Cooper, C. L. (1984). Executive Stress: A ten-country comparison. Human Resource
Management, 23(4), 395-407.

Cooper, C. L., & Roden, J. (1985). Mental health and satisfaction among tax officers.
Social Science & Medicine, 21(7), 747-751.

Crampton, S. M. & Wagner, J. A. (1994). Percept-percept inflation in micro-
organizational research: An investigation of prevalence and effect. Journal of Applied
Psychology, 79(1), 67-75.

Cranny, C. J ., Smith, P. C., & Stone, E. F. (1992). Jobsatisfaction: How people feel
about their jobs and how it aflects their performance. New York: Lexington Press.

Crocker, J ., Luhtanen, R., Blaine, B., & Broadnax, S. (1994). Collective self-esteem and
psychological well-being among White, Black, and Asian college students. Personality
and Social Psychology Bulletin, 20, 503-513.

Cross, S. E. (1995). Self-construals, coping, and stress in cross-cultural adaptation.
Journal of Cross-Cultural Psychology, 26(6), 673-697.

Cummings, T. G. (1981). Designing effective work groups. In P. C. Nystrom & W. H.
Starbuck (Eds), Handbook of organizational design (V 01. 2). London: Oxford University
Press.

Cyert, R. M., & March, J. G. (1963). A behavioral theory of the firm. Englewood Cliffs,
NJ : Prentice-Hall.

Dawis, R. V., Lofquist, L. H., & Weiss, D. J. (1968). A theory of work adjustment
[Revision] (Minnesota Studies in Vocational Rehabilitation, No. 23). Minneapolis:
University of Minnesota.

Doran, L. 1., Stone, V. K., Brief, A. P., & George, J. M. (1991). Behavioral intentions as
predictors of job attitudes: The role of economic choice. Journal of Applied Psychology,
76, 40-45.

Barley, P. C., & Erez, M. (1997). New perspectives on international
industrial/organizational psychology. San Francisco, CA: The New Lexington Press.

Barley, P. C., & Gibson, C. B. (1998). Taking stock in our progress on
individualism/collectivism: 100 years of solidarity and community. Journal of
Management, 24(3), 265-304.

Erez, M. (1986). The congurence of goal-setting strategies with socio-cultural values and
its effect on performance. Journal of Management, 12(4) 585-592.

104

Erez, M. (1994). Toward a model of cross-cultural industrial and organizational
psychology. In H. C. Triandis, M. D. Dunnette, & L. M. Hough (Eds), Handbook of
industrial and organizational psychology vol. 4, (pp. 559-608). Palo Alto: Consulting
Psychologists Press, Inc.

Erez, M. & Arad, R. (1986). Participative goal setting: Social, motivational and
cognitive factors. Journal of Applied Psychology, 71(4), 591-599.

Erez, M, & Barley, P. C. (1987). Comparative analysis of goal setting strategies across
cultures. Journal of Applied Psychology. 72(4), 658-665.

Erez, M, & Barley, P. C. (1993). Culture, self-identity, and work. New York: Oxford
University Press.

Etzioni, A. (1975). A comparative analysis of complex organizations: On power,
involvement, and their correlates. New York, NY: Free Press.

Farrell, D., & Starnm, C. L. (1988). Meta-analysis of the correlates of employee absence.
Human Relations, 41, 211-227.

Franke, R. H., Hofstede, G., & Bond, M. H. (1991). Cultural roots of economic
performance: A research note. Strategic Management Journal, 12, 165-173.

French, J. R. P. Jr., & Caplan, R. (1972). Organization stress and individual strain. In A.
J. Marrow (Ed.), The failure of success, New York, NY: AMACOM.

French, J. R., Kay, E., & Meyer, H. H. (1966). Participation and the appraisal system.
Human Relations, 19(1), 3-20.

Frese, M., & Zapf, D. (1988). Methodological issues in the study of work stress:
Objective vs. subjective measurement of work stress and the question of longitudinal

studies. In C. L. Cooper & R. Payne (Eds), Causes, coping and consequences of stress
at work (pp. 375-409). Chichester, UK: John Wiley.

Fried, Y., & Ferris, G. R. (1987). The validity of the job characteristics model.
Personnel Psychology, 40, 287-322.

Frijda, N. H., & Mesquita, B. The social roles and fimctions of emotions. In S.
Kitayama, H. R. Markus (Eds), Emotion and culture: Empirical studies of mutual
influence (pp. 51-87).

Furnham, A. (1993). Just world beliefs in twelve societies. Journal of Social
Psychology, 133(3), 317-329.

Ganster, D. C., Fusilier, M., & Mayes, B. T. (1986). Role of social support in the
experience of stress at work. Journal of Applied Psychology, 71, 102-110.

105

Gerhart, B. (1987). How important are dispositional factors as determinants of job
satisfaction? Implications for job design and other personnel programs. Journal of
Applied Psychology, 72, 366-373.

Gladstein, D. L. (1984). Groups in context: A model of task group effectiveness
Administrative Science Quarterly, 29, 499-517.

Goffrnan, B. (1961). Asylums. Garden City, NY: Doubleday.

Gordon, M. E., Slade, A. L., & Schmitt, N. (1986). The “science of the sophomore”
revisited: From conjecture to empiricism. Academy of Management Review, 11(1), 191 -
207.

Graen, G. B., Hui, C., Wakabayashi, M., & Wang, Z.-M. (1997). Cross-cultural research
alliances in organizational research: Cross cultural partnership making in action. In P. C.
Barley & M. Erez (Eds), New perspectives on international industrial/organizational
psychology. San Francisco, CA: The New Lexington Press.

Griffeth, R. W., Horn, P. W., DeNisi, A. S., & Kirchner, W. K. (1985). A comparison of
different methods of clustering countries on the basis of employee attitudes. Human
Relations, 38(9), 813-840.

Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S., Nishida, T., & Karimi, H. (1994,
July). Measuring self-construals across cultures: A derived etic analysis. Paper
presented at the convention of the International Communication Association, Sydney.

Guzzo, R. A., & Salas, B. (1995). Team effectiveness and decision making in
organizations. San Francisco, CA: Jossey-Bass.

Guzzo, R. A., & Shea, G. P. (1992). Group performance and intergroup relations in
organizations. In M. A. Dunnette & L. M. Hough (Eds), Handbook of industrial and
organizational psychology. Palo Alto, CA: Consulting Psychologists Press.

Hackett, R. D., & Guion, R. M. (1985). A reevaluation of the absenteeism-
jobsatisfaction relationship. Organizational Behavior and Human Decision Processes,
35, 340-381.

Hackrnan, J. R. (1980). Work redesign and motivation. Professional Psychology
Research and Practice, 11(3), 445-455.

Hackrnan, J. R. (1987). The design of work teams. In J. W. Lorsch (Ed.), Handbook of
organizational behavior. Englewood Cliffs, NJ: Prentice-Hall.

Hackrnan, J. R. (Ed.). (1990). Groups that work (and those that don ’0: Creating
conditions for eflecfive teamwork. SanFrancisco: J ossey-Bass.

106

Hackrnan, J. R. (1992). Group influences on individuals in organizations. In M. D.
Dunnette, & L. M. Hough (Eds), Handbook of Industrial and Organizational Psychology
(2“cl ed., Vol 3). Palo Alto, CA: Consulting Psychologists Press.

Hackrnan, J. R., & Oldham, G. R. (1980). Work redesign. Reading, MA: Addison-
Wesley.

Hofrnann, D. A. (1997). An overview of the logic and rationale of hierarchical linear
models. Journal of Management, 23(6), 723-744

Hofstede, G. (1980). Culture’s consequences: International differences in work-related
values. Thousand Oaks, CA: Sage.

Hofstede, G. (1991). Culture and organizations: Software of the mind. London:
McGraw-Hill.

Hofstede, G. (1993). Cultural constraints in management theories. Academy of
Management Executive, 7(1), 81-94.

Holt, J ., & Keats, D. M. (1992). Work cognitions in multi-cultural interaction. Journal
of cross-cultural psychology, 123(4), 421-443.

Hoppe, M. H. (1998). Validating the masculinity/femininity dimension on elites from 19
countries. In G. Hofstede (Ed.), Masculinity and femininity. The taboo dimension of
national cultures. Cross-cultural psychology series (V 01. 3, pp. 29-43). Thousand Oaks,
CA: Sage.

Horvath, M., Ryan, A. M., & Ployhart, R. (1998). An exploration for global factors in
the ford pulse. Report to Ford Motor Company.

Horvath, M., Ployhart, R., & Ryan, A. M. (1998). Construct validation study of the
pulse. Report to Ford Motor Company.

House, R. J ., Hanges, P., & Ruiz-Quintanilla, A. (1994). Conference on Global
Leadership and Organizational Behavior (GLOBE). Calgary, Canada.

House, R. J ., & Hanges, P. (1997). Cross-cultural leadership and organizational research
program. The Global Leadership & Organizational Behavior Research Program. The
Warton School, University of Pennsylvania.

Hui, C. H., & Triandis, H. C. (1985). Measurement in cross-cultural psychology.
Journal of Cross-Cultural Psychology, 16, 131-152.

107

Hui, C. H., & Yee, C. (1999). The impact of psychological collectivism and workgroup
atmosphere on Chinese employees’ job satisfaction. Applied Psychology: An
International Review, 48(2), 175-185.

Hulin, C. L. (1991). Adaptation, persistence, and comrrritrnent in organizations. In M.
D. Dunnette & L. M. Hough (Eds), Handbook of industrial and organizational
psychology (Vol. 2, pp. 445-505). Palo Alto, CA: Consulting Psychologists Press.

Hulin, C. L., Roznowski, M., & Hachiya, D. (1985). Alternative opportunities and
withdrawal decisions: empirical and theoretical discrepancies and an integration.
Psychological Bulletin, 97, 233-250.

Hyatt, D. B., & Ruddy, T. M. (1997). An examination of the relationship between work
group characteristics and performance: Once more into the breech. Personnel
Psychology, 50, 553-585.

Iaffaldano, M. T., & Muchinsky, P. M. (1985). Job satisfaction and job performance: A
meta-analysis. Psychological Bulletin, 97, 251-273.

Ip, G. W. M., & Bond, M. H. (1995). Culture, values, and the spontaneous self-concept.
Asian Journal of Psychology, 1, 30-36.

Isabella, L. A. & Waddock, S. A. (1994). Top management team certainty:
Environmental assessments, teamwork, and performance implications. Journal of
Management, 20(4), 835-858.

Ivancevich, J ., & Matteson, M. T. (1980). Stress and work: A managerial perspective.
Glenview, IL: Scott Foresman.

Ivancevich, J ., Matteson, M. T., & Preston, C. (1982). Occupational stress, Type A
behavior, and physical well-being. Academy of Management Journal, 25(2), 373-391 .

Jackson, S. E. (1983). Participation in decision making as a strategy for reducing job-
related strain. Journal of Applied Psychology, 68(1), 3-19.

J ahoda, G., Krewer, B. (1997). History of cross-cultural and cultural psychology. J>
W. Berry, Poortinga, Y. H. (Eds), Handbook of cross-cultural psychology, (2u ed., pp.
1 -42).

Janz, B. D., Colquitt, J. A., & Noe, R. A. (1997). Knowledge worker team effectiveness:
The role of autonomy, interdependence, team development, and contextual support
variables. Personnel Psychology, 50, 877-904.

J ex, S. M., & Beehr, T. A. (1991). Emerging theoretical and methodological issues in the
study of work-related stress. Research in Personnel and Human Resources Management,
9, 3 1 1-517.

108

Jones, J. W. (1984). A cost evaluation for stress management. EAP Digest, 1, 34-39.

Judge, T. A., & Hulin, C. L. (1993). Job satisfaction as a reflection of disposition: A
multiple source causal analysis. Organizational Behavior and Human Decision
Processes, 56, 388-421.

Kahn, R. L., & Byosiere, P. (1992). Stress in organizations. In Dunnette, M. D., &
Hough, L. M. (Eds), Handbook of Industrial and Organizational Psychology (2“‘1 ed.,
Vol 3). Palo Alto, CA: Consulting Psychologists Press.

Kanter, R. M. (1989). The new managerial work. Harvard Business Review, 66, 85-92.

Kaufman, G. N., & Beehr, T. A. (1986). Interactions between job stressors and social
support: Some counterintuitive results. Journal of Applied Psychology, 71 (3), 522-526.

Kiggundu, M. N. (1983). Task interdependence and job design: Test of a theory.
Organizational Behavior and Human Performance, 31, 145-172.

Kirkcaldy, B. D., & Cooper, C. L. (1992). Cross-cultural differences in occupatioinal
stress among British and German managers. Work & Stress, 6(2), 177-190.

Kirkcaldy, B. D., Brown, J ., & Cooper, C. L. (1994). Occupational stress profiles of
senior police managers: Cross-cultural study of officers from Berlin and Northern
Ireland. Stress Medicine, 10, 127-130.

Kirkrnan, B. L., & Rosen, B. (1999). Beyond self-management: Antecedents and
consequences of team empowerment. Academy of Management Journal, 42(1), 58-74.

Klein, K. J. & Kozlowski, S. W. (2000). From micro to meso: Critical steps in
conceptualizing and conducting multilevel research. Organizational Research Methods,
3(3), 211-236.

Kluckhohn, F. R., & Strodtbeck, F. L. (1961). Variations in value orientations.
Westport, CT: Greenwood Press.

Koberg, C. 8., Boss, R. W., Senjem, J. C., & Goodman, E. A. (1999). Antecedents and
outcomes of empowerment: Empirical evidence from the health care industry. Group and
Organizational Management 24(1), 71-91.

Kobrin, S. J. (1984). International expertise in American business: how to learn to play
with the Icids on the street. New York, NY: Institute of International Education.

LaRocco, J. M., House, J. S., & French, J. R. P., Jr. (1980). Social support, occupational
stress and health. Journal of Health and Social Behavior, 21, 202-216.

109

Lawler, E. E., III. (1986). High-involovment management: Participative stratefies for
improving organizational performance. San Francisco: J ossey-Bass.

Lazarus, R. S. (1991). Psychological stress in the workplace. Journal of Social-Behavior
and Personality, 6(2), 1-13.

Levine, 1., & Stokes, J. P. (1989). Dispositional approach to job satisfaction: Role of
negative affectivity. Journal of Applied Psychology, 74, 752-758.

Lincoln, J. R., Hanada, M., & Olson, J. (1981). Cultural orientations and individual
reactions to organizations: A study of employees of J apanese-owned firms.
Administrative Science Quarterly, 26, 93-115.

Linderman, P. W. (1998). Performance outcomes related to the implementation of
teaming in a telecommunications organization (Doctoral dissertation) Dissertation

Abstracts International, 59( 1-A): 0231.

Locke, B. A. (1976). The nature and causes of job satisfaction. In M. D. Dunnette (Ed.),
Handbook of industrial and organizational psychology (pp. 1297-1349). Chicago: Rand
McNally.

Locke, E. A., Schweiger, D. M., & Latham, G. P. (1986). Participation in decision
making: When should it be used? Organizational Dynamics, 14(3), 65-79.

Magjuka, R. J ., & Baldwin, T. T. (1991). Tearn-based employee involvement programs
for continuous organizational improvement: Effects of design and administration.
Personnel Psychology, 44, 793-812.

Mangione, B. L., & Quinn, R. P. (1975). Job satisfaction, counterproductive behavior,
and drug use at work. Journal of Applied Psychology, 63, 114-116.

Mishra, A. K., & Spreitzer, G. M. (1998). Explaining how survivors respond to
downsizing: The roles, of trust, empowerment, justice & work re-design. Academy of
Management Review, 23(3) 567-588.

Misra, G., & Gergen, K. J. (1993). On the place of culture in the psychological sciences.
International Journal of Psychology, 28, 225-243.

Morgan, B. B., & Bowers, C. A. (1995). Teamwork stress: Implications for team
decision making. In Guzzo, R. A., & Salas, E. (Eds), Team eflectiveness and decision
mala'ng in organizations. (pp. 262-290). San Francisco, CA: J ossey-Bass.

Morgan, B. B., Jr., & Lassiter, D. (1992). Team composition and staffing. In R. W.

Swezey & E. Salas (Eds), Teams: Their training and performance (pp. 75-100).
Norwood, NJ: Ablex Publishing.

110

Mossholder, K. W., Bedeian, A. G., & Armenakis, A. A. (1982). Group process-work
outcome relationships: A note on the moderating impact of self-esteem. Academy of
Management Journal, 25(3), 575-585.

Mulder, M., Rietsema, Van Eck, J. R., & de Jong, R. D. (1971). An organization in crisis
and non-crisis situations. Human Relations, 24(1), 19-41.

Near, J. P. (1986). Work and non-work attitudes among Japanese and American workers.
In Advances in International Comparative Management, Greenwich, CT: J AI Press.

O’Driscoll, M. P., & Beehr, T. A. (1994). Supervisor behaviors, role stressors and
uncertainty as predictors of personal outcomes for subordinates. Journal of
Organizational Behavior, 15, 141-1 55.

Organ, D. W., & Ryan, K. (1995). A meta-analytic review of attitudinal and
dispositional predictors of organizational citizenship behavior, Personnel Psychology, 48,
775-802.

Parsons, T., & Shils, B. A. (1951). Toward a general theory of action. Cambridge, MA:
Harvard University Press.

Peterson, M. F., & Smith, P. B. (1995). Role conflict, ambiguity, and overload: A 21-
nation study. Academy of Management Journal, 38(2), 429-452.

Peterson, M. F., & Smith, P. B. (1997). Does national culture or ambient temperature
explain cross national differences in role stress? No sweat! Academy of Management
Journal, 40(4), 930-946.

Petty, M. M., McGee, G. W., & Cavender, J. W. (1984). A meta-analysis of the
relationships between individual job satisfaction and individual performance. Academy
of Management Review, 9, 712-721.

Pike, K. L. (1966). Language in relation to a unified theory of the structure of human
behavior. The Hague: Mouton.

Pulakos, B. D., & Schmitt, N. (1983). A longitudinal study of a valence model approach
for the prediction of job satisfaction of new employees. Journal of Applied Psychology,
68, 307-312.

Rhodewalt, F., & Agustsdottir, S. (1984). On the relationship of hardiness to the Type A
behavior pattern: Perception of life events versus coping with life events. Journal of
Research in Personality, 18(2), 21 1-223.

Robert, C., Probst, T. M., Martocchio, J. J., Drasgow, F., & Lawler, J. J. (2000).
Bmpowerrnent and continuous improvement in the United States, Mexico, Poland, and

111

India: Predicting fit on the basis of the dimensions of power distance and individualism.
Journal of Applied Psychology, 85(5), 643-658.

Rodrigues, C. A. (1990). The situation and national culture as contingencies for
leadership behavior: Two conceptual models. In Advances in International Comparative
Management, Greenwich, CT: J AI Press.

Ronen, S. & Kraut, A. I. (1977). Similarities among countries based on employee work
values and attidudes. Columbia Journal of World Business, 12(2), 89-96.

Ronen, S. & Shenkar, O. (1985). Clustering countries on attitudinal dimensions: A
review and synthesis. Academy of Management Review, 10, 43 5-454.

Ruddy, D. E., & Hyatt, J. M. (1997). An examination of the relationship between work
group characteristics and performance: Once more into the breech. Personnel
Psychology, 50(3), 553-585".

Sagie, A., & Koslowsky, M. (1994). Organizational attitudes and behaviors as a function
of participation in strategic and tactical change decisions: An application of path goal
theory. Journal of Organizational Behavior, 15(1), 37-47.

Salas, B., Dickinson, T. L., Converse, S. A., & Tannenbaum, S. I. (1992). Toward an
understanding of team performance and training. In R. W. Swezey, & B. Salas (Eds),
Teams: Their training and performance (pp. 3-29). Norwood, NJ: ABLBX.

Sashkin, M. (1984). Participative management is an ethical imperative. Organizational
Dynamics, Spring, 5-22.

Schmidt, S. M., & Yeh, R. S. (1992). The structure of leader influence: A cross-national
comparison. Journal of Cross-Cultural Psychology, 23(2), 251-264.

Schneider, B., Goldstein, H. W., & Smith, B. D. (1995). The ASA framework: An
update. Personnel Psychology, 48(4), 747-773.

Schwartz, S. H. (1990). Individualism-collectivism: Critique and prposed refinements.
Journal of Cross-Cultural Psychology, 21, 139-157.

Schwartz, S. H. (1992). Universals in the content and structure of values: theoretical
advances and empirical tests in 20 countries. In M. Zanna (Ed.), Advances in
Experimental Social Psychology (Vol. 25, pp. 1-65). Orlando: Academic Press.

Schwartz, S. H. (1994). Beyond individualixm and collectivism: New cultural
dimensions of values. In U. Kim, H. C. Triandis, C. Kagitcibasi, S. C. Choi, & G. Yoon
(Eds), Individualism and collectivism: Theory, method, and applications (pp. 85-119).
Thousand Oaks, CA: Sage.

112

Schwartz, 8. H. & Bilsky. (1987). Toward a universal psychological structure of human
values. Journal of Personality & Social Psychologi, 53(3), 550-562.

Schwartz, S. H. & Bilsky. (1990). Toward a theory of the universal content and structure
of values: Extensions and cross-cultural replications. Journal of Personality and Social
Psychology, 58, 878-891.

Sego, D. J ., Hui, C., & Law, K. S. (1997). Operationalizing cultural values as the mean
of individual values: Problems and suggestions for research. In P. C. Barley & M. Erez
(Eds), New perspectives on international industrial/organizational psychology. San
Francisco, CA: The New Lexington Press.

Shea, G. P., & Guzzo, R. A. (1987). Groups as human resources. In K. M. Rowland, &
G. R. Ferris (Eds), Research in human resources and personnel management (V 01. 5, pp.
323-356). Greenwich, CT: JAI Press.

Shweder, R. A. (Ed.) & Levine, R. A. (Ed.). (1985). Culture theory. Cambridge,
England: Cambridge University Press.

Simon, J ., & Carey, K. B. (1998). A structural analysis of attitudes toward alcohol and
marijuana use. Personality and Social Psychology Bulletin, 24, 727-736.

Slocum, J. W., Jr., & Topichak, P. M. (1972). Do cultural differences affect job
satisfaction? Journal of Applied Psychology, 56, 177-178.

Smith, P. C., Kendall, L. M., & Hulin, C. L. (1969). The measurement of satisfaction in
work and retirement: A strategy for the study of attitudes. Chicago: Rand McNally.

Smith, P. B., & Misumi, J. (1989). Japanese management: A sun rising in the West? In
C. L. Cooper & I. T. Robertson (Eds), Intematioinal review of industrial and
organizational psychology, 1989 (pp. 330-369). Chichester, UK: John Wiley.

Sondergaard, D. M. (1994). Research note: Hofstede’s consequences: A study of
reviews, citations and replications. Organizational Studies, 15, 447-456.

Spector, P. E., & Wimalasiri, J. (1986). A cross-cultural comparison of jab satisfaction
dimensions in the United States and Singapore. International Review of Applied
Psychology, 35, 147-158.

Spreitzer, G. M. (1995). Psychological empowerment in the workplace: Dimensions,
measurement, and validation. Academy of Management Journal, 38, 1442-1465.

Spreitzer, G. M., Kizilos, M. A., & Nason, S. W. (1997). A dimensional analysis of the
relationship between psychological empowerment and effectiveness, satisfaction, and
strain Journal of Management, 23, 679-704.

113

Staw, B. M., Bell, N. E., & Clausen, J. A. (1986). The dispositional approach to job
attitudes: A lifetime longitudinal test. Administrative Science Quarterly, 31, 56-77.

Staw, B. M., & Ross, J. ( 1985). Stability in the midst of change: A dispositional
approach to job attitudes. Journal of Applied Psychology, 70, 469-480.

Sundstrom, B., DeMuse, K., & Futrell, D. (1990). Work teams: Applications and
effectiveness. American Psychologist, 45, 120-133.

Sutherland, L. F ., Fogerty, G. J ., & Pithers, R. T. (1995). Congruence as a predictor of
occupational stress. Journal of Vocational Behavior, 46(3), 292-309.

Tannenbaum, S. I., Beard, R. L., & Salas, E. (1992). Team building and its influence on
team effectiveness: An examination of conceptual and empirical developments. In K.
Kelley (Ed), Issues, theory, and research in industrial/organizational psychology (pp.
117-153). Amsterdam, Holland: Elsevier.

Tate, R. L. & Wongbundhit, Y. (1983). Random versus nonrandom coefficient models
for multilevel analysis. Journal of Educational Statistics, 8, 103-120.

Tetrick, L. E., LaRocco, J. M. (1987). Understanding, prediction, and control as
moderators of the relationships between perceived stress, satisfaction, and psychological
well being. Journal of Applied Psychology, 72(4), 538-543.

Thomas, K. W., & Tymon, W. G., Jr. (1994). Does empowerment always work:
Understanding the role of intrinsic motivation and personal interpretation. Journal of
Management Systems, 6, 39-54.

Thomas, K. W., & Velthouse, B. A. (1990). Cognitive elements of empowerment: An

“interpretive” model of intrinsic task motivation. Academy of Management Review, 15,
666-681.

Triandis, H. C. (1980). Values, attitudes, and interpersonal behavior. In H. B. Howe &
M. M. Page (Eds), Nebraska Symposium on Motivation, 1979 (pp. 41-133). Lincoln:
University of Nebraska Press.

Triandis, H. C. (1990). Toward cross-cultural studies of individualism and collectivism
in Latin America. Revista Interamericana de Psicologia, 24(2) 199-210.

Triandis, H. C. (1993). Collectivism and individualism as cultural syndromes. Cross-
cultural research: The journal of comparative social science, 27(3-4) 155-180.

Triandis, H. C. (1997). Culture theory and the meaning of relatedness. In P. C. Barley &
M. Erez (Eds), New perspectives on international industrial/organizational psychology.
San Francisco, CA: The New Lexington Press.

114

Triandis, H. C. & Bhawuk, D. P. S. (1997). Culture theory and the meaning of
relatedness. In Earley, C. P. & Erez, M. (Eds), New perspectives on international
industrial/organizational psychology. (pp. 13-52). San Francisco, CA: The New
Lexington Press/Jossey-Bass Inc.

Triandis, H. C., Bonternpo, R., Villareal, M. J ., Masaaki, A., & Lucca, N. (1988).
Individualism and collectivism: Cross-cultural perspective on self-ingroup relationships.
Journal of Personality and Social Psychology, 54, 328-338.

Triandis, H. C., McCusker, C., Betancourt, H., & Iwao, S. (1993). Journal of Cross-
cultural psychology, 24(3), 366-383.

Triandis, H. C. (1994). Culture: Theoretical and methodological issues. In H.C.
Triandis, M. D. Dunnette, & Hough, L. (Eds), Handbook of Industrial and
Organizational Psychology (2“d ed., Vol. 4). Palo Alto, CA: Consulting Psychologists
Press.

Urban, J. M., Bowers, C. H., Monday, S. D., & Morgan, B. B. (1995). Workload, team
structure, and communication in team performance. Military Psychology, 7(2), 123-139.

Van de Vliert, E. & Van Yperen, N. (1996). Why cross-national differences in role
overload? Don’t overlook ambient temperature! Academy of Management Journal, 39(4),
986-1 004.

Wall, T. D., Demp, N. J., Jackson, P. R., & Clegg, C. W. (1986). Outcomes of
autonomous workgroups: A long-term field experiment. Academy of Management
Journal, 29, 281-304.

Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: A theoretical
discussion of the structure, causes and consequences of affective experiences at work. In
Staw, B. M. & Cummings, L. L. (Eds), Research in organizational behavior: An annual
series of analytical essays and critical reviews, Vol. 18. (pp. 1-74). Stamford, CT: JAI
Press, Inc.

Weiss, H. M., Nicholas, J. P., & Daus, C. S. (1999). An examination of the joint effects
of affective experiences and job beliefs on job satisfaction and variations in affective
experiences over time. Organizational Behavior and Human Decision Processes, 78(1),

1-24.

Weitz, J. (1952). A neglected concept in the study of job satisfaction. Personnel
Psychology, 5, 201-205.

Williams, J. M., & Stout, J. K (1985). The effect of high and low assertiveness on locus
of control and health problems. Journal of Psychology, 119(2), 169-173.

115

APPENDIX A

SCALE ITEMS

Job Satisfaction (all years)
Considering everything, how satisfied are you with your job?

Considering everything, how would you rate your overall satisfaction in the
company at the present time?
I feel valued as an employee of the company.

My job makes good use of my skills and abilities.
How satisfied are you with the recognition you receive for doing a good job?

Stress (1996. 1998 & 1999 items are noted)

I do not feel excessive work-related stress. (all years)

Work-related stress does not interfere with doing my job well. (all years)
I do not feel excessive work-related stress. (all years)

The amount of work I am expected to do on my job is: (96)

Sufficient efforts are being made to manage work-related stress in my
Workgroup. (98 & 99)
Work-related stress does not affect my job satisfaction. (98 & 99)

My workload allows me to satisfy my customers’ requirement. (98 & 99)

The workload is distributed fairly among the people in my workgroup. (98 & 99)
My workload does not interfere with my ability to do a quality job. (98 & 99)
Normally, I am able to do my job during regular business hours. (98 & 99)

kamwork (1996gnd 1998 items are noted)
The people I work with cooperate to get the job done. (all years).

Diversity among employees (e.g., race, sex, nationality, age, background,
personality, thinking style) is valued in my workgroup. (96)
My workgroup receives adequate feedback fi'om internal customers. (all years).

There is close cooperation among departments to achieve quality. (all years).
I know my departrnent’s objectives (quality, cost, timing, etc.). (all years).
In my workgroup, expectations about the quality of our work are clear. (96).

My workgroup used feedback from our internal customers to improve the quality
of our work (all years).

116

Empgwerment (all years)

My supervisor communicates clear measures of accomplishment.

My supervisor encourages decisions to be made at the lowest appropriate
levels in the organization.

My supervisor gives employees the necessary authority to accomplish tasks.

117