Profession Comm. 8

Societal-Level Versus Individual-Level Predictions of Ethical Behavior: A 48-Society Study of Collectivism and Individualism

David A. Ralston • Carolyn P. Egri • Olivier Furrer • Min-Hsun Kuo •

Yongjuan Li • Florian Wangenheim • Marina Dabic • Irina Naoumova •

Katsuhiko Shimizu • Marı́a Teresa de la Garza Carranza • Ping Ping Fu •

Vojko V. Potocan • Andre Pekerti • Tomasz Lenartowicz • Narasimhan Srinivasan •

Tania Casado • Ana Maria Rossi • Erna Szabo • Arif Butt • Ian Palmer •

Prem Ramburuth • David M. Brock • Jane Terpstra-Tong • Ilya Grison •

Emmanuelle Reynaud • Malika Richards • Philip Hallinger • Francisco B. Castro •

Jaime Ruiz-Gutiérrez • Laurie Milton • Mahfooz Ansari • Arunas Starkus •

Audra Mockaitis • Tevfik Dalgic • Fidel León-Darder • Hung Vu Thanh •

Yong-lin Moon • Mario Molteni • Yongqing Fang • Jose Pla-Barber •

Ruth Alas • Isabelle Maignan • Jorge C. Jesuino • Chay-Hoon Lee •

Joel D. Nicholson • Ho-Beng Chia • Wade Danis • Ajantha S. Dharmasiri •

Mark Weber

Received: 29 August 2012 / Accepted: 8 May 2013 / Published online: 4 June 2013

� Springer Science+Business Media Dordrecht 2013

Abstract Is the societal-level of analysis sufficient today

to understand the values of those in the global workforce?

Or are individual-level analyses more appropriate for

assessing the influence of values on ethical behaviors

across country workforces? Using multi-level analyses for

a 48-society sample, we test the utility of both the societal-

level and individual-level dimensions of collectivism and

individualism values for predicting ethical behaviors of

business professionals. Our values-based behavioral anal-

ysis indicates that values at the individual-level make a

more significant contribution to explaining variance in

ethical behaviors than do values at the societal-level.

D. A. Ralston (&) University Fellow International Research Consortium, Ft. Myers,

FL, USA

e-mail: [email protected]

C. P. Egri

Simon Fraser University, Burnaby, BC, Canada

O. Furrer

University of Fribourg, Fribourg, Switzerland

M.-H. Kuo

National Central University, Zhongli, Taiwan

Y. Li

Chinese Academy of Sciences, Beijing, China

F. Wangenheim

ETH Zurich, Zurich, Switzerland

M. Dabic

University of Zagreb, Zagreb, Croatia

I. Naoumova

University of Hartford, Hartford, CT, USA

K. Shimizu

Keio University, Yokohama, Japan

M. T. de la Garza Carranza

Instituto Tecnológico de Celaya, Celaya, Mexico

P. P. Fu

Chinese University of Hong Kong, Hong Kong SAR, China

V. V. Potocan

University of Maribor, Maribor, Slovenia

A. Pekerti

University of Queensland, Brisbane, QLD, Australia

T. Lenartowicz

Florida Atlantic University, Boca Raton, FL, USA

N. Srinivasan

University of Connecticut, Storrs, CT, USA

T. Casado

University of São Paulo, São Paulo, Brazil

123

J Bus Ethics (2014) 122:283–306

DOI 10.1007/s10551-013-1744-9

Implicitly, our findings question the soundness of using

societal-level values measures. Implications for interna-

tional business research are discussed.

Keywords Cultural values � Influence ethics � Hierarchical linear modeling � Collectivism � Individualism

Is the use of societal-level values for cross-cultural analy-

ses both acceptable and sufficient in today’s global econ-

omy? To begin to address this question, we examine the

extent to which values predict the ethical behaviors of

16,229 business professionals from 48 societies. Specifi-

cally, we conducted multi-level analyses to simultaneously

assess relationships at the societal and individual levels of

analysis for the collectivism and the individualism values

dimensions of the cross-culturally validated Schwartz val-

ues survey (SVS) construct (Schwartz 1992) with the four

dimensions (pro-organizational, image management, self-

serving, and maliciously intended) of the cross-culturally

validated subordinate influence ethics (SIE) construct

(Ralston and Pearson 2010).

We begin by briefly reviewing the study constructs and

providing an overview of the debate on appropriate levels

of analysis for predicting the behavior of professionals in

the global workforce. Having thus framed our research

question, we present the study methods and results. We

conclude with a discussion of reasons why our findings

support using individual-level analyses, as well as our

observations on future directions in work values research.

Overview of the Literature

Ethical Behavior in Organizations

Ethics has been defined as ‘‘…the discipline that examines one’s moral standards’’ (Alas 2006, p. 238). As such, ethics

are the standards of appropriate conduct that individuals

use to guide decisions in both their work and non-work

environments (Ralston et al. 2009). In the organizational

context, ‘‘ethical behavior’’ is an encompassing category

that includes and/or relates to an array of behaviors that

occur in organizational settings, such as leadership, fol-

lowership, organizational citizenship, decision-making,

and communication (Collins 2000; Trevino et al. 2006).

However, most cross-national research on ethical judg-

ments and values has focused on the normative aspects of

ethical beliefs rather than the individual-level driving for-

ces of managerial attitudes regarding what is ethical (e.g.,

Forsyth et al. 2008). While the link between moral phi-

losophy and established cultural values dimensions (e.g.,

individualism and collectivism) needs further exploration,

it is apparent that within and between countries, there is

variance in the extent to which people and organizations

engage in ethically questionable behavior.

Whereas behavior in organizations may be viewed as

ranging from highly ethical to highly unethical, much of

the organizational research has focused either on the ethical

or the unethical ends of the continuum (e.g., Forsyth et al.

2008; Kish-Gephart et al. 2010; Martin et al. 2007). In this

study, we provide a broad perspective on both ethical and

A. M. Rossi

International Stress Management Association, Porto Alegre,

Brazil

E. Szabo

Johannes Kepler University, Linz, Austria

A. Butt

Lahore University of Management Sciences, Lahore, Pakistan

I. Palmer

RMIT University, Melbourne, VIC, Australia

P. Ramburuth

University of New South Wales, Sydney, NSW, Australia

D. M. Brock

Ben-Gurion University of the Negev, Beer-Sheva, Israel

J. Terpstra-Tong

Monash University, Bandar Sunway, Malaysia

I. Grison

University of Westminster, London, UK

E. Reynaud

IAE Marseille Université, Aix-en-Provence, France

M. Richards

Pennsylvania State University, Reading, PA, USA

P. Hallinger

Hong Kong Institute of Education, Hong Kong, China

F. B. Castro

CEMPRE-Universidade do Porto, Porto, Portugal

J. Ruiz-Gutiérrez

Universidad de los Andes, Bogotá, Colombia

L. Milton

University of Calgary, Calgary, AB, Canada

M. Ansari

University of Lethbridge, Lethbridge, AB, Canada

A. Starkus

CIBER-Vilnius, Vilnius, Lithuania

A. Mockaitis

Monash University, Melbourne, VIC, Australia

T. Dalgic

University of Texas at Dallas, Richardson, TX, USA

284 D. A. Ralston et al.

123

unethical behaviors that engenders both theory generation

(Martinko et al. 2002) and a better understanding of indi-

vidual ethicality across global workforces (Ralston et al.

2009). And, as we know, organizations, per se, do not plan,

or make decisions, or lead, or follow or communicate; it is

the individuals in organizations who engage in these

behaviors. Consequently, individual ethical behavior is

relevant to numerous aspects of organizational life that

involve human capital (Gratton 2000; Painter-Morland and

Ten Bos 2011; Trevino et al. 1999). Consequently, under-

standing the relationship between individual values (e.g.,

collectivism and individualism) and ethical behavior (e.g.,

pro-organizational, image management, self-serving and

maliciously intended) is very important for understanding

work behavior in organizations. Thus, the overarching goal

of this study is to take a twenty-first century, globally ori-

ented, multi-level perspective of the contributions that the

societal and individual levels of analysis bring to under-

stand the ethicality of work behaviors in organizations.

The Relationships of Collectivism and Individualism

Values with Ethical Behavior

‘‘Values are multifaceted standards that guide conduct in a

variety of ways. They lead us to take particular positions on

social issues and they predispose us to favor one ideology

over another. They are standards employed to evaluate and

judge others and ourselves’’ (Rokeach 1973, p. 79). Lind-

eman and Verkasalo (2005) also note that values make a

unique contribution to understanding psychological phe-

nomena that connect to the evaluation, justification or

selection of actions.

In the work environment, managers’ values have been

found to be predictive of a variety of ethics-based behav-

iors including leadership (Illies and Reiter-Palmon 2008;

Offermann and Hellmann 1997; Voegtlin et al. 2012),

cooperation (Chen et al. 1998), organizational citizenship

(Kabasakal et al. 2011; Kirkman et al. 2009), influence (Fu

et al. 2004; Ralston et al. 2009), and work-related per-

ceptions and decisions (Erez and Earley 1987; Wagner

1995). Thus, understanding the values that businesspeople

hold is directly relevant for understanding the ethicality of

their work behaviors.

Across a variety of values typologies, the two values of

collectivism and individualism have been predominant

(Chhokar et al. 2008; Hofstede 2001; Kluckhohn and

Strodtbeck 1961; Oyserman et al. 2002; Ralston et al.

1999b; Schimmack et al. 2005; Schwartz 1994). Based on

their review of the literature, Husted and Allen (2008)

concluded that collectivism and individualism affect ethi-

cal behavior more than any other cultural dimensions

because they most directly deal with ‘‘…the way people resolve conflicts in human interests and optimize mutual

benefits’’ (p. 294). In essence, these two values determine

how individuals prioritize and weigh the importance of

self- and group-interests, which in turn has ethical impli-

cations for decisions and behaviors undertaken (Robertson

and Fadil 1999; Robertson et al. 2012; Vitell et al. 1993).

Specific to the impact of the collectivism and individ-

ualism values on ethics-based behaviors in organizations,

Earley (1993) and Erez and Somech (1996) studied how

these values relate to individual and group performance at

work. Subsequently, Oyserman et al. (2002) reported that

one contrast between collectivists and individualists is their

F. León-Darder � J. Pla-Barber University of Valencia, Valencia, Spain

H. V. Thanh

National Economics University, Hanoi, Vietnam

Y. Moon

Seoul National University, Seoul, South Korea

M. Molteni

Catholic University of Milan, Milan, Italy

Y. Fang

University of Canberra, Canberra, ACT, Australia

R. Alas

Estonia Business School, Tallinn, Estonia

I. Maignan

VU University Amsterdam, Amsterdam, The Netherlands

J. C. Jesuino

Instituto Superior de Ciencias do Trabalho e da Empresa,

Lisbon, Portugal

C.-H. Lee

Keppel Offshore & Marine, Singapore, Singapore

J. D. Nicholson

San Francisco State University, San Francisco, CA, USA

H.-B. Chia

National University of Singapore, Singapore, Singapore

W. Danis

University of Victoria, Victoria, BC, Canada

A. S. Dharmasiri

Postgraduate Institute of Management, Colombo, Sri Lanka

M. Weber

Argosy University-Twin Cities, Eagan, MN, USA

Societal-Level Versus Individual-Level Predictions 285

123

conflicting priorities between maintaining good group

relations and completing work tasks, with collectivists

being more concerned about maintaining relationships and

individualists being more concerned about task completion.

Collectivism and individualism values have also been

argued to be associated with the cultural orientations of

pre-industrial and industrial economies, respectively (In-

glehart and Welzel 2010). Further, other research has

shown that societal-level collectivism/individualism

impacts the extent to which individuals use their peers as

primary referents for ethical decision-making (Westerman

et al. 2007). Although the values orientation of study par-

ticipants or their peers were not directly assessed, this

study’s findings argue both for the importance of values

and for the importance of individual influence on ethical

decision-making. Another study by Cullen et al. (2004)

found the cultural values of individualism and achievement

orientation to be negatively related to managers’ willing-

ness to justify ethically suspect behavior. Thus, we focus

on collectivism and individualism values first and foremost

because previous research has consistently found these

values to be relevant antecedents of ethical behavior.

Second, these values are the most commonly used

dimensions to differentiate cultures, groups, and individu-

als (Ralston 2008; Schimmack et al. 2005; Triandis 1995).

Another particularly relevant factor for our study is Ralston

et al.’s (2011) 50-country assessment of the individual-

level SVS dimensions (Schwartz 1994) that identified the

collectivism and individualism measures to be two robust

values dimensions for the study of business professionals.

The Societal-Level vis-à-vis the Individual-Level

of Analysis

Are societal-level or individual-level analyses more pre-

dictive of the relationships between values and ethics-

based phenomena? We raise this question as a continuation

of previous discussions on the appropriate level of analysis

(e.g., Au and Cheung 2004; Bond et al. 2004; Lenartowicz

and Roth 2001; Tung and Verbeke 2010) and the relevance

of multi-level analyses (e.g., Peterson et al. 2012; Ralston

et al. 2009; Tsui et al. 2007). To address this question, we

employ the framework introduced by Klein and Kozlowski

(2000) to discuss the various properties of a group (e.g., a

country).

In their conceptualization of the group, Klein and

Kozlowski (2000) identify three types of properties that a

group may possess: global, shared and configural. Global

properties encompass the properties that are most objec-

tively recognizable (e.g., GDP per capita or political sys-

tem of a country). While shared and configural properties

each emerge from the characteristics of the members of the

group (e.g., country), the shared properties of the group are

‘‘shared’’ or embraced by all members of the group,

whereas configural properties are not shared by all mem-

bers of the group. Configural properties may be classified

as differentiations in the group composition that are caused

by either meso-level (e.g., ethnic group or region) or

individual-level (e.g., age or gender) differences.

Over the past several decades, the norm in cross-cultural

research appears to have been to rely heavily on the global

orientation, as is reflected in the Klein and Kozlowski (2000)

typology. To this point, Tsui et al. (2007, p. 461) noted in

their review of cross-cultural studies that: ‘‘It is curious that

culture researchers continue to treat culture as a global

property by using nation as a proxy or assume a shared

property of culture by using mean scores of culture values.’’

Likewise, Au and Cheung (2004, p. 1339) observed that

‘‘…the dispersion of individuals within a culture is not often the focus of international management compared to the

[shared] cultural mean.’’ Similarly, Kirkman et al. (2006,

p. 313) report that ‘‘the relatively low amount of variance

explained by the cultural values in many studies underscores

the existence of the many other forces besides culture that

determine the behavior and attitudes of individuals in soci-

eties.’’ And in their study of employee-manager relations in

China and the U.S., Kirkman et al. (2009) concluded that the

old adage, when in Rome do as the Romans do, should likely

be revised to ‘‘When in Rome, get to know Romans as

individuals’’ (p. 757). Finally, based on their meta-analysis

of 598 cultural values studies, Steel and Taras (2010, p. 211)

reported that ‘‘up to 90 % of the variance in cultural values is

found to reside within countries, stressing that national

averages poorly represent specific individuals.’’ The con-

sensus from these statements is that a case can be made that

researchers should consider looking more to configural

properties to better understand organizational phenomena.

While within-country differences have long been recog-

nized to be important (e.g., Au 2000; Wallace 1970), only a

few cross-cultural studies have taken the step to adopt a

configural perspective by directly conducting within-society

analyses across multiple countries (e.g., Au and Cheung

2004; Egri and Ralston 2004; Fischer et al. 2011; Fu et al.

2004; Gurven et al. 2008; Lenartowicz and Roth 2001;

Ralston et al. 1996). The findings of these studies demon-

strate the need to explore micro- and/or meso-level differ-

ences within societies to fully understand the behaviors of

those in the workforce.

Given these findings one might ask why have

researchers continued to employ societal-level cultural

values analyses? The answer appears to be multi-faceted.

In part, it has been due to methodological research issues

(Fischer 2009); and in part, it also appears to have been due

to a lack of appreciation of the contribution that individual-

level analyses bring to our understanding of organizational

phenomena (Au 1999; Buchholz et al. 2009; Tung 2008).

286 D. A. Ralston et al.

123

Thus, we next examine both the methodological research

issues and the relevance of the individual-level of analysis.

Methodological Research Issues

In the 1960s and 1970s, cross-cultural management was a

burgeoning field of inquiry seeking answers to fundamental

questions about value differences (Kluckhohn and Strodt-

beck 1961). The values dimensions developed by Hofstede

(2001) constituted the first major breakthrough toward

answering ‘‘values differences’’ questions. However, due to

methodological design limitations, the four Hofstede

dimensions have been considered to be valid only at the

societal-level (Spector and Cooper 2002), if valid at all

(McSweeney 2002; Spector et al. 2001b). In respect to the

validity of Hofstede’s values survey module (VSM),

Spector et al.’s (2001a) cross-national study reported that

the Hofstede individualism dimension measure had unac-

ceptably low scale reliabilities (Cronbach a). Specifically, 16 of the 24 countries (67 %) had individualism a levels below .60 (with the U.S. a at .26), while other variables in their study (e.g., job satisfaction) had a levels that were consistently above the generally accepted .70 level. The

subsequent GLOBE project constituted another major

effort to develop societal-level values measures applicable

for the study of global workforces (House et al. 2004).

However, this international project has also received its

share of methodological criticism (e.g., Peterson 2004;

Peterson and Castro 2006; Taras et al. 2010b). Nonetheless,

in spite of the methodological concerns raised regarding

these two dominant societal-level measures of values, the

consensus appears to be that societal-level analyses were

the only viable approach to use, at least until recently.

However, that thinking may be ready for change.

Relevance of Analysis at the Individual-Level

Individual-level measures of values can be traced back to

the work of Rokeach (1973), and the literature has shown

that the behavior of individuals is best predicted by

studying the demographic aspects of the individual (e.g.,

Bielby 2000; Egan and Bendick Jr. 2008). Further, previ-

ous research has shown that individually held values

influence a variety of individuals’ behaviors at work (e.g.,

Gelfand et al. 2007; Tsui et al. 2007). Although individuals

have characteristics that can be attributed to their societal

culture, ultimately, individuals are individuals.

Inherent Within-Country Differences

To exemplify differences in values orientations within a

society and across societies, we consider two individual-

level factors: age and gender. Inherent within every

country, these individual-level differences significantly

influence the values and subsequent work behaviors of

individuals across the global workplace. While age and

gender are certainly not the only two individual-level

factors that shape values/behaviors, with others including

education, cognitive ability, and occupation, they do tend

to be the most frequently studied factors that identify val-

ues differences in societies (Peng and Lin 2009; Taras et al.

2010a).

Age differences have been approached primarily from

two theoretical perspectives: life stage theory and genera-

tion subculture theory. Both approaches cluster individuals

by age (birth year); however, they employ different criteria

for categorization. Life stage theory proposes that there is a

universally consistent pattern of human development over

the life of an individual (Erikson 1968), and that this pat-

tern consists of four stages, with young adulthood

(20–39 years old) and middle adulthood (40–59 years old)

being the most pertinent for a discussion of business pro-

fessionals (Erikson 1997; Ralston et al. 1999a; Settersten

and Mayer 1997). Young adulthood tends to focus on how

to become self-sufficient and make decisions regarding

professional and personal growth. Middle adulthood is a

time to consolidate one’s accomplishments, both profes-

sionally and personally, as one becomes more aware of and

concerned for others and society as a whole. As such,

people in the young and middle adulthood stages tend to

have different priorities and goals, with young adults being

more concerned about the well-being of self and family and

middle-aged adults being more concerned about the well-

being of all. Given their different priorities, the young and

middle age groups tend to have different perspectives and

to seek different goals. Cross-cultural life stage research

has shown that ‘‘chronological age has the strongest sal-

ience in communities that are part of modern, industrial-

ized societies’’ (Settersten and Mayer 1997, p. 237).

Nonetheless, the similar life stage patterns in values ori-

entations found in comparative studies of Chinese and

Americans (Pan et al. 1994) and of Thais and Americans

(Ralston et al. 2005) suggest that life stage differences may

transcend both industrialized and industrializing countries.

Generation subculture theory parallels life stage theory

to the extent that it also predicting that age groups will

differ on their values and behaviors. However, in contrast

to life stage theory, generation subculture theory proposes

that a person’s values and behaviors are influenced by the

socio-economic and political context of one’s formative

pre-adult years. Significant macro-level events demarcate

different generation cohorts, each of which share a set of

beliefs and values that comprise a generational identity that

remains relatively intact throughout one’s lifetime (Egri

and Ralston 2004). Inglehart (1997) proposed a structure to

operationalize the values that one might expect from a

Societal-Level Versus Individual-Level Predictions 287

123

specific generation cohort in a particular country, with

intergenerational values differences being premised on two

hypotheses: the socialization hypothesis and the scarcity

hypothesis. The socialization hypothesis proposes that the

values that one acquires and retains throughout one’s life

reflect the socioeconomic conditions experienced during

one’s formative years. The scarcity hypothesis proposes

that one’s values priorities are derived from those envi-

ronmental aspects that had limited availability during one’s

formative years (Inglehart 1997; Meglino and Ravlin

1998). As such, one learns modernist survival values (e.g.,

materialism, conformity) if one grew up during a period of

economic or physical insecurity (e.g., war, economic

depression). Conversely, one learns postmodernist values

(e.g., individualism, trust, self-transcendence) if one grew

up during a period of economic security. Hence, generation

subculture theory predicts substantial within-country vari-

ation in individual values and behaviors due to macro-level

historical events occurring in a society during a particular

period in time.

The study of gender differences has been an important

research topic since its genesis in the 1970s (Eagly et al.

2003; Jaffee and Hyde 2000). Previous research indicates

that cross-national differences in how women’s work

behaviors are perceived are based on culturally influenced

roles, norms, context and stereotypes (e.g., Costa et al.

2001; Eagly and Johannesen-Schmidt 2001; Fullagar et al.

2003). Other research has shown that women tend toward

supportive and considerate workplace behaviors more than

men (cf. Groves 2005), and that regardless of labor market

constraints, female managers continue to hold their moral

perspective and take a more ethical stance (Weeks et al.

1999). Thus, Eagly et al. (2003) concluded that while

female managers are astute organizational players, their

behavioral patterns differ from the typical male patterns.

Bartol et al. (2003) asked: how consistent are gender dif-

ferences across the range of cultures and economic devel-

opment levels found in the business world? A definitive

answer to this question has yet to be found given the mixed

results of cross-national values and ethics studies (e.g.,

Choi and Chen, 2006; Ma 2010; Roxas and Stoneback

2004; Schwartz and Rubel-Lifschitz 2009). Moreover, as

countries change socio-economically, are the values and

manifestations of these values also changing (Steel and

Taras 2010)? While resolving these questions provides an

opportunity for future research endeavors, what does

appear to be resolved today is that differences in gender do

exist and that these differences do influence the behaviors

of the genders when functioning in the work world. In sum,

individual-level differences such as age and gender, which

are inherent within every country, significantly influence

the values and subsequent work behaviors of individuals

across the global workplace.

Evolving Socio-Political Change

Our discussion of inherent within-country differences

might be sufficiently compelling for some to accept the

importance of conducting analyses at the individual-level

rather than the societal-level of analysis. However, it is

clear from a survey of the research designs being used in

the cross-cultural literature that, to date, all are not con-

vinced that analyses are more meaningful at the individual

level of analysis. In support of the prior use of the societal-

level of analysis, it might be acknowledged that one reason

is methodologies did not previously exist, which now do,

that would facilitate individual-level and multi-level anal-

yses (Fischer, 2009). However, even with the previous

discussion of inherent within-country differences and with

the new methodologies available, a skeptic still might

query: What, if anything, has changed over the past several

decades that has made individual-level analyses far more

relevant than societal-level analyses when studying the

values/behaviors of members of today’s global workforce?

Our answer is that what has changed is within-country

workforce demographics. These changes are due to both

within-society changes and globalization of the workforce.

While the primary emphasis in this section of the paper is

on the influences of globalization, we would be remiss not

to note voluntary, as well as legislated, increases in

workforce diversity (e.g., gender, race, ethnicity) across the

globe irrespective of the globalization phenomenon (Za-

noni et al. 2010).

Globalization of the workplace has been argued to be one

of the most significant factors affecting the way we do

business (Earley and Gibson 2002), and over the past few

decades increased globalization has directly contributed to

diversity/heterogeneity within today’s workforces in many

countries around the world (e.g., Tung 2008). Furthermore,

globalization has increasingly accelerated due to the expo-

nential growth of new and improved communication (e.g.,

internet, social media) and transportation (e.g., air travel)

technologies (Amin 2002; Hummels 2007; Janson et al.

2007; Schumann et al. 2012), as well as sweeping changes in

political ideologies (e.g., transitioning economies in Europe

and Asia) across the globe (Ferdinand 2007; Gartin et al.

2009; Miller and Tenev 2007). As a result, we are experi-

encing a variety of somewhat-disassociated phenomena that

are converging to increase within-country diversity/hetero-

geneity. First, enhanced communication technologies such

as Internet access have led to an unprecedented level of

interaction among individuals across political (country)

boundaries. The Internet is fostering its own type of multi-

cultural effect, as its social networking features allow those

in their formative years to experience virtual travel to

interact with other-culture individuals and institutions at

levels never experienced before.

288 D. A. Ralston et al.

123

Second, we are also witnessing an unprecedented

migration of individuals across political boundaries, with

one estimate that there are over 160 million expatriates

worldwide (Malecki and Ewers 2007). Often, this migra-

tion is to seek new economic opportunities (e.g., East

Europeans migrating to Western Europe in pursuit of

work). We also see multinational corporations—compa-

nies, boundaries of which are more far-reaching than the

political boundaries that identify a country—relocating

employees to other countries and cultures in record num-

bers. Colakoglu and Caligiuri (2008) report that there are

over 65,000 MNCs with over 850,000 subsidiaries oper-

ating worldwide, and these numbers are expected to

increase even more due to the continuing growth in glob-

alization (Haslberger and Brewster 2009).

Third, while the trans-border movement of individuals

has been occurring for many centuries, international man-

agement research has increasingly recognized that this

phenomenon creates another group that contributes to the

diversity within a society (Taras et al. 2009). These are the

next-generation bi-cultural and/or multi-cultural individu-

als who have two or more cultural heritages (e.g., mother

and father from different cultures) that result in crossver-

gent individual values orientations reflecting the ‘‘mixed’’

cultural influences experienced during their youths (Tho-

mas et al. 2010). These, in addition to other factors, are

diversifying the ‘‘societal faces’’ of today’s global

workforce.

In summary, ever since societies have existed, there

have been inherent individual-level differences (e.g., gen-

der) within them. Today, in the context of understanding

the values/behaviors of the members of the global work-

force, we now must integrate with these inherent individ-

ual-level differences the impact of the range of phenomena

that are changing the cultural, ethnic and/or religious

make-up of a society’s membership. To picture the impact

of this integration of factors, we might envision a matrix

with the inherent individual-level differences on one axis

and the technology-driven socio-political differences on

the other. The multiplicity of cells in this illustration dictate

a substantial level of heterogeneity, with the global result

being a collection of highly diverse societies that are

trending toward being even more so over the coming years

and decades. In our view, these inherent differences in

conjunction with the increasingly changing face of today’s

global workforce begs re-examination of the question:

Does a societal-level mean score of workplace values truly

represent the values of all workers in a particular society?

This is the essence of our research question. Accordingly,

we hypothesize:

Hypothesis 1 For collectivism values, the individual-

level of analysis has more explanatory power than the

societal-level of analysis in predicting perceptions of eth-

ical behavior.

Hypothesis 2 For individualism values, the individual-

level of analysis has more explanatory power than the societal-

level of analysis in predicting perceptions of ethical behavior.

Methods

Participants

Our sample consists of 16,229 business professionals

across 48 societies that represent a wide diversity of socio-

economic contexts. To collect data, a mail survey of a

cross-section of individuals and industries was conducted

(average response rate was 23 %, range of 15–43 %). In a

few exceptions (e.g., Costa Rica), surveys were conducted

prior to continuing education classes. Of paramount

importance, in all cases, participation was voluntary and

participants were provided assurances of anonymity. The

sample sizes and demographic characteristics of study

participants for each society are presented in Table 1.

Measures

The survey questionnaire was constructed in English and

standard translation/back-translation procedures were used

to develop surveys in the native language of a society. One

exception was India for which an English language ques-

tionnaire was administered since English is the language of

business in this country. To minimize socially desirable

responses, participants were instructed that it was their

perceptions that were important, and that there were no

right or wrong answers (Anastasi 1982).

Independent Variables

We used the SVS (Schwartz 1992) to measure collectivism

and individualism values. In Schwartz’s (1992) typology,

collectivism is comprised of the tradition, conformity, and

benevolence values components, whereas individualism is

comprised of the openness to change (self-direction and

stimulation) and self-enhancement (achievement, hedo-

nism, and power) values components. Individualism was

measured using 18 items and collectivism was measured

using 14 items found to be cross-culturally valid in the SVS

instrument (Schwartz 1994). For each item, respondents

were asked to indicate the importance of a value to them on

a 9-point Likert scale (-1 = opposed to one’s principles,

to 7 = of supreme importance).

The initial confirmatory factor analysis (CFA) for the

total sample (societies counterweighted to be of equal size)

Societal-Level Versus Individual-Level Predictions 289

123

Table 1 Demographic characteristics for the respondents in the 48 societies

N Age Gender Education Position Company size Industry

Mean s.d. (% female) Mean Mean Mean (% manuf.)

Algeria 100 32.7 7.8 18 3.5 2.1 1.7 15

Argentina 96 44.4 9.6 31 n.a. 2.4 2.1 22

Australia 173 29.4 7.4 33 3.8 2.0 2.1 17

Austria 119 33.0 8.6 62 3.7 1.3 2.2 30

Brazil 1,115 39.3 9.5 50 3.9 1.8 2.3 10

Canada 264 39.8 10.8 41 4.3 2.1 2.1 5

China 1,087 31.9 7.6 40 3.7 2.0 2.0 26

Colombia 184 41.0 11.6 47 3.5 3.1 2.3 21

Costa Rica 70 32.6 7.3 42 3.7 2.2 1.9 23

Croatia 285 38.4 9.6 55 3.8 2.1 1.8 18

Czech Rep. 309 39.0 10.9 56 3.9 1.8 1.7 39

Estonia 270 31.6 10.7 71 3.0 1.6 1.9 7

France 662 39.9 10.7 39 3.7 2.7 2.2 21

Germany 414 39.9 11.5 33 3.9 2.4 1.9 22

Greece 170 37.5 8.6 35 n.a. 1.9 2.2 17

Hong Kong 447 34.2 8.9 57 3.0 2.1 1.8 7

Hungary 128 38.3 10.9 42 4.6 2.3 1.6 22

India 285 38.3 12.0 27 4.5 2.8 2.2 33

Indonesia 132 37.1 7.5 25 n.a. 2.1 2.3 30

Israel 135 33.1 6.5 35 4.8 2.0 2.4 16

Italy 297 43.2 10.7 23 4.7 2.4 2.3 25

Japan 135 42.6 5.8 5 4.2 2.7 2.6 51

Lebanon 101 33.6 8.4 42 4.1 2.9 1.9 23

Lithuania 316 43.7 11.4 44 4.3 2.9 1.3 28

Macau 609 35.0 8.2 35 n.a. 2.2 2.1 2

Malaysia 329 34.6 7.3 40 3.8 2.1 3.0 100

Mexico 492 33.6 10.3 44 3.6 2.3 1.8 27

Netherlands 207 37.0 7.0 24 3.4 2.7 2.1 51

New Zealand 113 43.6 12.4 44 4.0 2.6 1.8 12

Pakistan 334 32.5 8.8 13 4.5 2.5 2.2 34

Peru 383 34.2 6.8 35 4.3 2.3 2.1 9

Portugal 823 35.0 11.1 42 4.1 2.2 2.1 14

Russia 338 37.6 8.5 37 5.1 2.5 2.1 44

Singapore 899 35.3 9.6 50 3.8 1.9 2.0 20

Slovakia 82 40.3 8.2 55 n.a. 1.8 2.0 4

Slovenia 300 28.5 7.4 71 3.2 1.3 1.5 31

South Africa 303 40.5 9.0 40 3.8 2.3 2.5 11

South Korea 283 39.5 9.2 20 4.2 2.0 2.4 20

Spain 84 40.2 10.4 16 3.4 2.6 1.3 25

Sri Lanka 120 31.4 6.1 23 4.3 2.6 2.3 35

Switzerland 357 40.9 13.9 23 4.1 2.8 2.0 26

Taiwan 300 41.3 11.0 31 4.0 2.2 2.2 32

Thailand 280 37.1 9.9 58 4.3 2.3 2.0 18

Turkey 124 40.9 9.3 23 4.1 3.2 2.0 52

U.K. 443 40.8 10.1 48 4.1 2.9 2.2 18

U.S. 1,136 38.2 10.7 50 4.7 2.2 2.1 13

Venezuela 134 31.6 6.4 69 4.0 1.6 2.0 23

Vietnam 462 37.9 8.6 35 n.a. 2.2 1.9 16

Coding Education (highest level completed): 1 4 or fewer years, 2 5–8 years, 3 9–12 years, 4 13–16 years, 5 masters degree, 6 doctorate degree; position: 1

professional, 2 1st level management, 3 middle management, 4 top management; company size: 1 less than 100 employees, 2 100–1000 employees, 3 more than

1000 employees

290 D. A. Ralston et al.

123

showed an acceptable fit for the model with two higher-

order factors (collectivism and individualism) and 8 first-

order factors [v2ð455Þ = 24511.42, CFI = .940, NNFI =

.935, RMSEA = .062]. To address cross-cultural differ-

ences in scale response style with the SVS instrument

(Fischer 2004), we used within-subject standardized

adjusted scores in the analyses (per Hanges 2004). For the

48 societies, the adjusted means, standard deviations, and

scale reliabilities (Cronbach’s a) for the two values mea- sures are presented in Table 2. Across societies, the aver-

age of scale reliabilities (Cronbach’s a) was .83 for collectivism and .82 for individualism, with all society

scale reliabilities above .70.

Dependent Variables

We used the SIE instrument (Ralston and Pearson 2010) to

assess participants’ perceptions of the ethicality of influ-

ence behaviors. Participants were asked to indicate how

acceptable (ethical) their co-workers would consider 38

scenario items to be, using an 8-point Likert-type scale

(1 = extremely unacceptable to 8 = extremely accept-

able). The four SIE dimensions are pro-organizational

behaviors (6 items), image management (5 items), self-

serving behaviors (6 items), and maliciously intended

behaviors (5 items). The description of the four SIE

dimensions is presented in Table 3. Further, a previous

41-country study of the SIE dimensions identified a uni-

versally consistent hierarchy for these dimensions (Ralston

et al. 2009). This hierarchy of highly ethical to highly

unethical behavior is, respectively: pro-organizational,

image management, self-serving, and maliciously intended

behavior. Thus, the SIE covers the full spectrum of ethi-

cality from the highly ethical (pro-organizational) to the

highly unethical (maliciously intended).

The initial confirmatory factor analysis (CFA) for the 48

societies (samples counterweighted to be of equal size)

showed a good fit for the 4-factor 22-item model [v2ð203Þ =

6440.01, CFI = .971, NNFI = .968, RMSEA =.048]. The

one-factor 22-item model had an unacceptable fit

[v2ð209Þ = 81008.27, CFI = .813, NNFI = .794, RMSEA =

.168]. Multi-group CFAs were conducted to test for

between-group measurement invariance (cf. Steenkamp and

Baumgartner 1998). Per Cheung and Rensvold (2002),

model fit comparisons were based on changes in CFI with

DCFI B .010 indicating no significant difference, DCFI between .010 and .020 indicating a marginal difference, and

DCFI [ .020 indicating a significant difference in model fits. The baseline (unconstrained) CFA model had a mar-

ginal level of between-group configural invariance

[v2ð9744Þ = 30053.47, CFI = .887, NNFI = .872, RMSEA =

Table 2 Collectivism and individualism: society adjusted means, standard deviations and scale reliabilities (Cronbach a)

Collectivism Individualism

Mean s.d. a Mean s.d. a

Algeria 4.51 0.48 .73 3.60 0.45 .77

Argentina 4.38 0.61 .89 3.90 0.43 .88

Australia 4.02 0.69 .85 4.34 0.50 .82

Austria 3.98 0.59 .77 3.95 0.46 .84

Brazil 4.33 0.55 .78 3.92 0.44 .73

Canada 4.15 0.61 .86 4.16 0.53 .86

China 4.01 0.57 .86 4.11 0.44 .79

Colombia 4.44 0.54 .85 4.02 0.45 .77

Costa Rica 4.22 0.63 .86 4.13 0.49 .86

Croatia 3.97 0.68 .79 3.96 0.53 .76

Czech Rep. 3.97 0.62 .79 3.99 0.57 .85

Estonia 3.89 0.63 .79 4.07 0.48 .85

France 4.05 0.59 .81 4.07 0.50 .79

Germany 4.04 0.62 .82 4.17 0.51 .82

Greece 4.48 0.53 .85 3.77 0.54 .77

Hong Kong 4.22 0.58 .86 3.95 0.48 .82

Hungary 3.93 0.66 .79 3.95 0.55 .85

India 4.45 0.54 .84 3.95 0.44 .75

Indonesia 4.52 0.50 .87 3.99 0.39 .90

Israel 4.13 0.59 .79 4.22 0.46 .88

Italy 4.38 0.65 .82 3.76 0.51 .83

Japan 3.89 0.67 .80 3.83 0.50 .83

Lebanon 4.18 0.67 .86 4.09 0.52 .79

Lithuania 3.97 0.54 .82 4.02 0.50 .83

Macau 4.17 0.61 .88 4.05 0.46 .84

Malaysia 4.49 0.52 .82 3.93 0.33 .79

Mexico 4.42 0.52 .84 4.00 0.47 .77

Netherlands 4.00 0.59 .83 4.34 0.46 .87

New Zealand 4.02 0.65 .82 4.15 0.51 .81

Pakistan 4.37 0.56 .84 4.17 0.45 .84

Peru 4.35 0.58 .86 4.06 0.46 .80

Portugal 4.18 0.61 .82 3.99 0.52 .78

Russia 3.93 0.55 .80 4.08 0.55 .80

Singapore 4.43 0.64 .87 3.97 0.55 .84

Slovakia 4.15 0.55 .74 3.64 0.46 .72

Slovenia 3.80 0.54 .80 4.13 0.45 .83

South Africa 4.32 0.69 .88 3.94 0.51 .88

South Korea 4.10 0.59 .79 4.14 0.48 .82

Spain 4.48 0.49 .76 3.92 0.45 .85

Sri Lanka 4.07 0.58 .89 3.80 0.42 .89

Switzerland 4.00 0.63 .78 4.13 0.48 .83

Taiwan 4.22 0.56 .90 4.02 0.47 .90

Thailand 4.58 0.57 .76 3.81 0.51 .75

Turkey 4.25 0.62 .86 3.95 0.54 .86

UK 3.90 0.67 .85 4.21 0.50 .78

US 4.30 0.62 .83 4.09 0.48 .79

Venezuela 4.57 0.54 .85 4.00 0.41 .88

Vietnam 4.43 0.54 .86 3.85 0.47 .80

Societal-Level Versus Individual-Level Predictions 291

123

.090]. The metric invariance model (factor loadings con-

strained) had a marginal difference in model fit (DCFI = -.014) whereas the partial metric invariance model with

four factor loadings unconstrained (one for each SIE factor)

was not significantly different from the baseline model

(DCFI = -.009). The partial scalar invariance model (intercepts unconstrained for the four items) had a signifi-

cant change in model fit (DCFI = -.050), and freeing additional intercepts did not yield a nonsignificant change.

Hence, within-subject standardized adjusted scores were

used in analyses (per Hanges 2004).

Table 4 presents the societies’ adjusted scores, standard

deviations, and scale reliabilities for the SIE variables. For

the 48 societies, the average of scale reliabilities (Cron-

bach’s a) was .70 for pro-organizational, .73 for image management, .81 for self-serving, and .72 for maliciously

intended. The number of societies with scale reliabilities

below the .60 cutoff level used in previous cross-cultural

research (e.g., Fu and Yukl 2000; Parboteeah et al. 2009)

was six for pro-organizational, one for self-serving, and

two for maliciously intended. Parallel analyses to test

hypotheses without these societies showed no substantive

differences in results. Therefore, we report the results for

all 48 societies.

Common Method Variance

We took a number of preventive measures to address com-

mon method variance issues. First, we provided assurances of

anonymity and confidentiality of responses to participants,

and used different response formats for measures previously

shown to be reliable and valid (Podsakoff et al. 2003). To

assess this potential biasing effect, we used the total coun-

terweight sample to conduct CFAs for the eight first-order

SVS values factors and the four SIE factors. The fit of the

baseline model was: v2(1311) = 33410.48, CFI = .951, NNFI = .947, RMSEA = .042. The CFA model with an

additional unmeasured latent method common factor showed

a nonsignificant change in model fit (DCFI = .009), while the CFA model for the Harman one-factor test had a significantly

poorer fit (DCFI = -.092). In sum, these analyses indicate that common method variance was not a significant issue.

Analyses

We used hierarchical linear modeling (Raudenbush and

Bryk 2002) to assess the effectiveness of the individual-

level versus the societal-level of the collectivism and

individualism values dimensions in predicting the per-

ceived ethicality of subordinate influence behaviors. In

these analyses, the dependent variables were the four SIE

variables. The intraclass correlation coefficients (ICC) for

the null models indicated sufficient between-group vari-

ance to proceed with HLM analyses (31.6 % for pro-

organizational, 10.9 % for image management, 14.2 % for

self-serving, and 13.2 % for maliciously intended, all v2

significant at p \ .001 level). The independent variables were collectivism and individualism values scores at both

the individual-level and the societal-level. For the aggre-

gated societal-level collectivism and individualism values

scores, we estimated Brown and Hauenstein’s (2005)

awg(1) interrater agreement statistics. The high level of

Table 3 Subordinate Influence Ethics dimensions

Pro-organizational ethics behavior

These behaviors may be defined as the ‘‘organizational person’’ approach to gain influence in that these behaviors reflect those that are

typically prescribed and/or sanctioned by organizations for their subordinates. These may be viewed as behaviors that are most ethical and

that tend to be directly beneficial to the organization. Pro-organizational behaviors include acts such as getting the job done, behaving in an

appropriate manner, developing good working relationships, and working overtime

Image management behavior

These behaviors may be defined as the ‘‘get others to like me’’ approach to gain influence in that they are intended to be non-confrontational

in nature (e.g., ingratiatory). As such, they have a ‘soft’ self-orientation. Image management behaviors include acts such as volunteering for

undesirable tasks to make themselves appreciated by the superior, and attempting to act in a manner that they believe will result in others

admiring them

Self-serving ethics behavior

These behaviors may be defined as the ‘‘it’s me first’’ approach to gain influence in that these behaviors show self-interest being of paramount

importance, and thus being above the interests of others and the organization. Thus, they have a ‘‘hard’’ self-orientation. Whether these

behaviors help or harm the organization is secondary to the individual meeting his/her goals and thus are likely to be determined by the

situation. Self-serving behaviors include acts such as blaming others for mistakes, spreading rumors, and taking credit for others’ work

Maliciously intended ethics behavior

These behaviors may be defined as the ‘‘burn, pillage, and plunder’’ approach to gain influence in that they are intended to directly hurt others

and/or the organization, to facilitate personal gain. These may be viewed as behaviors that are most unethical, and in many industrialized

societies these behaviors would also be considered illegal. Maliciously intended behaviors include acts such as making threatening phone

calls to co-workers, blackmail, and stealing corporate documents

292 D. A. Ralston et al.

123

Table 4 Perceptions of SIE: society adjusted means, standard deviations and scale reliabilities

Pro-organizational Image management Self-serving Maliciously intended

Mean s.d. a Mean s.d. a Mean s.d. a Mean s.d. a

Algeria 4.41 0.81 .77 4.99 1.52 .73 3.39 1.01 .65 2.22 0.53 .80

Argentina 6.44 0.63 .74 4.30 1.06 .73 2.16 0.96 .85 1.70 0.35 .73

Australia 6.11 0.80 .80 5.12 1.07 .70 2.71 1.15 .86 1.50 0.66 .83

Austria 6.33 0.56 .74 4.68 1.07 .75 2.44 1.00 .84 1.58 0.46 .79

Brazil 6.53 0.57 .66 4.15 1.36 .76 2.35 0.82 .80 1.70 0.40 .75

Canada 6.44 0.52 .76 5.08 0.97 .75 2.22 1.02 .91 1.52 0.32 .75

China 6.20 0.57 .76 4.88 1.11 .76 2.60 0.91 .82 1.65 0.48 .75

Colombia 6.45 0.53 .68 4.44 1.21 .80 1.99 0.68 .77 1.77 0.36 .69

Costa Rica 6.23 0.62 .74 4.45 1.05 .62 2.15 0.99 .86 1.86 0.41 .70

Croatia 5.91 0.83 .66 4.37 1.13 .70 3.18 1.38 .87 1.80 0.63 .77

Czech Rep. 6.36 0.75 .66 4.66 1.09 .70 2.55 1.19 .90 1.64 0.39 .70

Estonia 6.29 0.68 .78 4.40 1.20 .73 2.59 0.96 .80 1.72 0.48 .79

France 6.51 0.49 .66 4.18 1.33 .81 2.41 0.89 .75 1.77 0.43 .71

Germany 6.20 0.67 .73 4.99 1.18 .82 2.56 1.09 .86 1.56 0.44 .79

Greece 6.44 0.55 .58 3.87 1.24 .72 2.18 0.76 .74 1.78 0.37 .69

Hong Kong 6.23 0.60 .82 5.06 1.04 .73 2.73 1.12 .88 1.46 0.41 .77

Hungary 6.17 0.68 .70 5.11 1.03 .69 2.90 1.17 .87 1.49 0.40 .69

India 6.14 0.88 .78 4.20 1.29 .80 2.79 1.33 .92 1.74 0.66 .63

Indonesia 6.16 0.53 .64 5.07 1.17 .68 2.53 0.82 .66 1.73 0.47 .81

Israel 6.24 0.55 .75 5.23 0.92 .65 2.18 1.04 .86 1.49 0.40 .83

Italy 6.39 0.60 .69 4.97 0.99 .77 2.83 1.16 .88 1.52 0.38 .60

Japan 6.05 0.36 .61 5.16 0.87 .72 1.80 0.45 .71 1.36 0.27 .64

Lebanon 6.32 0.66 .73 4.45 1.13 .76 2.62 1.00 .87 1.67 0.51 .73

Lithuania 6.19 0.52 .53 4.44 1.01 .64 2.41 1.02 .82 1.71 0.37 .52

Macau 6.20 0.67 .73 4.94 1.08 .72 2.65 1.04 .87 1.55 0.53 .88

Malaysia 6.15 0.62 .72 5.00 1.03 .71 2.69 1.01 .80 1.64 0.56 .87

Mexico 6.35 0.69 .70 4.50 1.10 .66 2.10 0.74 .81 1.87 0.47 .71

Netherlands 6.45 0.33 .60 5.15 0.81 .66 1.76 0.39 .77 1.58 0.26 .60

New Zealand 6.61 0.51 .76 4.99 1.05 .81 2.22 0.90 .86 1.53 0.28 .65

Pakistan 5.68 0.92 .70 4.46 1.35 .71 3.34 1.46 .85 1.96 0.79 .80

Peru 6.38 0.51 .64 4.42 1.05 .73 2.08 0.76 .82 1.70 0.39 .75

Portugal 6.40 0.52 .64 4.74 1.08 .75 2.27 0.89 .86 1.58 0.36 .67

Russia 6.16 0.68 .71 4.63 1.21 .74 2.90 0.89 .63 1.80 0.51 .71

Singapore 6.31 0.58 .76 4.82 1.12 .82 2.34 0.96 .89 1.57 0.49 .86

Slovakia 6.34 0.61 .81 4.18 1.12 .78 2.68 0.92 .57 1.68 0.37 .62

Slovenia 6.15 0.72 .51 4.71 1.11 .70 2.84 1.11 .79 1.66 0.47 .75

South Africa 6.17 0.82 .69 4.84 1.13 .78 2.98 1.34 .90 1.60 0.49 .85

South Korea 6.37 0.54 .79 5.18 1.02 .74 1.98 0.68 .75 1.76 0.38 .81

Spain 6.41 0.60 .79 4.43 1.18 .75 2.20 0.76 .81 1.84 0.53 .72

Sri Lanka 5.85 0.56 .76 4.15 1.01 .75 2.21 0.99 .85 1.45 0.50 .82

Switzerland 6.44 0.43 .59 4.82 1.07 .77 1.97 0.66 .77 1.60 0.30 .61

Taiwan 6.21 0.52 .68 5.61 1.12 .73 2.44 0.93 .83 1.61 0.51 .78

Thailand 6.45 0.44 .58 5.24 0.88 .65 2.24 0.73 .75 1.52 0.35 .65

Turkey 6.61 0.44 .54 4.50 1.20 .69 1.97 0.63 .72 1.68 0.33 .65

UK 6.44 0.52 .62 5.08 0.93 .76 2.28 1.00 .86 1.41 0.33 .66

US 6.49 0.48 .66 5.21 0.97 .72 1.90 0.73 .78 1.55 0.29 .62

Venezuela 6.41 0.56 .75 4.23 1.20 .72 1.98 0.61 .74 1.87 0.38 .52

Vietnam 6.31 0.53 .66 4.95 1.05 .64 2.45 0.77 .67 1.71 0.39 .63

Societal-Level Versus Individual-Level Predictions 293

123

interrater agreement for both the collectivism (awg(1) mean = .87, range of .79–.93) and individualism (awg(1) mean = .88, range of .82–.94) measures across the 48

societies (LeBreton and Senter 2008) supported using

aggregated societal-level measures.

We estimated a series of intercepts-as-outcomes HLM

models for each SIE dependent variable. Model 1 was the

baseline comparison model with the three covariates.

Individual-level covariates were participant age and gender

while the societal-level covariate was the logarithm of

GDP per capita (purchasing power parity). Models 2 and 3

added the collectivism and individualism values at the

individual-level separately. Models 4 and 5 entered col-

lectivism and individualism (aggregated scores) at the

societal-level separately. Model 6 had collectivism at both

the individual- and societal-levels, whereas model 7 had

individualism at both levels. Model 8 was the full model

with the two values scores at both the individual and

societal levels. Since our interest was whether a values

predictor had a differential impact at both levels, the

individual-level and societal-level variables were grand-

mean-centered (Enders and Tofighi 2007; Hofmann and

Gavin 1998). Given different society sample sizes, we

counterweighted the society samples to be of equal size in

the analyses.

Our tests of Hypotheses 1 and 2 were based on com-

parisons of model deviance index statistics and on exami-

nation of the parameter estimates for the level-1 and level-

2 values variables. We compared the deviance index

(-2 9 log likelihood of a maximum-likelihood estimate)

of models 2–8 with the baseline covariate model. Using the

full maximum-likelihood estimation, the difference in

deviance statistics has a Chi square distribution with

degrees of freedom being the difference in the number of

estimated parameters in comparison models (Raudenbush

and Bryk 2002). As identified by Kreft (2000), deviance

tests account for the multilevel nature of errors in HLM

models and hence are an appropriate way to represent

effect size for multi-level models. We also calculated

pseudo-R 2

statistics (Snijders and Bosker 1994) to indicate

proportional reduction of explained variance. In respect to

the values variables, a significant level-2 parameter esti-

mate for models 6 and 7 indicates a contextual (societal)

effect that significantly differs from that at the individual

level (Enders and Tofighi 2007).

Alternative Societal-Level Values Scores

To more fully examine the influence of societal-level val-

ues on ethical behaviors, we conducted additional HLM

analyses using publicly available societal-level collectiv-

ism and individualism related values scores for the socie-

ties in this study. These included: Hofstede’s (2001) VSM

individualism/collectivism value scores (N = 46 societies);

Taras et al.’s (2012) meta-analytic VSM individualism/

collectivism value scores derived from studies conducted

during the 2000s (N = 29) 1 ; the GLOBE project’s (House

et al. 2004) in-group collectivism and institutional collec-

tivism values (N = 36); and Schwartz’s (Licht et al. 2007)

societal embeddedness and autonomy (affective and intel-

lectual) values (N = 30) which have been identified as

conceptually similar to collectivism and individualism,

respectively (House et al. 2004; Schwartz 1999).

Intercepts-as-outcomes HLM analyses were conducted

for each of the SIE dependent variables. The first model

consisted of the three covariates, the second model added

the two individual-level collectivism and individualism

values, and the third model added each societal-level val-

ues variable separately. All ICC statistics for the reduced

society samples were significant (v2 at the p \ .001 level), and covariate and predictor variables were grandmean-

centered with society samples counterweighted to be of

equal size. Change in deviance statistics for successive

nested models are reported.

Results

As illustrated in Fig. 1a, the societal-level collectivism

values had a range of 0.78 (3.80–4.58), while individualism

had a range of 0.74 (3.60–4.34). As illustrated in Fig. 1b,

the individual-level collectivism values had a range of 5.47

(0.89–6.36), while individualism had a range of 3.64

(2.40–6.04). Figure 2 identifies the locations of the 48

societies presented in Fig. 1a.

Table 5 presents the descriptive statistics (means, stan-

dard deviations, and correlations) at both the individual and

societal levels of analyses. Hypothesis 1 and Hypothesis 2

proposed that for collectivism and individualism values

(respectively), the individual-level of analysis would be

more predictive of perceptions of ethical behavior than the

societal-level of analysis. The HLM results presented in

Table 6 provide strong support for both hypotheses.

1 Taras et al. (2012) provide only regional scores for countries

located in a number of geographic regions (e.g., Arab countries,

Baltic USSR, Central America, and South America). Rather than

assigning geographic region scores, which assumes a lack of inter-

societal cultural variation within a geographic region, we only used

values scores identified for individual countries. Taras et al. (2012)

also provide meta-analytic VSM individualism/collectivism scores

based on the full set of studies conducted 1970–2010. The results of

HLM analyses (N = 29 societies) using these scores were the same

(nonsignificant for all SIE dependent variables) as for the more recent

set of scores based on studies conducted during the 2000s which we

used to provide a more updated assessment of the VSM individual-

ism/collectivism value.

294 D. A. Ralston et al.

123

In respect to the individual-level of analysis, collectiv-

ism values (model 2) were positively related to pro-orga-

nizational and maliciously intended behavior (at p \ .001 level), negatively related to image management (p \ .001), and not significantly related to self-serving behaviors.

Individual-level individualism values (model 3) were pos-

itively related to image management and self-serving

behaviors (p \ .001), and negatively related to pro-orga- nizational (p \ .001) and maliciously intended (p \ .05) behaviors. The addition of these individual-level values

variables in models 2 and 3 resulted in a significant change

in explained variance for each type of ethical behavior

(D deviance at p \ .001 level). In respect to the societal-level of analysis, collectivism

values (model 4) were positively related to maliciously

intended behaviors (p \ .05), and not significantly related to pro-organizational, image management, and self-serving

behaviors. Societal-level individualism (model 5) was not

significantly related to any of the four types of influence

behaviors. The addition of the societal-level values vari-

ables in the HLM models resulted in a significant change in

explained variance for only collectivism in respect to

maliciously intended behaviors (D deviance at p \ .05 level).

The HLM models 6 (collectivism) and 7 (individualism)

included both the individual- and societal-level values. In

these models, the HLM analyses for pro-organizational,

image management, and self-serving behaviors showed

similar results to those of the models in which these vari-

ables were entered separately. For maliciously intended

behaviors, societal-level collectivism (model 6) and indi-

vidual-level individualism (model 7) were no longer sig-

nificant predictors.

And finally, the results for models 2 through 7 were very

similar to those for the full models (model 8) for pro-

organizational and image management behaviors. The full

Fig. 1 Comparison of the societal-level and individual-level matrices for collectivism by individualism: a matrix of the 48 societies in the study, b matrix of the 16,229 participants in the study

Fig. 2 Locations of the 48 societies on the matrix of collectivism by individualism

Societal-Level Versus Individual-Level Predictions 295

123

model for self-serving behaviors showed that both indi-

vidual-level values were positively related (p \ .001) and both societal-level values were negatively related

(p \ .05). For maliciously intended behaviors, both indi- vidual-level values were positively related (p \ .01) whereas both societal-level values were not significant

predictors.

Alternative Societal-Level Values

The results of the HLM analyses using alternative collec-

tivism and individualism societal-level scores are provided

in Table 7. The VSM individualism value was negatively

related to maliciously intended behaviors (p \ .05), and not significantly related to other types of ethical behaviors.

There were no significant relationships for meta-analytic

VSM individualism. For the GLOBE cultural values, in-

group collectivism and institutional collectivism were

positively related to pro-organizational behaviors (respec-

tively, p \ .05, p \ .01). In addition, institutional collec- tivism was negatively related to image management

(p \ .001) and positively related to maliciously intended

(p \ .01) behaviors. For the three Schwartz societal values, the only significant relationship was the positive relation-

ship between intellectual autonomy and pro-organizational

behaviors (p \ .001). Results consistent with those for the aggregated values

scores would be nonsignificant relationships for the pro-

organizational, image management, and self-serving

behaviors. As shown in Table 7, this was found for the

VSM and the meta-analytic VSM individualism/collecti-

vism measures, as well as for the embeddedness and

affective autonomy measures. In addition, in-group col-

lectivism and intellectual autonomy were not significantly

related to image management and self-serving behaviors,

and institutional collectivism was not significantly related

to self-serving behaviors. However, significant contextual

effects were found in that in-group collectivism,

institutional collectivism and intellectual autonomy (an

individualistic value) were positively related to pro-orga-

nizational behaviors, and institutional collectivism was

negatively related to image management behaviors.

For maliciously intended behaviors, we found the

aggregated collectivism value was positively related and

Table 5 Descriptive statistics: means, standard deviations, and correlations

Mean s.d. 1 2 3 4 5 6 7 8 9 10 11 12 13

Individual-level

1. Collectivism 4.20 .63

2. Individualism 4.01 .50 -.62

3. Pro-organizational 6.25 .69 .03 -.05

4. Image management 4.71 1.18 -.10 .12 -.18

5. Self-serving 2.42 1.03 .01 .01 -.56 -.04

6. Maliciously intended 1.67 .47 .12 -.05 -.24 -.44 .01

7. Age 37.13 10.16 .13 -.23 .11 -.11 -.03 .02

8. Gender .39 .48 -.01 -.11 .04 -.03 .02 -.04 -.15

Societal-level

1. Collectivism 4.20 .21

2. Individualism 4.01 .15 -.43

3. Pro-organizational 6.24 .33 -.06 .29

4. Image management 4.73 .43 -.17 .27 -.09

5. Self-serving 2.46 .39 -.03 -.23 2.65 -.02

6. Maliciously intended 1.70 .19 .39 -.19 2.44 2.50 .35

7. GDP pc ppp (log) 9.47 .82 2.43 .28 .30 .26 -.33 -.45

8. VSM individualism 45.02 23.69 2.38 .22 .14 .09 -.05 2.33 .49

9. Meta-analytic VSM individualism .24 .55 2.46 .62 .14 .02 -.18 -.05 .55 .69

10. In-group collectivism (GLOBE) 5.63 0.32 .06 .13 .26 -.05 -.15 .13 .11 .12 .17

11. Institutional collectivism (GLOBE) 4.76 0.46 .51 2.40 .39 2.58 -.08 .46 -.21 -.20 -.27 .07

12. Embeddedness (Schwartz) 2.28 0.52 .11 -.00 -.19 .06 -.01 -.01 2.55 2.46 2.42 2.52 -.22

13. Affective autonomy (Schwartz) 4.46 0.41 2.40 .15 .21 -.11 -.21 .00 .46 .45 .53 .14 .20 2.55

14. Intellectual autonomy (Schwartz) 4.80 0.32 -.01 .16 .40 -.01 -.25 -.09 .52 .47 .50 .53 .31 2.63 .59

Total N = 48 societies (16,229 respondents); N = 46 VSM value, N = 29 meta-analytic VSM, N = 36 GLOBE values, N = 30 Schwartz values. For individual-

level correlations, society samples are counterweighted to be of equal size. Gender coded as 1 = female and 0 = male. Individual-level correlations r [ |.03| significant at p \ .01 level; societal-level correlations in bold font significant at p \ .05 level

296 D. A. Ralston et al.

123

T a

b le

6 H

L M

c o

m p

a ri

so n

s o

f in

d iv

id u

a l-

le v

e l

a n

d so

c ie

ta l-

le v

e l

v a lu

e s

a s

p re

d ic

to rs

o f

p e rc

e iv

e d

e th

ic a l

b e h

a v

io rs

(N =

4 8

so c ie

ti e s)

M o

d e l

1 M

o d

e l

2 M

o d

e l

3 M

o d

e l

4 M

o d

e l

5 M

o d

e l

6 M

o d

e l

7 M

o d

e l

8

P ro

-o rg

a n

iz a

ti o

n a

l

In te

rc e p

t 6

.2 1

5 *

* *

6 .2

5 6

* *

* 6

.2 0

7 *

* *

6 .2

1 4

* *

* 6

.2 3

3 *

* *

6 .2

1 6

* *

* 6

.2 3

6 *

* *

6 .2

3 5

* *

*

In d

iv id

u a l-

le v

e l

A g

e .0

0 3

� .0

0 2

.0 0

1 .0

0 3

� .0

0 3 �

.0 0

2 .0

0 1

.0 0

1

G e n

d e r

.0 0

6 .0

0 4

- .0

1 8

.0 0

6 .0

0 9

.0 0

4 -

.0 1

5 -

.0 1

6

C o

ll e c ti

v is

m .0

5 5

* *

* .0

5 5

* *

* -

.0 2

4

In d

iv id

u a li

sm -

.1 4

2 *

* *

- .1

4 3

* *

* -

.1 6

1 *

* *

S o

c ie

ta l

le v

e l

G D

P p

c p

p p

(l o

g )

.1 1

2 .1

2 6

.1 2

6 .1

2 1

.0 8

5 .1

1 8

.0 8

0 .1

0 3

C o

ll e c ti

v is

m .0

6 4

- .0

6 1

.1 1

1

In d

iv id

u a li

sm .5

1 0

.5 5

0 .6

0 5

D e v

ia n

c e

2 7

,9 8

4 .6

5 2

7 ,9

0 4

.0 3

2 7

,7 7

1 .8

6 2

7 ,9

8 4

.6 0

2 7

,9 8

2 .0

1 2

7 ,9

0 3

.9 8

2 7

,7 6

0 .7

7 2

7 ,7

4 9

.9 6

N u

m b

e r

o f

p a ra

m e te

rs 1

1 1

6 1

6 1

2 1

2 1

7 1

7 2

4

D d

e v

ia n

c e

v e rs

u s

m o

d e l

1 8

0 .6

2 *

* *

2 1

2 .7

9 *

* *

0 .0

5 2

.6 4

8 0

.6 7

* *

* 2

2 3

.8 8

* *

* 2

3 4

.6 9

* *

*

P se

u d

o -R

2 .0

1 3

.0 3

8 .0

2 3

.0 1

3 .0

1 9

.0 3

8 .0

3 1

.0 6

7

Im a

g e

m a

n a

g e m

e n

t

In te

rc e p

t 4

.6 6

7 *

* *

4 .6

6 9

* *

* 4

.6 7

5 *

* *

4 .6

6 6

* *

* 4

.6 8

1 *

* *

4 .6

6 8

* *

* 4

.6 8

7 *

* *

4 .6

8 3

* *

*

In d

iv id

u a l-

le v

e l

A g

e -

.0 1

9 *

* *

- .0

1 8

* *

* -

.0 1

6 *

* *

- .0

1 9

* *

* -

.0 1

9 *

* *

- .0

1 8

* *

* -

.0 1

6 *

* *

- .0

1 7

* *

*

G e n

d e r

- .0

6 1

* -

.0 5

4 *

- .0

2 4

- .0

6 1

* -

.0 6

0 *

- .0

5 5

* -

.0 2

4 -

.0 3

6

C o

ll e c ti

v is

m -

.1 2

5 *

* *

- .1

2 6

* *

* -

.0 7

2 *

*

In d

iv id

u a li

sm .1

6 3

* *

* .1

5 9

* *

* .1

0 1

*

S o

c ie

ta l

le v

e l

G D

P p

c p

p p

(l o

g )

.1 7

4 *

.1 7

4 *

* .1

8 3

* .1

8 0

* .1

3 0 �

.2 0

9 *

.1 2

5 .1

5 7

*

C o

ll e c ti

v is

m .0

4 5

.2 3

6 .3

5 3

In d

iv id

u a li

sm .5

9 7

.4 4

3 .6

1 7

D e v

ia n

c e

4 5

,6 7

4 .4

2 4

5 ,5

5 2

.7 3

4 5

,5 6

7 .9

0 4

5 ,6

7 4

.3 9

4 5

,6 7

0 .6

2 4

5 ,5

5 1

.9 9

4 5

,5 6

5 .8

0 4

5 ,3

5 5

.7 8

N u

m b

e r

o f

p a ra

m e te

rs 1

1 1

6 1

6 1

2 1

2 1

7 1

7 2

4

D d

e v

ia n

c e

v e rs

u s

m o

d e l

1 1

2 1

.6 9

* *

* 1

0 6

.5 2

* *

* 0

.0 3

3 .8

0 �

1 2

2 .4

3 *

* *

1 0

8 .6

2 *

* *

3 1

8 .6

4 *

* *

P se

u d

o -R

2 .0

2 9

.0 4

0 .0

3 9

.0 2

9 .0

3 5

.0 3

5 .0

4 0

.0 3

5

S e lf

-s e rv

in g

In te

rc e p

t 2

.4 1

5 *

* *

2 .4

3 4

* *

* 2

.4 2

1 *

* *

2 .4

1 8

* *

* 2

.4 0

7 *

* *

2 .4

1 8

* *

* 2

.4 0

9 *

* *

2 .4

1 3

* *

*

In d

iv id

u a l-

le v

e l

A g

e .0

0 4

� .0

0 3 �

.0 0

5 �

.0 0

4 �

.0 0

4 �

.0 0

3 �

.0 0

5 *

.0 0

5

G e n

d e r

.0 3

8 .0

3 9

.0 5

6 *

.0 3

8 .0

3 8

.0 3

9 .0

5 6

� .0

6 9

*

C o

ll e c ti

v is

m .0

1 7

.0 1

8 .1

1 3

* *

*

Societal-Level Versus Individual-Level Predictions 297

123

T a

b le

6 c o

n ti

n u

e d

M o

d e l

1 M

o d

e l

2 M

o d

e l

3 M

o d

e l

4 M

o d

e l

5 M

o d

e l

6 M

o d

e l

7 M

o d

e l

8

In d

iv id

u a li

sm .1

0 4

* *

* .1

0 4

* *

* .1

9 3

* *

*

S o

c ie

ta l

le v

e l

G D

P p

c p

p p

(l o

g )

- .1

5 4

* -

.1 5

8 *

- .1

7 3

* -

.2 0

8 *

- .1

3 6

* -

.2 2

1 *

* -

.1 4

8 *

- .4

5 0

* *

C o

ll e c ti

v is

m -

.3 8

4 -

.4 3

0 -

.5 4

2 *

In d

iv id

u a li

sm -

.3 2

0 -

.4 5

0 -

.2 0

4 *

D e v

ia n

c e

4 1

,4 6

7 .3

5 4

1 ,4

2 8

.1 4

4 1

,3 8

5 .7

2 4

1 ,4

6 5

.3 0

4 1

,4 6

6 .5

1 4

1 ,4

2 5

.5 7

4 1

,3 8

4 .0

1 4

1 ,3

2 2

.2 1

N u

m b

e r

o f

p a ra

m e te

rs 1

1 1

6 1

6 1

2 1

2 1

7 1

7 2

4

D d

e v

ia n

c e

v e rs

u s

m o

d e l

1 3

9 .2

1 *

* *

8 1

.6 3

* *

* 2

.0 5

0 .8

4 4

1 .7

8 *

* *

8 3

.3 4

* *

* 1

4 5

.1 4

* *

*

P se

u d

o -R

2 .0

3 2

.0 3

3 .0

4 1

.0 3

9 .0

3 8

.0 4

0 .0

4 2

.0 5

6

M a

li c io

u sl

y in

te n

d e d

In te

rc e p

t 1

.6 8

1 *

* *

1 .6

8 1

* *

* 1

.6 8

1 *

* *

1 .6

7 8

* *

* 1

.6 7

8 *

* *

1 .6

7 8

* *

* 1

.6 7

8 *

* *

1 .6

7 7

* *

*

In d

iv id

u a l-

le v

e l

A g

e .0

0 3

* *

.0 0

3 *

* .0

0 3

* *

.0 0

3 *

* .0

0 3

* *

.0 0

3 *

* .0

0 3

* *

.0 0

3 *

*

G e n

d e r

- .0

3 7

* -

.0 3

8 *

- .0

3 9

* -

.0 3

7 *

- .0

3 7

* -

.0 3

8 *

- .0

3 9

* -

.0 3

2 �

C o

ll e c ti

v is

m .0

4 9

* *

* .0

4 8

* *

* .0

6 9

* *

*

In d

iv id

u a li

sm -

.0 2

5 *

- .0

1 5

.0 4

1 *

*

S o

c ie

ta l

le v

e l

G D

P p

c p

p p

(l o

g )

- .1

1 0

* *

- .1

0 9

* *

* -

.1 1

2 *

* *

- .0

6 8

* *

- .0

9 8

* *

- .0

7 6

* -

.1 0

0 *

* -

.0 7

2 *

C o

ll e c ti

v is

m .2

8 6

* .2

1 7

.1 8

4

In d

iv id

u a li

sm -

.1 5

1 -

.1 8

6 -

.1 6

1

D e v

ia n

c e

1 8

,5 2

5 .4

4 1

8 ,3

9 5

.9 1

1 8

,4 8

8 .2

6 1

8 ,5

2 0

.1 3

1 8

,5 2

4 .1

6 1

8 ,3

9 3

.0 8

1 8

,4 8

6 .5

1 1

8 ,3

6 9

.9 7

N u

m b

e r

o f

p a ra

m e te

rs 1

1 1

6 1

6 1

2 1

2 1

7 1

7 2

4

D d

e v

ia n

c e

v e rs

u s

m o

d e l

1 1

2 9

.5 3

* *

* 3

7 .1

8 *

* *

5 .3

1 *

1 .2

8 1

3 2

.3 6

* *

* 3

8 .9

3 *

* *

1 5

5 .4

7 *

* *

P se

u d

o -R

2 .0

3 5

.0 4

5 .0

4 0

.0 5

1 .0

4 1

.0 5

0 .0

4 3

.0 5

9

U n

st a n

d a rd

iz e d

c o

e ffi

c ie

n t

e st

im a te

s a re

re p

o rt

e d

� p \

.1 0

; *

p \

.0 5

; *

* p \

.0 1

; *

* *

p \

.0 0

1

298 D. A. Ralston et al.

123

the aggregated individualism value was not significantly

related. Consistent with these results, VSM individualism/

collectivism was negatively related and institutional col-

lectivism was positively related to maliciously intended

behaviors, and the two individualistic Schwartz societal

values (affective autonomy and intellectual autonomy)

were not significantly related to maliciously intended

behaviors. Inconsistent with the aggregated values results,

meta-analytic VSM individualism/collectivism, in-group

collectivism, and embeddedness were not significantly

related to maliciously intended behaviors.

Although these alternative measures of societal-level

individualism and collectivism have been identified as

similar at a construct level (e.g., Hofstede 2001; House

Table 7 HLM results for alternative collectivism and individualism societal values dimensions as predictors of perceived ethical behavior

Model Pro-organizational Image

management

Self-serving Maliciously

intended

I VSM individualism/collectivism (c01) -.0001 .001 .001 -.003*

1 Deviance: covariate model 26,634.43 43,553.79 39,407.15 17,422.34

2 D deviance: individual-level values 196.42*** 152.33*** 123.32*** 120.69***

3 D deviance: VSM individualism 0.00 0.27 1.66 2.56

II Meta-analytic VSM individualism/collectivism (c01) .033 .030 .150 .036

1 Deviance: covariate model 17,290.49 29,232.82 26,767.82 11,061.01

2 D deviance: individual-level values 174.88*** 106.25*** 109.35*** 95.48***

3 D deviance: meta-analytic individualism 0.35 0.06 2.09 0.81

III GLOBE cultural values

a In-group collectivism (c01) .166* -.075 -.219 .055

b Institutional collectivism (c01) .182** -.428*** -.189 .138**

1 Deviance: covariate model 20,616.49 35,172.10 31,219.54 13,213.50

2 D deviance: individual-level values 175.33*** 150.39*** 112.46*** 129.50***

3a D deviance: in-group collectivism 4.82* 0.18 1.78 0.90

3b D deviance: institutional collectivism 10.69** 12.26*** 2.13 10.60**

IV Schwartz societal values

a Embeddedness (c01) -.033 .109 -.150 -.010

b Affective autonomy (c01) .037 -.180 -.077 .054

c Intellectual autonomy (c01) .445*** -.119 -.083 .097

1 Deviance: covariate model 18,617.43 31,754.66 28,383.87 12,030.86

2 D deviance: individual-level values 191.64*** 124.03*** 119.49*** 104.78***

3a D deviance: embeddedness 0.30 0.61 1.22 0.05

3b D deviance: affective autonomy 0.16 1.28 0.28 2.69

3c D deviance: intellectual autonomy 17.25*** 0.21 0.14 1.80

Unstandardized coefficient estimates are reported. Deviance statistics df = 11 for model 1, df = 2 for model 2, df = 1 for model 3 �

p \ .10; * p \ .05; ** p \ .01; *** p \ .001

Table 8 Summary of relationships for collectivism

and individualism values with

ethics behaviors at the

individual and societal levels of

analyses

n.s. nonsignificant

Values Ethics behaviors Significance/direction of relationship

Individual-level Societal-level

Collectivism Pro-organizational Positive n.s.

Image management Negative n.s.

Self-serving n.s. n.s

Maliciously intended Positive Positive

Individualism Pro-organizational Negative n.s.

Image management Positive n.s.

Self-serving Positive n.s.

Maliciously intended Negative n.s.

Societal-Level Versus Individual-Level Predictions 299

123

et al. 2004; Schwartz 1999), there is substantial variability

in results when using the operationalized values scores to

predict ethical behaviors.

Discussion

A Summary of the Individual-Level and Societal-Level

Findings

For the societal-level and individual-level analyses, two

patterns emerged. At the societal-level, with only one

exception, the pattern is one of nonsignificance for the

relationships between both collectivism and individualism

with the ethics behaviors dimensions. As noted, the only

significant finding at the societal-level was a positive

relationship between collectivism and maliciously intended

ethics behaviors. Conversely, for the individual-level

analyses, the only nonsignificant relationship was between

collectivism and self-serving ethics behavior. Thus, as the

hypotheses proposed and Table 8 summarizes, the indi-

vidual-level analysis was found to have significantly

greater predictive power than the societal-level analysis for

estimating both collectivism and individualism values

predictions of ethical behavior.

A Longitudinal Perspective of the Individual-Level vis-

à-vis the Societal-Level

First, it is interesting to note that our results are consistent

with those of Taras et al. (2010a) whose meta-analysis of

Hofstede’s VSM cultural values found that the predictive

power of values is higher for primary data compared to

secondary data. Since Taras et al. (2010a) conducted a

meta-analysis, they could not directly compare levels of

analysis with the same data sets. However, our findings

based on multi-level analyses yield a similar conclusion.

Whereas a substantial proportion of the studies in Taras

et al.’s (2010a) comprehensive meta-analysis were con-

ducted during the latter quarter of the twentieth century

(with some studies dating back to the 1950s), our sub-

sidiary analyses revealed a lack of predictive power for the

updated VSM individualism/collectivism societal scores

(Taras et al. 2012).

The subsidiary analyses using collectivism and indi-

vidualism societal-level scores from alternative cultural

values typologies raise some cautions about their use in

cross-cultural research. While proposed to be representing

similar theoretical constructs, there were inconsistent and

sometimes contradictory findings across these measures of

collectivism and individualism values. Of the five alter-

native sets of individualism and collectivism societal-level

scores, only institutional collectivism was found to be a

significant predictor of more than one type of ethical

behavior. Thus, our empirical findings provide support for

the extensive array of conceptual and methodological cri-

tiques that have previously been reported for these cultural

values typologies and their measures (e.g., Brewer and

Venaik 2011; Ralston et al. 2011; Taras et al. 2010b; Tung

and Verbeke 2010). Whereas Taras et al. (2010a) proposed

a moratorium on the use of Hofstede’s VSM scores based

on 1960s–1970s data, one implication of our findings is

that with one exception, the relationship between values

and ethical behaviors is more a function of the level of

analysis with the individual-level being more predictive

than the societal-level.

The trends we reported in respect to evolving socio-

political change lead to another implication of our study

concerning changes in intra-cultural variation. We agree

with Au and Cheung (2004) that a possible explanation as to

why individual-level values have substantially higher pre-

dictive power is because of the existence of large intra-

cultural variations in many countries, with technology-led

trends predicting even greater intra-cultural variation within

societies in the future. Hence, we present Fig. 3 as a lon-

gitudinal representation of the apparent trends in societal

diversity. As illustrated in Fig. 3, our discussion implies:

first, the mean of a society’s values is not a good repre-

sentation of the values within the society today; second, the

mean of a society’s values will become an even poorer

representation in the future; and third, organization success

will increasingly depend on the organization’s ability to

manage intra-societal values-based cultural diversity.

A Comparison of Collectivism vis-à-vis Individualism

Values as Predictors of Ethics

Based on our findings, it appears clear that the focus of

further discussion should be on the individual-level of

analysis findings. Of particular interest is the nature of the

relationships of collectivism and individualism values with

the four ethical behavior dimensions, and the distribution

pattern of collectivism vis-à-vis individualism values

responses depicted in Fig. 1b. On one hand, we see an

inverse relationship between the collectivism and individ-

ualism dimensions, while on the other hand, we see that

this relationship is too dispersed to justify considering

these two dimensions as points on a single continuum (see

Ralston 2008).

As previously identified, a hierarchy of the SIE behav-

iors has been consistently found across a wide range of

cultures (Ralston et al. 2009; Ralston and Pearson 2010).

Pro-organizational behavior is viewed as the most ethical,

image management as the next most ethical, self-serving as

the third most ethical, and maliciously intended as the least

ethical behaviors. We found that collectivistic business

300 D. A. Ralston et al.

123

professionals had significantly more positive views of the

extreme ethics behaviors (organizationally beneficial and

maliciously intended) while they had more negative views

of the intermediate image management ethics behavior.

Conversely, individualistic professionals had significantly

more positive views of the intermediate ethics behaviors

(image management and self-serving) and more negative

views of the extreme ethics behaviors (organizationally

beneficial and maliciously intended). This certainly begs

the question: Why might we expect to find this set of

relationships for collectivism and individualism with ethi-

cal behaviors? We next postulate as to the ‘‘why’’ and

propose topics for future research.

Collectivism

Embracing the ethical extremes that we see for collectivism

is perhaps the more interesting, as well as less intuitively

obvious, of the two sets of relationships. However, a

potential explanation emerges when considering these

collectivism findings from the in-group/out-group context

(Pekerti and Kwantes 2011). As Triandis (1995) noted, the

in-group and out-group differentiation is greater for allo-

centric (collectivistic) individuals than it is for ideocentric

(individualistic) individuals. Thus, there is a very clear

distinction as to whether you are one of us or you are not. If

you are one of us, you are treated in a very benevolent way

and if you are not, malicious treatment is deemed accept-

able behavior. Our findings suggest that when collectivistic

individuals consider ethical behaviors they are compart-

mentalizing these into three categories: (1) things you

would do for in-group members (e.g., pro-organizational

ethical behavior); (2) things you would be willing to do to

out-group members (e.g., maliciously intended ethical

behavior); and (3) things you would do for yourself. In

regard to this third category, collectivism is about the

welfare of the group and an individual’s needs should be

subservient to those of the group. Thus, image management

behaviors or self-serving behaviors are viewed as sub-

stantially unacceptable. Consequently, group-orientation

may explain why collectivists evaluate these behaviors

negatively.

Individualism

Conversely, individualism is oriented toward self-needs

and individualists do not view these needs to be subservient

to those of the group. The individualist embraces the tril-

ogy of me, myself, and I. Our findings indicate that the

individualists appear to focus upon the self-promoting

image management and self-serving ethical behaviors.

Also as Triandis (1995) noted, ideocentric (individualistic)

individuals do not discriminate between the in-group and

out-group nearly as much as do allocentric (collectivistic)

individuals. In fact, Triandis went beyond this two-group

classification to add a third group, the like-group, particu-

larly for the individualists. The like-group might be

described as a temporary in-group that is not accorded the

intense commitment accorded to the true in-group. In that

individualists see less distinction between types of group

membership, there are many more shades of gray for the

individualist than there are for the collectivist when

defining the meaning of group membership. Given the

lesser importance of the group and the lesser distinctions

between types of groups in conjunction with the high level

of importance attached to self-promotion, our findings may

be showing that individualists, as contrasted with collec-

tivists, have only one primary category: self. Thus, it may

be argued that investing time in doing things for others

(e.g., pro-organizational behavior) or to others (e.g.,

Fig. 3 A longitudinal approximation of intra-cultural variation (ICV). Adapted from Au and Cheung (2002)

Societal-Level Versus Individual-Level Predictions 301

123

maliciously intended behavior), while certainly not absent,

may not be the most efficient way to promote a ‘‘self’’

agenda.

A related discussion on the influence of individualism

and collectivism on ethical behavior is provided by Chen

et al. (2002), in the context of opportunistic propensity,

where ‘‘opportunism refers to the incomplete or distorted

disclosure of information, especially to calculated efforts to

mislead, distort, disguise, obfuscate, or otherwise confuse’’

(Williamson 1985, p. 47). They argue that collectivistic

and individualistic societies have different levels of

opportunistic propensity depending on the group mem-

bership of the target. That is, collectivistic societies tend to

be more opportunistic in out-group relationships and less

opportunistic in in-group relationships than are individu-

alistic societies. This argument is philosophically consis-

tent with the individual-level, theoretical perspective of

Triandis (1995) and with the individual-level, empirical

findings of our study. However, not integrated into Chen

et al.’s (2002) thinking is the argument for intra-cultural

variation that Au and Cheung (2004) later propose and that

Tung (2008) subsequently identified as crucial for devel-

oping an understanding of the values/behaviors in a

society.

Thus, to integrate the intra-cultural variation perspective

into opportunistic argument, we might transform the level

of their argument from the societal-level to the individual-

level. Then, we would find different opportunistic patterns

of ethical behavior within a society comprised of intra-

cultural variation (i.e., heterogeneous) populations that

included both collectivists and individualists. An individ-

ual-level of analysis will fully capture these distinctions in

behavior while a societal-level analysis will average-out

the findings, with the results being skewed in the direction

of whichever orientation—collectivistic or individualis-

tic—was the more prevalent in that society. One implica-

tion of the findings from both Chen et al. (2002) and our

study is that the relationship between values and ethical

behavior is more complex than initially thought, and it is

one that appears too complex to be accurately deciphered

using societal-level analyses. However, this complexity

also poses a question for future research to explore: Are

collectivistic individuals in predominantly individualistic

societies the same/more/less opportunistic with in-group/

out-group members as collectivistic individuals in pre-

dominantly collectivistic societies? A similar question

might also be posed for individualistic individuals in

individualistic vis-à-vis collectivistic societies.

Concluding Thoughts

The purpose of this study was to explore the question: Is

the use of societal-level values for cross-cultural analyses

acceptable and sufficient in the global economy of the

twenty-first century? Our findings based on business pro-

fessionals from 48 diverse societies were resoundingly

clear. For collectivism and individualism values, the indi-

vidual-level of analysis exhibited substantially more pre-

dictive power of ethical behavior than did the societal-level

of analysis. Thus, from highly ethical (e.g., supportive

teamwork) to highly unethical (e.g., corporate espionage)

behaviors, individual-level analyses present a more com-

plete picture of reality in the global business environment.

As such, our findings seriously challenge the use of soci-

etal-level values scores to predict managerial behavior.

Implicitly, these findings affirm our prediction of values

heterogeneity within the workforces of the societies in this

study.

Over a decade ago, Ralston et al. (1997) proposed that a

societal crossvergence-effect occurs when differing socio-

cultural and business ideology influences impact an indi-

vidual. The more recent bicultural concept (e.g., Thomas

et al. 2010) proposed a perspective that is philosophically

consistent with the crossvergence perspective, albeit at the

individual-level. As we have discussed, both societal

crossvergence and individual bi-culturalism are widespread

phenomena in the current world of business. Thus, an

integration of these two concepts appears to explain much

of the impetus behind the dynamic values evolution

occurring within societal workforces across the globe.

Consequently, as a result of exponentially growing tech-

nological advancements and transitioning economies, in

conjunction with the offspring of interpersonal-mergers,

we are in an accelerating state of crossverging values

evolution of workforce members within societies. The

result is a high degree of heterogeneity of work values

within societies (i.e., intra-cultural variation) that will

continue to increase into the future.

In sum, the modern reality is that political boundaries

are not surrogates for the work values of its inhabitants.

These boundaries do not well define the thinking of the

workforces within them and to assume the contrary will

likely lead to erroneous conclusions. Thus, it is our view

that to thoroughly understand the values/behaviors of those

in the global workforce, we need to conduct our empirical

investigations at the individual-level of analysis. As such,

the research methodologies to be applied should likely be

rethought, if they are to be fully relevant. However, this is

much easier said than done. There are few individual-level

databases available for all to use, and it is truly challenging

for individual researchers to develop their own databases.

Accordingly, there may be the temptation to ‘‘look the

other way’’ and to continue to use societal-level values

simply because they are readily available. However, our

findings do not support societal-level analyses as being a

viable alternative. Moreover, current trends (e.g.,

302 D. A. Ralston et al.

123

technological advancement) suggest that the societal-level

will be an even less viable approach in the future. Conse-

quently, instead of looking the other way, perhaps we

should proactively view this situation as a challenge for

cross-cultural researchers to find ways to move forward in

developing public, individual-level databases to elevate the

cross-cultural management discipline to a higher level.

References

Alas, R. (2006). Ethics in countries with different cultural dimensions.

Journal of Business Ethics, 69, 237–247.

Amin, A. (2002). Spatialities of globalization. Environment and

Planning, 34, 385–399.

Anastasi, A. (1982). Psychological testing (2nd ed.). New York:

MacMillian.

Au, K. Y. (1999). Intra-cultural variation: Evidence and implications

for international business. Journal of International Business

Studies, 30(4), 700–812.

Au, K. Y. (2000). Intra-cultural variation as another construct of

international management: A study based on secondary data of

42 countries. Journal of International Management, 6, 217–238.

Au, K., & Cheung, M. W. L. (2004). Intra-cultural variation and job

autonomy in 42 countries. Organization Studies, 25, 1339–1362.

Bartol, K. M., Martin, D. C., & Kromkowski, J. A. (2003). Leadership

and the glass ceiling: Gender and ethnic group influences on

leader behaviors at middle executive managerial levels. Journal

of Leadership and Organizational Studies, 9, 8–19.

Bielby, W. (2000). Minimizing workplace gender and racial bias.

Contemporary Sociology, 9(1), 120–130.

Bond, M. H., Leung, K., Au, A., Tong, K. K., Reimel de Carrasquel,

S., Murakami, F., et al. (2004). Culture-level dimensions of

social axioms and their correlates across 41 cultures. Journal of

Cross-Cultural Psychology, 35, 548–570.

Brewer, P., & Venaik, S. (2011). Individualism–collectivism in

Hofstede and GLOBE. Journal of International Business Stud-

ies, 42, 436–445.

Brown, R. D., & Hauenstein, N. M. A. (2005). Interrater agreement

reconsidered: An alternative to rWG indices. Organizational

Research Methods, 8, 165–184.

Buchholz, S., Hofäcker, D., Mills, M., Blossfeld, H. P., Kurz, K., &

Hofmeister, H. (2009). Life courses in the globalization process:

The development of social inequalities in modern societies.

European Sociological Review, 25(1), 53–71.

Chen, C. C., Chen, X.-P., & Meindl, J. R. (1998). How can

cooperation be fostered? The cultural effects of individualism–

collectivism. Academy of Management Review, 23, 285–304.

Chen, C. C., Peng, M. W., & Saparito, P. A. (2002). Individualism,

collectivism, and opportunism: A cultural perspective on trans-

action cost economics. Journal of Management, 28, 567–583.

Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit

indexes for testing measurement invariance. Structural Equation

Modeling, 9, 233–255.

Chhokar, J. S., Brodbeck, F. C., & House, R. J. (Eds.). (2008). Culture

and leadership across the world: The GLOBE book of in-depth

studies of 25 societies. Mahwah, NJ: Lawrence Erlbaum.

Choi, J., & Chen, C. C. (2006). Gender differences in perceived work

demands, family demands and life stress among married Chinese

employees. Management and Organization Review, 2(2), 209–

229.

Colakoglu, S., & Caligiuri, P. (2008). Cultural distance, expatriate

staffing and subsidiary performance: The case of US subsidiaries

of multinational corporations. International Journal of Human

Resource Management, 19, 223–239.

Collins, D. (2000). The quest to improve the human condition: The

first 1500 articles published in Journal of Business Ethics.

Journal of Business Ethics, 26(1), 1–73.

Costa, P, Jr, Terracciano, A., & McCrae, R. R. (2001). Gender

differences in personality traits across cultures: Robust and

surprising findings. Journal of Personality and Social Psychol-

ogy, 81, 322–331.

Cullen, J. B., Parboteeah, K. P., & Hoegl, M. (2004). Cross-national

differences in managers’ willingness to justify ethically suspect

behavior: A test of institutional anomie theory. Academy of

Management Journal, 47, 411–421.

Eagly, A. H., & Johannesen-Schmidt, M. C. (2001). The leadership

styles of women and men. Journal of Social Issues, 57, 781–797.

Eagly, A. H., Johannesen-Schmidt, M. C., & van Engen, M. L.

(2003). Transformational, transactional, and laissez-faire lead-

ership styles: A meta-analysis comparing women and men.

Psychological Bulletin, 129, 569–591.

Earley, P. (1993). East meets West meets Mideast: Further explora-

tions of collectivistic and individualistic work groups. Academy

of Management Journal, 36, 319–348.

Earley, P. C., & Gibson, C. B. (2002). Multinational teams: A new

perspective. Mahwah, NJ: Lawrence Erlbaum.

Egan, M. L., & Bendick, M, Jr. (2008). Combining multicultural

management and diversity into one course on cultural competence.

Academy of Management Learning & Education, 7(3), 387–393.

Egri, C. P., & Ralston, D. A. (2004). Generation cohorts and personal

values: A comparison of China and the U.S. Organization

Science, 15, 210–220.

Enders, C. K., & Tofighi, D. (2007). Centering predictor variables in

cross-sectional multilevel models: A new look at an old issue.

Psychological Methods, 12(2), 121–138.

Erez, M., & Earley, P. C. (1987). Comparative analysis of goal-setting

strategies across cultures. Journal of Applied Psychology, 72,

658–665.

Erez, M., & Somech, A. (1996). Is group productivity loss the rule or

the exception? Effects of culture and group-based motivation.

Academy of Management Journal, 39(6), 1513–1537.

Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton.

Erikson, E. H. (1997). The life cycle completed. New York: W.W.

Norton.

Ferdinand, P. (2007). Russia and China: Converging responses to

globalization. International Affairs, 83, 655–680.

Fischer, R. (2004). Standardization to account for cross-cultural

response bias: A classification of score adjustment procedures

and review of research in JCCP. Journal of Cross-Cultural

Psychology, 35, 263–282.

Fischer, R. (2009). Where is culture in cross cultural research? An

outline of a multilevel research process for measuring culture as

a shared meaning system. International Journal of Cross-

Cultural Management, 9(1), 25–29.

Fischer, R., Milfont, T. L., & Gouveia, V. V. (2011). Does social

context affect value structures? Testing the within-country

stability of value structures with a functional theory of values.

Journal of Cross-Cultural Psychology, 42, 253–270.

Forsyth, D., O’Boyle, E., & McDaniel, M. (2008). East meets west: A

meta-analytic investigation of cultural variations in idealism and

relativism. Journal of Business Ethics, 83, 813–833.

Fu, P. P., Kennedy, J., Tata, J., Yukl, G., Bond, M. H., Peng, T.-K.,

et al. (2004). The impact of societal cultural values and

individual social beliefs on the perceived effectiveness of

managerial influence strategies: A meso approach. Journal of

International Business Studies, 35, 284–305.

Societal-Level Versus Individual-Level Predictions 303

123

Fu, P. P., & Yukl, G. (2000). Perceived effectiveness of influence

tactics in the United States and China. Leadership Quarterly, 11,

251–266.

Fullagar, C. J., Sverke, M., Sumer, H., & Slick, R. (2003). Managerial

sex-role stereotyping: A cross cultural analysis. International

Journal of Cross Cultural Management, 3, 93–107.

Gartin, T., Shroyer, E., & Neidermeyer, P. (2009). Transitional

economies of Europe and the development of financial reporting

standards: A look at the correlation between a successful

economic transition and the development of financial reporting

standards and laws. International Business & Economics

Research Journal, 8(10), 19–33.

Gelfand, M. J., Erez, M., & Aycan, Z. (2007). Cross-cultural

organizational behavior. In M. J. Posner and M. K. Rothbart

(Eds.). Annual Review of Psychology, 58, 479–514.

Gratton, L. (2000). Living strategy: Putting people in the heart of

corporate purpose. Upper Saddle River, NJ: Financial Times Press.

Groves, K. S. (2005). Gender differences in social and emotional

skills and charismatic leadership. Journal of Leadership and

Organizational Studies, 11, 30–46.

Gurven, M., Zanolini, A., & Schniter, E. (2008). Culture sometimes

matters: Intra-cultural variation in pro-social behavior among

Tsimane Amerindians. Journal of Economic Behavior & Orga-

nization, 62, 587–607.

Hanges, P. L. (2004). Response bias correction procedure used in

GLOBE. In R. J. House, P. J. Hanges, M. Javidan, P.

W. Dorfman, & V. Gupta (Eds.), Culture, leadership, and

organizations: The GLOBE study of 62 societies (pp. 737–751).

Thousand Oaks, CA: Sage.

Haslberger, A., & Brewster, C. (2009). Capital gains: Expatriate

adjustment and the psychological contract in international

careers. Human Resource Management, 23, 324–346.

Hofmann, D. A., & Gavin, M. B. (1998). Centering decisions in

hierarchical linear models: Implications for research in organi-

zations. Journal of Management, 24(5), 623–641.

Hofstede, G. (2001). Culture’s consequences: Comparing values,

behaviors, institutions, and organizations across nations (2nd

ed.). Thousand Oaks, CA: Sage.

House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., & Gupta, V.

(Eds.). (2004). Culture, leadership, and organizations: The

GLOBE study of 62 societies. Thousand Oaks, CA: Sage.

Hummels, D. (2007). Transportation costs and international trade in

the second era of globalization. Journal of Economic Perspec-

tives, 21(3), 131–154.

Husted, B. W., & Allen, D. B. (2008). Toward a model of cross-

cultural business ethics: The impact of individualism and

collectivism on the ethical decision-making process. Journal of

Business Ethics, 82, 293–305.

Illies, J., & Reiter-Palmon, R. (2008). Responding destructively in

leadership situations: The role of personal values and problem

construction. Journal of Business Ethics, 82, 251–272.

Inglehart, R. (1997). Modernization and postmodernization: Cultural,

economic, and political change in 43 societies. Princeton, NJ:

Princeton University Press.

Inglehart, R., & Welzel, C. (2010). Changing mass priorities: The link

between modernization and democracy. Perspectives on Politics,

8(2), 551–567.

Jaffee, S., & Hyde, J. S. (2000). Gender differences in moral

orientation: A meta-analysis. Psychological Bulletin, 126(5),

703–726.

Janson, M., Cecez-Kecmanovic, D., & Zupančič, J. (2007). Prosper-

ing in a transition economy through information technology-

supported organizational learning. Information Systems Journal,

17, 3–36. Kabasakal, H., Dastmalchian, A., & Imer, P. (2011). Organizational

citizenship behaviour: A study of young executives in Canada,

Iran, and Turkey. International Journal of Human Resource

Management, 22, 2703–2729.

Kirkman, B. L., Chen, G., Farh, J. L., Chen, Z. X., & Lowe, K. B.

(2009). Individual power distance orientation and follower

reactions to transformational leaders: A cross-level, cross-

cultural examination. Academy of Management Journal, 52,

744–764.

Kirkman, B. I., Lowe, K. B., & Gibson, C. B. (2006). A quarter

century of culture’s consequences: A review of empirical

research incorporating Hofstede’s cultural values framework.

Journal of International Business Studies, 37(3), 285–320.

Kish-Gephart, J. J., Harrison, D. A., & Trevino, L. K. (2010). Bad

apples, bad cases, and bad barrels: Meta-analytic evidence about

sources of unethical decisions at work. Journal of Applied

Psychology, 95(1), 1–31.

Klein, K. J., & Kozlowski, S. W. J. (2000). From micro to meso:

Critical steps in conceptualizing and conducting multilevel

research. Organization Research Methods, 3, 211–236.

Kluckhohn, C., & Strodtbeck, K. (1961). Variations in value

orientations. Westport, CT: Greenwood Press.

Kreft, I. G. G. (2000). Using random coefficient linear models for the

analysis of hierarchically nested data. In H. E. A. Tinsey & S.

D. Brown (Eds.), Handbook of applied multivariate statistics and

mathematical models (pp. 613–639). New York: Academic

Press.

LeBreton, J. M., & Senter, J. L. (2008). Answers to 20 questions

about interrater reliability and interrater agreement. Organiza-

tional Research Methods, 11, 815–852.

Lenartowicz, T., & Roth, K. (2001). Does subculture within a country

matter? A cross-cultural study of motivational domains and

business performance in Brazil. Journal of International Busi-

ness Studies, 32(2), 305–326.

Licht, A. N., Goldschmidt, C., & Schwartz, S. H. (2007). Culture rules:

The foundations of the rule of law and other norms of

governance. Journal of Comparative Economics, 35(4), 659–688.

Lindeman, M., & Verkasalo, M. (2005). Measuring values with the

short Schwartz’s value survey. Journal of Personality Assess-

ment, 85(2), 170–178.

Ma, Z. (2010). The SINS in business negotiations: Explore the cross-

cultural differences in business ethics between Canada and

China. Journal of Business Ethics, 91, 123–135.

Malecki, E. J., & Ewers, M. C. (2007). Labor migration to world

cities: With a research agenda for the Arab Gulf. Progress in

Human Geography, 31, 467–484.

Martin, K. D., Cullen, J. B., Johnson, J. L., & Parboteeah, K. P.

(2007). Deciding to bribe: A cross-level analysis of firm and

home country influences on bribery activity. Academy of

Management Journal, 50(6), 1401–1422.

Martinko, M. J., Gundlach, M. J., & Douglas, S. C. (2002). Toward an

integrative theory of counterproductive workplace behavior: A

causal reasoning perspective. International Journal of Selection

and Assessment, 10, 36–50.

McSweeney, B. (2002). Hofstede’s model of national cultural

differences and their consequences: A triumph of faith—a

failure of analysis. Human Relations, 55(1), 89–118.

Meglino, B. M., & Ravlin, E. C. (1998). Individual values in

organizations: Concepts, controversies, and research. Journal of

Management, 24, 351–389.

Miller, J., & Tenev, S. (2007). On the role of government in

transition: The experiences of China and Russia compared.

Comparative Economic Studies, 49, 543–571.

Offermann, L. R., & Hellmann, P. S. (1997). Culture’s consequences

for leadership behavior: National values in action. Journal of

Cross-Cultural Psychology, 28, 342–351.

Oyserman, D., Coon, H. M., & Kemmelmeier, M. (2002). Rethinking

individualism and collectivism: Evaluation of theoretical

304 D. A. Ralston et al.

123

assumptions and meta-analyses. Psychological Bulletin, 128,

3–72.

Painter-Morland, M., & Ten Bos, R. (Eds.). (2011). Business ethics

and continental philosophy. Cambridge: Cambridge University

Press.

Pan, Z., Chaffee, S. H., Chu, G. C., & Ju, Y. (1994). To see ourselves:

Comparing traditional Chinese and American cultural values.

Boulder, CO: Westview Press.

Parboteeah, P. K., Hoegl, M., & Cullen, J. (2009). Religious

dimensions and work obligation: A country institutional profile

model. Human Relations, 62, 119–148.

Pekerti, A. A., & Kwantes, C. (2011). The effect of self-construals on

perceptions of organizational events. International Journal of

Cross-Cultural Management, 11(3), 303–323.

Peng, Y., & Lin, S. (2009). National culture, economic development,

population growth and environmental performance: The medi-

ating role of education. Journal of Business Ethics, 90, 203–219.

Peterson, M. F. (2004). Culture, leadership and organizations: The

GLOBE study of 62 societies. Administrative Science Quarterly,

49, 641–647.

Peterson, M. F., Arregle, J.-L., & Martin, X. (2012). Multilevel

models in international business research. Journal of Interna-

tional Business Studies, 43(5), 451–457.

Peterson, M. F., & Castro, S. L. (2006). Measurement metrics at

aggregate levels of analysis: Implications for organization

culture research and the GLOBE project. Leadership Quarterly,

17, 506–521.

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P.

(2003). Common method biases in behavioral research: A

critical review of the literature and recommended remedies.

Journal of Applied Psychology, 88, 879–903.

Ralston, D. A. (2008). The crossvergence perspective: Reflections and

projections. Journal of International Business Studies, 39,

27–40.

Ralston, D. A., Egri, C. P., de la Garza-Carranza, M. T., Ramburuth,

P., Terpstra-Tong, J., Pekerti, A., et al. (2009). Ethical prefer-

ences for influencing superiors: A 41-society study. Journal of

International Business Studies, 40, 1022–1045.

Ralston, D. A., Egri, C. P., Reynaud, E., Srinivasan, N., Furrer, O.,

Brock, D., et al. (2011). A 21st century assessment of values

across the global workforce. Journal of Business Ethics, 104,

1–31.

Ralston, D. A., Egri, C. P., Stewart, S., Terpstra, R. H., & Yu, K. C.

(1999a). Doing business in the 21st century with the new

generation of Chinese manager: A study of generational shifts in

work values in China. Journal of International Business Studies,

30, 415–428.

Ralston, D. A., Hallinger, P., Egri, C. P., & Naothinsuhk, S. (2005).

The effects of culture and life stage on workplace strategies of

upward influence: A comparison of Thailand and the United

States. Journal of World Business, 30, 321–337.

Ralston, D. A., Holt, D. H., Terpstra, R. H., & Yu, K. C. (1997). The

impact of national culture and economic ideology on managerial

work values: A study of the United States, Russia, Japan, and

China. Journal of International Business Studies, 28, 177–208.

Ralston, D. A., Nguyen, V. T., & Napier, N. K. (1999b). A

comparative study of the work values of North and South

Vietnamese managers. Journal of International Business Studies,

30, 655–672.

Ralston, D. A., & Pearson, A. (2010). The Cross-cultural evolution of

the subordinate influence ethics measure. Journal of Business

Ethics, 96, 149–168.

Ralston, D. A., Yu, K. C., Wang, X., Terpstra, R. H., & He, W.

(1996). The cosmopolitan Chinese manager: Findings of a study

on managerial values across the six regions of China. Journal of

International Management, 2, 79–109.

Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear

models: Applications and data analysis methods (2nd ed.).

Thousand Oaks, CA: Sage.

Robertson, C., & Fadil, P. A. (1999). Ethical decision making in

multinational organizations: A culture-based model. Journal of

Business Ethics, 19, 385–392.

Robertson, C., Ralston, D. A., & Crittenden, W. (2012). The

relationship between cultural values and moral philosophy: A

generational subculture theory approach. Academy of Marketing

Science Review, 12(2), 99–107.

Rokeach, M. (1973). The nature of human values. New York: Free Press.

Roxas, M. L., & Stoneback, J. Y. (2004). The importance of gender

across cultures in ethical decision-making. Journal of Business

Ethics, 50, 149–165.

Schimmack, U., Oishi, S., & Diener, E. (2005). Individualism: A

valid and important dimension of cultural differences between

nations. Personality and Social Psychology Review, 9(1), 17–31.

Schumann, J. H., Wünderlich, N. V., & von Wangenheim, F. (2012).

Technology mediation in service delivery: A new typology and

an agenda for managers and academics. Technovation, 32(2),

133–143.

Schwartz, S. H. (1992). Universals in the content and structure of

values: Theoretical advances and empirical tests in 20 countries.

In M. P. Zanna (Ed.), Advances in experimental social psychol-

ogy (pp. 1–65). San Diego, CA: Academic Press.

Schwartz, S. H. (1994). Beyond individualism/collectivism: New

dimensions of values. In U. Kim, H. C. Triandis, C. Kagitçibasi,

S. C. Choi, & G. Yoon (Eds.), Individualism and collectivism:

Theory, applications, and methods (pp. 85–119). Newbury Park,

CA: Sage.

Schwartz, S. H. (1999). A theory of cultural values and some

implications for work. Applied Psychology: An International

Review, 48(1), 23–47.

Schwartz, S. H. (2006). A theory of cultural value orientations:

Explication and applications. Comparative Sociology, 5, 137–182.

Schwartz, S. H., & Rubel, T. (2005). Sex differences in value

priorities: Cross-cultural and multimethod studies. Journal of

Personality and Social Psychology, 89(6), 1010–1028.

Schwartz, S. H., & Rubel-Lifschitz, T. (2009). Cross-national

variation in the size of sex differences in values: Effects of

gender equality. Journal of Personality and Social Psychology,

97(1), 171–185.

Settersten, R. A, Jr, & Mayer, K. U. (1997). The measure of age, age

structuring, and the life course. Annual Review of Sociology, 23,

233–261.

Snijders, T. A. B., & Bosker, R. J. (1994). Modeled variance in two-

level models. Sociological Methods and Research, 22, 342–363.

Spector, P. E., & Cooper, C. L. (2002). The pitfalls of poor

psychometric properties: A rejoinder to Hofstede’s reply to us.

Applied Psychology: An International Review, 51(1), 174–178.

Spector, P. E., Cooper, C., Sanchez, J., O’Driscoll, M., et al. (2001a).

Do national levels of individualism and internal locus of control

relate to well-being: An ecological level international study.

Journal of Organizational Behavior, 22(8), 815–832?.

Spector, P. E., Cooper, C. L., & Sparks, K. (2001b). An international

study of the psychometric properties of the Hofstede Values

Survey Module 1994: A comparison of individual and country/

province level results. Applied Psychology: An International

Review, 50(2), 269–281.

Steel, P., & Taras, V. (2010). Culture as a consequence: A multi-level

multivariate meta-analysis of the effects of individual and

country characteristics on work-related cultural values. Journal

of International Management, 16, 211–233.

Steenkamp, J. B. M., & Baumgartner, H. (1998). Assessing

measurement invariance in cross-national consumer research.

Journal of Consumer Research, 25(1), 78–90.

Societal-Level Versus Individual-Level Predictions 305

123

Taras, V., Kirkman, B. L., & Steel, P. (2010a). Examining the impact

of culture’s consequences: A three-decade, multi-level, meta-

analytic review of Hofstede’s cultural value dimensions. Journal

of Applied Psychology, 95(3), 405–439.

Taras, V., Rowney, J., & Steel, P. (2009). Half a century of measuring

culture: Approaches, challenges, limitations, and suggestions

based on the analysis of 112 instruments for quantifying culture.

Journal of International Management, 15(4), 357–373.

Taras, V., Steel, P., & Kirkman, B. L. (2010b). Negative practice–

value correlations in the GLOBE data: Unexpected findings,

questionnaire limitations and research directions. Journal of

International Business Studies, 41, 1330–1338.

Taras, V., Steel, P., & Kirkman, B. L. (2012). Improving national

cultural indices using a longitudinal meta-analysis of Hofstede’s

dimensions. Journal of World Business, 47, 329–341.

Thomas, D. C., Brannen, M. Y., & Garcia, D. (2010). Bicultural

individuals and intercultural effectiveness. European Journal of

Cross-Cultural Competence and Management, 1, 315–333.

Trevino, L., Weaver, G. R., Gibson, D. G., & Toffler, B. (1999).

Managing ethics and legal compliance: What works and what

hurts. California Management Review, 41(2), 131–151.

Trevino, L. K., Weaver, G. R., & Reynolds, S. J. (2006). Behavioral

ethics in organizations: A review. Journal of Management,

32(6), 951–990.

Triandis, H. C. (1995). Individualism and collectivism. Boulder, CO:

Westview.

Tsui, A. S., Nifadkar, S. S., & Ou, A. Y. (2007). Cross-national, cross-

cultural organizational behavior research: Advances, gaps, and

recommendations. Journal of Management, 33(3), 426–478.

Tung, R. (2008). The cross-cultural research imperative: The need to

balance cross-national and intra-national diversity. Journal of

International Business Studies, 39, 41–46.

Tung, R. L., & Verbeke, A. (2010). Beyond Hofstede and GLOBE:

Improving the quality of cross-cultural research. Journal of

International Business Studies, 41, 1259–1274.

Vitell, S. J., Nwachukwu, S. L., & Barnes, J. H. (1993). The effects of

culture on ethical decision-making: An application of Hofstede’s

typology. Journal of Business Ethics, 12, 753–760.

Voegtlin, C., Patzer, M., & Scherer, A. (2012). Responsible leadership

in global business: A new approach to leadership and its multi-

level outcomes. Journal of Business Ethics, 105, 1–16.

Wagner, J. A. (1995). Studies of individualism-collectivism: Effects

on cooperation in groups. Academy of Management Journal, 38,

152–172.

Wallace, A. F. C. (1970). Culture and personality. New York:

Random House.

Weeks, W., Moore, C., McKinney, J., & Longenecker, J. (1999). The

effects of gender and career stage on ethical judgment. Journal

of Business Ethics, 20, 301–313.

Westerman, J. W., Beekun, R. I., Stedham, Y., & Yamamura, J.

(2007). Peers versus national culture: An analysis of antecedents

to ethical decision-making. Journal of Business Ethics, 75,

239–252.

Williamson, O. E. (1985). The economic institutions of capitalism.

New York: Free Press.

Zanoni, P., Janssens, M., Benschop, Y., & Nkomo, S. (2010).

Unpacking diversity, grasping inequality: Rethinking difference

through critical perspectives. Organization, 17(1), 9–29.

306 D. A. Ralston et al.

123

Copyright of Journal of Business Ethics is the property of Springer Science & Business Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.