Review of Empirical Research Incorporating Hofstede's Cultural Values Framework

7/27/2019 Review of Empirical Research Incorporating Hofstede's Cultural Values Framework

1/37

A Quarter Century of "Culture's Consequences": A Review of Empirical ResearchIncorporating Hofstede's Cultural Values FrameworkAuthor(s): Bradley L. Kirkman, Kevin B. Lowe, Cristina B. GibsonSource: Journal of International Business Studies, Vol. 37, No. 3 (May, 2006), pp. 285-320Published by: Palgrave Macmillan JournalsStable URL: http://www.jstor.org/stable/3875261Accessed: 24/09/2010 13:58

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

http://www.jstor.org/action/showPublisher?publisherCode=pal.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve and extend access toJournal of

International Business Studies.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=palhttp://www.jstor.org/stable/3875261?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=palhttp://www.jstor.org/action/showPublisher?publisherCode=palhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/3875261?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=pal


2/37

Journalf International usiness tudies 2006)37, 285-320? 2006 Academy f InternationalusinessAllrights eserved 047-2506$30.00www.jibs.net

PERSPECTIVE

A q u a r t e r c en tury o f Cul t u r e ' s Consequences:r e v i e w o f empir i ca l research incorporatingHofstede's c u l t u r a l v a l u e s framework

BradleyL. Kirkman',Kevin B. Lowe2 andCristinaB. Gibson3'Department f Management,MaysBusinessSchool,TexasA&MUniversity, ollegeStation,TX,USA; Departmentf BusinessAdministration,osephM.Bryan choolofBusiness ndEconomics, niversityf NorthCarolina t Greensboro, reensboro, orthCarolina,USA; Graduate choolofManagement,Universityf California,rvine,California, SACorrespondence:Bradley L Kirkman, Department ofManagement, Mays Business School, TexasA&M University, 4221 TAMU, CollegeStation, TX 77843-4221, USA.Tel: + 1 979 845 8813;Fax:+ 1 979 845 9641;E-mail:[email protected]

Received: 22 August 2002Revised: 8 May 2005Accepted: 17 May 2005Online publication date: 4 May 2006

AbstractSince GeertHofstede'sCulture's onsequences:nternational ifferencesn Work-RelatedValuesSage, 1980) was published,researchers ave utilizedHofstede'sculturalvaluesframeworkn a wide varietyof empirical tudies.We review180studiespublished n40 businessand psychology ournals nd two internationalannual volumes between 1980 and June 2002 to consolidate what isempiricallyverifiableabout Hofstede'sculturalvalues framework.We discusslimitations n the Hofstede-inspiredesearchand make recommendations orresearcherswho use Hofstede's rameworkn the future.Journal f International usiness tudies 2006) 37, 285-320.doi:I0.I057/palgrave.jibs.8400202Keywords:Hofstede;ulturalalues;ross-culturalanagementIntroductionResearch using a variety of frameworkshas shown that nationalcultural values are related to workplace behaviors, attitudes andother organizational outcomes (e.g., Kluckhohn and Strodtbeck,1961; Hall, 1976; Hofstede, 1980a; Trompenaars,1993; Schwartz,1994; Ronen and Shenkar, 1985). Perhapsthe most influential ofcultural classifications is that of Geert Hofstede. Over two decadeshave passed since the publication of Culture's Consequences:InternationalDifferences n Work-RelatedValues(Hofstede, 1980a),inspiring thousands of empirical studies; however, a comprehen-sive review of the impact of Hofstede'sframework s lacking.' Tofillthis gap, we summarize and synthesize empirical research pub-lished between January 1980 and June 2002 that has appliedHofstede's framework to organizations. We focus on Hofstede'sframework rather than others, given evidence that it has had fargreater impact (Sivakumarand Nakata, 2001). For example, theSocial Science Citations Index indicates that Hofstede's work ismore widely cited than others (cited 1,800 times through 1999;Hofstede, 2001). Trompenaars (1993, iii), who has a competingframework,credits Hofstede 'foropening management's eyes to theimportance of the [cross-cultural management] subject'. Ourpurpose is both to summarize existing research and to direct andinform future research,ratherthan provide an in-depth discussionof Hofstede's original study, a critique (e.g., Schwartz,1994; Smithand Bond, 1999; McSweeney, 2002; Smith, 2002), or a replication


3/37

A quarter century of Culture'sConsequences BradleyKirkmant al286

(e.g., Punnett and Withane (1990); Shackleton andAll, 1990; Merritt,2000; Spectoret al., 2001a).We focus on aspects of Hofstede's work notdiscussed in recent reviews and meta-analyses. Forexample, most researchers focused exclusively onindividualism-collectivism (IND-COL)at the indi-vidual level of analysis (e.g., Triandis, 1995; Earleyand Gibson, 1998; Oyserman et al., 2002).Consequently, their implications and conclusionsare based on a much narrowerband of Hofstede-inspired research. Hofstede (2001) recentlyreviewed hundreds of studies published since hisoriginal book appeared in 1980. However, consis-tent with his opposition to applying the frameworkto the individual level, Hofstede 'ignores every-thing.., but the culture level comparisons' (Smith,2002, 123), thus missing an opportunity to drawconclusions across levels. Indeed, an analysis ofreferences reveals very little overlap between stu-dies we reviewed and those contained in previousreviews. Therefore our conclusions and implica-tions should add value beyond previous reviews, asour purpose was to review studies at different levelsof analysis and direction of effects to comprehen-sively integrate and synthesize the findings for allfive cultural value dimensions in Hofstede's frame-work. As we shall show, without such a compre-hensive review, much of the Hofstede-inspiredresearch has remained fragmented and in somecases redundant, and researchers are unable tobenefit from the cumulative knowledge thataccruesfrom an integratedbody of quality research.First, we provide a brief overview of Hofstede'scultural value dimensions and how they werederived, and then we discuss our typology and therationale for including/excluding articles.Next, wereview findings, research accomplishments andchallenges within each major domain of ourtypology. Finally, we provide direction for futureHofstede-inspiredresearch.An overview of Hofstede's cultural valuedimensionsHofstede (1980a, 25) defined culture as 'the collec-tive programming of the mind which distinguishesthe members of one human group from another'.His framework was developed using data from over116,000 morale surveys from over 88,000 employ-ees from 72 countries (reduced to 40 countries thathad more than 50 responses each) in 20 languagesat IBM between 1967 and 1969 and again between1971 and 1973. He later expanded the databasewith 10 additional countries and three regions (i.e.,

Arab countries and Eastand West Africa).Basedona country level factor analysis, he classified theoriginal 40 countries along four dimensions. Thefirstis IND-COL,with INDdefined as 'a loosely knitsocial frameworkin which people are supposed totake care of themselves and of their immediatefamilies only', while COL 'is characterized by atight social frameworkin which people distinguishbetween ingroups and outgroups, they expect theiringroup to look after them, and in exchange forthat they feel they owe absolute loyalty to it'(Hof-stede, 1980b, 45). The second dimension is powerdistance (PD), defined as 'the extent to which asociety accepts the fact that power in institutionsand organizations is distributedunequally' (1980b,45). Third,uncertainty avoidance (UA)is defined as'the extent to which a society feels threatenedby uncertain and ambiguous situations and tries toavoid these situations by providing greater careerstability, establishing more formal rules, not toler-ating deviant ideas and behaviors, and believing inabsolute truths and the attainment of expertise'(1980b, 45). The fourth dimension is masculinity(MAS)-femininity (FEM),with MAS defined as 'theextent to which the dominant values in society are"masculine"- that is, assertiveness,the acquisitionof money and things, and not caring for others,the quality of life, or people' (1980b, 46) andFEMdefined as the opposite of MAS.Michael HarrisBond (Chinese Culture Connection, 1987) andlater Hofstede and Bond (1988) developed a fifthdimension, Confuciandynamism (or long-term vsshort-term orientation). Long-term orientationrefers to future-orientedvalues such as persistenceand thrift, whereas short-term orientation refersto past- and present-orientedvalues such as respectfor tradition and fulfilling social obligations.Hofstede's (1980a) work has been criticized for:reducing culture to an overly simplistic four or fivedimension conceptualization; limiting the sampleto a single multinational corporation; failing tocapture the malleability of culture over time; andignoring within-country cultural heterogeneity(Sivakumar and Nakata, 2001). In spite of criticism,researchers have favored this five-dimensionframework because of its clarity, parsimony,and resonance with managers. Yet, even given theproliferation of studies incorporating the frame-work, there have been few attempts to summarizethe empirical findings it has generated. Thus, apressing and practical need in the literature at thistime is a comprehensive review and discussion ofways to improve the use of Hofstede's framework.

Journalof International Business Studies


4/37

A quarter century of Culture's Consequences BradleyL Kirkman t a!l 287

Criteria used to select articles and organizethe reviewWe examine empirical research that assessed anyof the five cultural values published in top-tiermanagement and applied psychology journals(Extejtand Smith, 1990; Gomez-Mejiaand Balkin,1992; Johnson and Podsakoff, 1994; Tahai andMeyer, 1999) and in journals specializing in inter-national management and psychology (seeTable1).We excluded areas such as marketing or financebecause of the traditional delineation of these fields(in business schools and most academic journals)and the need to place limits on our comprehen-siveness. We conducted article title, abstract, andmethodology section searches from January 1980to June 2002. We firstconducted computer-assistedsearches (e.g., ABI-Inform, Ebsco Host AcademicFull-TextElite) when full text was available; other-wise, each journal was physically searched. Wethen conducted computer-assisted keywordsearches within journals using variations on Hof-stede's terms (e.g., COL,collectivists, collectivistic).We included articles only if the authors empiricallyassessedthe cultural values using either primaryorsecondary data. Primarydata include research thatassessed values through surveys, experiments, orother direct methods. Secondary data includeresearch that used Hofstede's country scores to,for example, create cultural distance measures(Kogut and Singh, 1988). We also included a fewstudies that used country scores to assign culturalvalues to individuals (e.g., Bochner and Hesketh,1994; Palich et al., 1995). A detailed analysis of theshortcomings of this particularmethod appears inour discussion. We mention these limitations hereto warn readersabout the variance in the quality ofmethodologies used in the studies we review.Previous reviews of the IND-COL literature havediscussed the difficulty of comparing studiesthat used a wide variety of different measuresand methods to operationalize cultural values(Earleyand Gibson, 1998; Oyserman et al., 2002).Thus, comparing means directly across studies maybe problematic. Space prohibits a critique ofmethodology for each of the studies, but readersshould note that authors took different degreesof care to control for methodological concerns suchas response bias or halo effects. Thus, as we notethroughout, caution should be exercised in inter-preting the findings.2To organize the review, we used a two-tierclassification scheme. The first tier pertained tothe role of cultural values in the relationships

Table 1 Journals searched, with corresponding number ofarticles oundJournalname NumberAcademyof ManagementJournal 22Administrative cienceQuarterly 7Advances n GlobalLeadership 1Advances n InternationalComparativeManagement 1AnnualReviewof Psychology 0BritishJournalof Psychology 0Britishournalof SocialPsychology 2European ournalof SocialPsychology 1Groupand OrganizationManagement 1HumanRelations 2InternationalJournalof Commerceand Management 2InternationalJournalof Comparative ociology 0InternationalJournalof ConflictManagement 1International ournalof Intercultural elations 5InternationalJournalof OrganizationalAnalysis 2InternationalJournalof Psychology 2InternationalStudiesof Managementand Organization 1Journalof AppliedBehavioralScience 1Journalof AppliedPsychology 7Journalof AppliedSocialPsychology 3Journalof Cross-Culturalsychology 16Journalof Experimental ocialPsychology 1Journalof InternationalBusinessStudies 41Journalof InternationalManagement 5Journalof Management 9Journalof ManagementStudies 0Journalof OrganizationalBehavior 6Journalof Personality 0Journalof Personality nd SocialPsychology 11Journalof Research n Personality 1Journalof WorldBusiness 1LeadershipQuarterly 1ManagementInternationalReview 6ManagementScience 1MultinationalBusinessReview 2OrganizationScience 0OrganizationalBehaviorand Human DecisionProcesses 4PersonnelPsychology 2PsychologicalBulletin 0PsychologicalReview 0SmallGroupResearch 1StrategicManagementJournal 11Total 180

investigated. Research on culture has examinedmain associations between values and outcomes, aswell as cultural values as moderators.Main effectstudies have been labeled 'Type I' and moderatorstudies 'Type II' (Lytle et al., 1995). Our secondclassification was by level of analysis, whetherindividual, group/organizational, or country. In



5/37

A quarter century of Culture'sConsequences BradleyLKirkmant a288

spite of Hofstede's (1980a, 2001) argument againstusing his dimensions for purposes other thancountry level studies, we found that a majority ofresearchers had adapted them for use at theindividual or group/organization levels. All of theselevels arevalid, depending on the researchquestionand on whether there is more commonality within,than between groups (Sivakumar and Nakata,2001). Thus, we argue that excluding these studieswould create an incomplete picture of Hofstede'simpact. In addition, analyses of the same data atdifferent levels of analysis do not necessarilylead toequivalent findings or replication of dimensions(Leung and Bond, 1989; Hofstede et al., 1993,2001). To accuratelysort the studies into the properlevel of analysis, we scrutinized each study's theoryand method section to determine the level at whichthe study was conceptualized and analyzed, respec-tively. The level at which the data were collectedwas not always as informative. For example,studies using individual level survey data could beclassified as individual, group/organizational, orcountry level. For individual level studies, datamust have been collected and analyzed at theindividual level of analysis and tied to individuallevel outcomes. Forgroup/organizational, the datahad to be aggregated to the group or organiza-tional level and linked to group/organizationaloutcomes. For country level studies, either indivi-dual level data had to be aggregatedby country andlinked to country level outcomes or pre-existingcountry level measures (such as Hofstede's, 1980acountry level scores)had to be used. Table2 depictsour 2 x 3 classification with the correspondingnumber of articles reviewed within each of the sixcells. To ease interpretation, we further organizefindings by topic using subject matter headingsfrom business and psychology. To facilitatereader comparisons of topics at respective levels,Table3 shows the topics on the vertical axis and thelevel of analysis on the horizontal. Also, Supple-mentary Appendices A and B contain detailedarticle summaries on culture as main and mod-erator effects, respectively, for all 180 articles inour review.3

Research incorporating Hofstede's culturalvalue dimensionsWe first review culture as a main effect (i.e., Type Istudies) at the individual level of analysis, and thenat the group/organization and country levels. Wethen review findings that incorporate culture asa moderator (i.e., Type II studies) at these levelsof analysis. When a study included multiple levelsor type of effect, we discuss each finding in theappropriatesection and note cross-listings.Type I studies of culture at the individual levelof analysisOur review uncovered two basic types of Type Iindividual level study: cross-cultural and mono-cultural. In both types of study, researchers typi-cally examine relationships between individuals'cultural values and various outcomes; however,in cross-culturalstudies two or more countries arenormally included, whereas in mono-cultural stu-dies all individuals emanate from the same country.Researchers have shown that there is plentyof within-country variation on cultural values(Hofstede, 1980a; Au, 1999). Clearly, 'people varyon pivotal psychological dimensions (e.g., PDbeliefs, traditionality) both on a between-countrybasis and on a within-country basis' (Brockner,2005: 355). Thus, even though mono-culturalstudies may not be viewed as technically cross-cultural, we still include those studies empiricallyassessing cultural values in only a single countrybecause of the theoretical importance of under-standing within country cultural variation.Cultural values were associated with outcomes inmanagement and applied psychology domains,including: change management; conflict manage-ment; decision-making; human resource manage-ment (HRM); eadership;organizational citizenshipbehavior (OCB);work-relatedattitudes;negotiationbehavior; reward allocation; and individual beha-vior relating to group processes and personality.Change management (4)3Collectivists in the US showed more positive groupattitudes toward a new technology and had better

Table 2 Classification cheme used for literature eviewand numberof articles ncludedIndividualevel Group/organizationevel Country evel Total

Cultureas a main effect 64 6 78 148Cultureas a moderator 23 5 4 32Total 87 11 82 180Note: fa studywaslisted n more han one section, twas countedonlyonce in the section n which tfirstappeared.



6/37

A quarter century of Culture'sConsequences BradleyKirkmantal 289

Table 3 Research ubject matter by level of analysisIndividual Group/organization Country

Change management James(1993)main,Geletkanycz(1997)main, Anakwe et al.(1999)main, Eby et al. (2000)main

Conflict management Leung (1987)main, Leung Elron (1997)main, Smith et al. (1998)main(1 988)main, Cocroft and Oeztzel (1998)mainTing-Toomey (1994)mainGabrielidis et al. (1997)main

Decision-making Ali (1993)main, Nooteboom et al.(1997)main, Mitchell et al.(2000)main, Steensma et al.(2000a)main

Human resource Ozawa et al. (1996)main, Cable and Newman and Nollen (1996)mainmanagement judge (1994)main, Ramamoorthy Roth and O'Donnell (1996)main,and Carroll(1998)main, Earley Schuler and Rogovsky(1998)main,

(1986)main, Earley et al. (1999)main Ryan et al. (1999)main

Leadership Casimir and Keats (1996)main, Pillai and Meindl (1998)main Shenkar and Zeira (1992)mainPillaiand Meindl (1998)main, Offermanand HellmannHelgstrand and Stuhlmacher (1997)main, House et al. (1999)main(1999)main, Jung and Avolio(1999)mod, Chan andDrasgow (2001)main

OCB Moorman and Blakely (1995)man,Van Dyne et al. (2000)main

Work-relatedattitudes Bochner and Hesketh (1994)main, Schneider and DeMeyerPalich et al. (1995)mod, Bennett (1991)main, Weber et al.(1999)main, Chiu (1999)main, (1996)main, Peterson et al.Clugston et al. (2000)main, Feldman (1995)main, Van de Vliert and Vanand Bolino (2000)main, Lee et al. Yperen (1996)main, Peterson and(2000a)main, Lee et al. (2000b)mod, Smith (1997)main, Robie et al.Martella and Maass (2000)mod, (1998)mod, Gong et al. (2001)main,Schaubroecket al. (2000)mod, Spector et al. (2001 b)main, SpectorVandenberghe et al. (2001)mod, et al. (2002)modHarpaz et al. (2002)main,Thomas andAu (2002)main' mod

Negotiation Arunachalam et al. (1998)main,Tinsley and Pillutla (1998)man,Brett and Okumura (1998)man,Pearson and Stephan (1998)mainMorris et al. (1998)main, Gelfandand Realo (1999)mod, Probst et al.(1999)main, Ng and Van Dyne(2001)mod, Tinsley (2001)mainTinsley and Brett (2001)main,Wade-Benzoni et al. (2002)main

journalof International usinessStudies


7/37

"Aquarter century of Culture's Consequences BradleyLKirkmant al290

Table 3 ContinuedIndividual Group/organization Country

Reward allocation Mann et al. (1985)ma"', Leung andIwawaki (1988)main, Hui et al.(1991)main, Tower et al. (1997)mainChen et al. (1997)main, C.C. Chenet al. (1998a, b)main,McLean-Parkset al. (1999)main, Gomez et al.(2000)main

Behavior relating to Triandis et al. (1988)main, Earley Chatman and Barsade (1995)mod, Krug and Nigh (1998)maingroup processes and (1989)mod, Huiand Villareal Chatman et al. (1998)mod, Oyserman et al. (2002)mainpersonality (1989)main, Earley (1993)mod, Cox et al. (1991)mainOyserman (1993)main, Wagner Erez and Somech (1996)mod(1995)mod, Yamaguchi et al. Eby and Dobbins (1997)main(1995)main, Eby and Dobbins Earley (1999)mod(1997)main, Kwan et al. Gibson (1999)mod(1997)main, Wojciszke (1997)main, Kirkman and Shapiro (2001 b)mainChen et al. (1998a, b)mod, Lam et al. (2002a)modOyserman et al. (1998)main, Grimmet al. (1999)main, Tafarodi et al.(1999)main, Thomas (1999)main,irkman and Shapiro (2000)main,Satterwhite et al. (2000)main,Carpenterand Radhakrishnan(2000)main, Gibson and Zellmer-Bruhn (2001)main, Kirkman andShapiro (2001 a)main

Entrepreneurship Morriset al. (1993)main, Thomas and Mueller(2000)mainMorris et al. (1994)mainSocial networks Zaheer and Zaheer

(1997)main,anev and Stevenson (2001)mainEntrymodes Kogut and Singh (1988)mainErramilli (1991)main, Kim and

Hwang (1992)main, Shane(1992)main, Shane (1994)mainErramilli (1996)mainPadmanabhan and Cho(1996)main, Pan (1996)mainPan and Tse (1996)mainAnand and Delios (1997)mainErramilliet al. (1997)mainBarkemaand Vermeulen(1998)main, Hennart and Larimo(1998)main, Pan and Tse(2000)main, Arora and Fosfuri(2000)main, Brouthers andBrouthers 2000)main, Brouthersand Brouthers (2001)main, Changand Rosenzweig (2001)mainHarzing (2002)main, Erramilli et al.(2002)mod, Pan (2002)main,Nachum (2003)main

Journal of International Business Studies


8/37

A quarter century of Culture's Consequences BradleyLKirkmant al 291

Table 3 ContinuedIndividual Group/organization Country

Foreigndirect Benito and Gripsrud 1992)main,investment Liand Guisinger (1992)mainLoree and Guisinger (1995)main,Kallunki et al. (2001)main, Thomasand Grosse (2001)main,Habib andZurawicki (2002)main

Jointventure Liand Guisinger (1991)main, Dattacharacteristics and and Puia (1995)main, Barkema et al.performance (1996)main, Barkema et al.

(1997)main, Barkema andVermeulen (1997)mainGomez-Mejiaand Palich(1997)main, Park and Ungson(1997)main, Morosini et al.(1998)main, Glaister and Buckley(1999)main, Luo and Peng(1999)main, Merchant andSchendel (2000)main, Hakansonand Nobel (2001)main, Luo(2001 a)main, Luo(2001 b)man,Luo and Park (2001)main, Luo et al.(2001)mod, Reuer 2001)mainLuo (2002)mann, Pothukuchi et al.(2002)main

Alliance formation Dickson and Weaver (1997)mod Kashlak et al. (1998)mainSteensma et al. (2000b)mainmod

Innovationand Shane (1995)main, Shane et al.research and (1995)main, Jones and Teegendevelopment (2001)main, Richards and De Carolis

(2003)mainSocietal outcomes Franke t al. (1991)main,Dienerand(e.g., wealth, national Diener (1995)mainmod, Diener et al.accounting systems, (1995)main, Diener et al. (2000)modnumber of intellectual Salterand Niswander (1995)mainpropertyviolations Riahi-Belkaoui1998)main,Husted

(1999)main, Ronkainen andGuerrero-Cusumano (2001)mainMotivation Erez and Earley (1987)mod,Dorfman and Howell (1988)mod,

Earley (1994)mod, Eylon and Au(1999)mod, Lam et al. (2002a)mod

Organizational justice Lind et al. (1997)mod, Au et al.(2002)mod, Brockner et al.(2000)mod, Brockner et al.(2001)mod, Lam et al. (2002b)mod

main, main effect study; mod, moderating effect study.retention when a technology presentation con-tained a group-relevancetheme (the same was truefor individualists when the presentation containeda strong self-relevance theme) (James, 1993). COLwas negatively related to receptivity to distance

learning technology (Anakwe et al., 1999), andwas positively related to perceptions of an organi-zation's readiness to change to team-based selling(Eby et al., 2000) in the US. Using Hofstede's(1980a) country scores to assign cultural values to

journalof International usiness tudies


9/37

SA uarter century of Culture'sConsequences BradleyKirkmantal292

managers in 20 countries, Geletkanycz (1997)found that managers were more likely to believethat the future chief executive officer (CEO) oftheir company should have the same expertise asthe current CEO, and that the company's futurecorporate strategy should be the same as theexisting corporate strategy (i.e., essentially demon-strating acceptance of the status quo) when theywere high on IND, and low on UA, PD, and long-term orientation.Conflictmanagement (4)Hong Kong (HK) subjects (who were significantlyhigher in COL)preferredbargainingand mediationmore than US subjects in a scenario-based experi-ment with undergraduateand non-student subjects(Leung, 1987). In both countries, subjects' proce-dure preferences were related to the extent towhich the procedurewas perceived as favorable tothe disputants, fair,capable of animosity reduction,and allowing process control. Leung (1988) foundthat HKsubjects were more likely to sue a strangerthan were USsubjects;and COLwas associated withhigher likelihood of suing when the dispute wasbetween strangers (but not friends). US students(who scored significantly lower on both IND andCOL; see our discussion section for debate onIND-COL as unipolar or bipolar at the individuallevel) used more antisocial, self-attribution, hint,and self-presentation strategies than Japanese sub-jects, who used more indirect face strategies(Cocroft and Ting-Toomey, 1994). Finally,Mexicanstudent subjects (who were significantly higher onCOLand IND and lower on FEM han US subjects)showed more concern for others' outcomes inconflict resolution than US students (Gabrielidiset al., 1997). More specifically, Mexican scores onaccommodation and collaboration were higherthan those in the US.Decision-making 4)IND was positively related to consultative, partici-pative, and autocratic decision-making styles andto attitudes toward risk in Saudi Arabian managers(Ali, 1993). Nooteboom et al. (1997) found that UAwas negatively related to the probability of apotential loss by transaction partners, but unrelatedto the size of loss among 97 firm-supplier alliancesin the Netherlands. Mitchell et al. (2000), usingHofstede's (1980a) country scores at the individuallevel and assessing business professionals' cognitivescripts, found that IND and PD were positivelyrelated to 'ability' scripts (i.e., given resources, the

person is able to carry out a goal); IND waspositively related to 'willingness' scripts (i.e., givenresources, the person will want to carryout a goal);and the relationship between 'arrangements'scripts(i.e., access to required materials) and starting anew business was stronger for individualists thanfor collectivists in the US, Mexico, China, Japan,and Chile. Using country scores, entrepreneursfrom seven countries (Australia, Finland, Greece,Indonesia, Mexico, Norway, and Sweden) viewedcooperative strategies with other firms as moreacceptable when they were lower in IND and MAS,but higher in UA (Steensma et al., 2000a). Thosehigher, ratherthan lower, in IND and UA preferredcontractual safeguardsmore, and perceived partnercommonality was positively linked to UA.HRM5)US undergraduates,who were significantly higheron IND than Japanese (using Singelis et al.'s (1995)self-construal measure), perceived a change to anaffirmative action program less favorably and fairthan Japanese subjects (Ozawa et al., 1996). COLwas negatively relatedto preferences for individual-based pay in the US (Cableand Judge, 1994) and tothe use of selection tests, formal appraisalpractices,and desire for promotions based on merit, andpositively related to preference for equality-basedrewardsand employment security among US busi-ness undergraduates (Ramamoorthy and Carroll,1998). Both COLand PD interacted with feedbacktype such that US subjects' performance increasedas a result of both positive and negative feedback,but in England (where COL and IND were sig-nificantly higher), only positive feedback resultedin performanceincreases(Earley,1986). In a secondsample, the importance of feedback received andtrust in supervisor partially mediated the maineffects of praise, criticism, and culture on perfor-mance. In a sample of managers completing amanagerial simulation in the US, Czech Republicand PRC(US subjects were significantly higher onIND than either Czech or PRCsubjectswho did notdiffer), Earley et al. (1999) found that individualfeedback played a role for both individualists andcollectivists, whereas group feedback was criticalonly for collectivists.Leadership (4)In a sample of Anglo- and Chinese-Australianmanagers, Casimir and Keats (1996) assessed pre-ferences for leadership styles from among fourchoices (i.e., created by crossing the extent to



10/37


which a leader is high or low on both concern forgroup performance and maintenance of grouprelations). Both cultures preferred leaders whoexpressed high concern for both performance andgroup relations, and did not differ significantly onIND-COL. In high-stress work environments, bothcultures maintained their preference for high-concern leaders; however, in low-stress environ-ments, Chinese respondents preferreda leaderwhoshowed concern for group relations equally toleaders high on both types of concern. In a secondstudy, COL was positively related to the level ofcharismatic leadership, which in turn was posi-tively related to supervisory ratings of work unitperformance, job satisfaction, satisfaction with theleader, and leader effectiveness in over 100 workunits of a US firm (Pillai and Meindl, 1998).Helgstrand and Stuhlmacher (1999) found thatboth Danish and US high school and undergradu-ate students rated leaders who were feminine andindividualistic as most effective (Danes were loweron PD). Finally, using horizontal (H) and vertical(V)aspects of IND and COL(i.e., VCOLrepresents atendency to view the self as an aspect of the group,see members of an ingroup as differentfrom theself, and accept inequality;HCOL represents thetendency to view the self as an aspect of the group,see members of an ingroup as similar o the self, andvalue equality;VINDrepresents the extent to whichan individual's self-concept is autonomous butexpects inequality;and HIND represents the extentto which an individual's self-concept is autono-mous and the individual is seen as equalto others),Chan and Drasgow (2001) surveyed militaryrecruits and junior college students in Singaporeand the US to explore links between culturalvalues and various dimensions of motivation tolead (MTL).They found that HIND was negativelyrelated to noncalculative MTL (i.e., people leadonly when they are not calculative about the costsof leading relative to the benefits) and social-normative MTL(i.e., people lead because they feela sense of responsibility or duty); and VIND waspositively related to affective identity MTL (i.e.,some people just like to lead others) and social-normative MTL, and negatively related to noncal-culative MTL.OCB2)In a sample of US financial services employees, COL(assessed using the values, norms, and beliefsdimensions of Wagner and Moch (1986)) waspositively related to several dimensions of OCB

after controlling for procedural justice (Moormanand Blakely, 1995). In a sample of cooperativehousing residents in the US, COL was positivelyrelatedto OCB,with organization-basedself-esteemfully mediating the relationship (Van Dyne et al.,2000).Work-related ttitudes (8)Bochner and Hesketh (1994) surveyed Australianbank employees representing 28 different nation-alities, assigned each a country score for IND-COLand PD, and placed them into high and low groupson the two values. Collectivists reported havingmore informal contact with fellow workers, knewstaffbetter, and were more likely to work on a teamthan alone compared with individualists. Thosehigh, rather than low, in PD were less open withtheir superiors, had more contact with them,described their supervision as being more closeand direct, were more task-oriented, and hadgreater beliefs in Theory X (i.e., a managementstyle favoring centralized decision-making, tightcontrol, and hierarchy). Bennett (1999), after firstconfirming that a USsample was significantly loweron COL and higher on MASthan a PRC sample,found that COLwas positively related to favorableattitudes towards group activities and cooperationin both countries. MAS was negatively relatedto attitudes towards human development, but onlyin the US sample.Chiu (1999) found that individualists scoredhigher on positive affect and job satisfaction andlower on work strain than collectivists in a sampleof Singaporean and HK nurses, who were signifi-cantly higher on COL and lower on IND thanAustralian and US nurses. Clugston et al. (2000)assessed the relationships among Hofstede'sfour original cultural values and three bases (i.e.,affective, continuance, and normative) and foci(i.e., organization, supervisor, and workgroup) ofcommitment using surveys in a US publicagency. COL was positively related to affectivecommitment to supervisors and the workgroup,continuance commitment to the workgroup, andnormative commitment to all foci; and PD waspositively related to affective commitment to theorganization and both continuance and normativecommitment to all foci. Lee et al. (2000a) foundthat individualists were more attuned toward apromotion focus (i.e., the pursuit of gains andaspiration toward ideals), whereas collectivists weremore attuned toward a prevention focus (i.e., theavoidance of losses, and the fulfillment of obliga-



11/37

A quarter century of Culture's Consequences BradleyLKirkmant al294

tions) in a sample of US undergraduate subjectswho were significantly higher on IND (i.e., inde-pendent self-construal) and lower on COL (i.e.,interdependent self-construal)than HKundergrad-uates. Harpaz et al. (2002) found that, over time,new work entrants from low UA countries (basedon Hofstede's (1980a) country scores assigned toindividuals) had increased work centrality. In astudy of behavioral responses to job dissatisfactionin a sample of executive education participants inNew Zealand and HK(New Zealanders were higherin HINDand VCOL han those in HK),Thomas andAu (2002) found that HIND was positively relatedto voice and that VCOLwas positively related toneglect and negatively related to loyalty. Finally,cultural distance (based on the degree of differencebetween an individual and others in their contextusing country scores) was unrelated to opportu-nities to learn or further develop skills, internshipsatisfaction, or organizational commitment amonga sample of US MBA students in overseas intern-ships (Feldman and Bolino, 2000).Negotiation (9)HKnegotiators (who scored significantly higher onCOL and lower on IND than US negotiators)obtained higher joint outcomes than those in theUS in a two-party negotiation experiment (Aruna-chalam et al., 1998). Negotiators with a high, ratherthan a low, best alternative to a negotiated agree-ment (BATNA)obtained larger outcomes in bothHK and the US. Mediation was associated withhigher joint outcomes and had a stronger effect inthe US than in HK. In a similar experiment, Tinsleyand Pillutla (1998) found that US subjects scoredsignificantly higher on self-enhancement (i.e., acultural value 'conceptually similar'to IND;p 713)and lower on self-transcendence (i.e., a culturalvalue conceptually similar to COL) than HK sub-jects using Schwartz's(1992) cultural value inven-tory. US subjects rated self-interested and jointproblem-solving as more appropriatethan their HKcounterparts. Further, cooperative instructionswere interpreted by US subjects as meaning theyshould strive for joint gain, whereas HK subjectsinterpreted them as meaning they should strive forequality. Finally, dyad sum was positively related tosatisfaction for US (but not HK) subjects, whereasdyad difference was positively related to satisfac-tion for HK (but not US) subjects.In a series of inter- and intra-cultural dyadexperiments, Brett and Okumura (1998), aftershowing that US negotiators were more individua-

listic but less hierarchical (i.e., lower in PD) thanthe Japanese, found that individualists endorsedself-interest in negotiations, and that negotiatorswith stronger hierarchical values endorsed distri-butive tactics and reported spending significantlymore time discussing power. Brazilians(who scoredsignificantly lower on IND than US subjects)preferred accommodation, collaboration, andwithdrawal negotiating styles more, whereas USsubjects preferred competition in student dyadnegotiation experiments (Pearson and Stephan,1998). US subjects preferred negotiation stylesreflectinga high concern for self, whereasBrazilianspreferred a style reflecting a high concern forothers. Finally, Brazilians made accommodationsand avoided conflict more when the conflict waswith an ingroup, rather than an outgroup, member,whereas US subjects treated ingroup and outgroupmembers similarly. Importantly, results were repli-cated using IND-COLrather than country.In a sample drawn from the US, the People'sRepublicof China (PRC), ndia, and the Philippines(with culture assessedusing Schwartz's 1994) valueinventory), Chinese managers preferred an avoid-ing style, with societal conservatism (i.e., lowopenness to change) fully mediating the effects ofcountry on conflict style; and US managers pre-ferred a competing style of conflict, with self-enhancement (i.e., IND) fully mediating the effectsof country on conflict style (Morriset al., 1998). In asample of US undergraduates, Probst et al. (1999)found type of prisoner'sdilemma (i.e., single-groupvs intergroup) moderated the main effects thatVIND and VCOL had with cooperation such thatvertical individualists were least cooperative in thesingle-group prisoner's dilemma but were morecooperative in the intergroup dilemma (whencooperation with the group maximized personaloutcomes). Vertical collectivists were most coop-erative in the single-group dilemma but less in theintergroup dilemma (when group defectionresulted in maximum group outcomes). Examiningbusinesspeople from Japan, Germany, and the US,Tinsley (2001) showed that IND was positivelyrelated to using interest strategies and negativelyrelated to using power strategies (Japanese respon-dents were significantly lower on IND than Germanor US respondents). Studying business students inHK and the US, Tinsley and Brett (2001) found thatUS students were more self-directed, less hierarch-ical, and less tradition-bound than HK students; USstudents placed greater emphasis on discussinginterests and synthesizing multiple interests than



12/37

A quarter century of Culture's Consequences BradleyLKirkmant al 295

did HK students; and American dyads were sig-nificantly more likely to reach an integrative out-come than were HK dyads. Finally, Wade-Benzoniet al. (2002) found that US MBAs (who weresignificantly higher on IND than Japanese under-graduates) were less cooperative, reached fewerequal solutions, and expected others to be lesscooperative than the Japanese.Reward allocation (8)AssessingCOL n both Japanand the US,LeungandIwawaki(1988) found that the more collectivistic asubject, the more he or she followed the equalityrule (everyone receives the same reward) and theless he or she followed the equity rule (rewardsallocated based on effort). Contrary to expecta-tions, Japanese and US subjects did not signifi-cantly differ on COL.After including South Koreain the dataset, for low input subjects (who partici-pated infrequently), there were no differencesacross countries in the use of the equity norm,nor did they allocate more equally with friends (i.e.,ingroups) and more equitably with strangers (i.e.,outgroups). Generous allocators (i.e., low-inputallocators using equity rules and high-input alloca-tors using equality rules) were better liked (in allthree countries) and rated as fairer than allocatorswho were less generous (findings held in Japanandthe US, but not in South Korea).Hui et al. (1991)found that compared with US subjects, whenresources were plentiful, HK subjects (who scoredsignificantly higher on COL)put a greater emphasison equal allocation of rewards; when the rewardwas fixed, HK subjects were more generous andtreated close friends, compared with co-workers,more generously, whereas US subjects made nosuch distinction.

Primary school Japanese subjects (who scoredsignificantly higher on COL) were more likely tofollow 'equal-say' rules than Australian subjects,who tended to follow more 'self-interest' rules in agame experiment (Mann et al., 1985). In a study ofBritish and Russian students (Tower et al., 1997),when British students (who scored significantlylower on COL) were low performers or allocatingrewards with a co-worker who was a friend, theyallocated more to themselves; and there were noallocation differences for the British when perfor-mance was high or when the co-worker was astranger. In contrast, when Russian students werehigh performers or allocating rewards to a co-worker who was a stranger, they allocated morereward to themselves; and there were no allocation

differences for the Russianswhen performancewaslow or the co-workerwas a friend.VCOLwas positively related to rewardallocationreform in an in-basket exercise of managers in thePRC (Chen et al., 1997). In contrast, HCOL wasmarginally and negatively related to rewardalloca-tion reform. An interaction effect for VCOL andHCOL ndicated that the negative relation betweenHCOLand support for reformholds only under thecondition of low VCOL.VCOLwas also negativelyrelated to egalitarian allocation preferences,whereasHCOLwas negatively related to differentialallocation preferences. In a second study, Chenet al. (1998a) asked undergraduate students in HKand the US to make reward allocation decisionsafterreading a case-vignette. Rewardallocation wasmore differential when task interdependence waslow rather than high, and when the goal wasproductivity or fairness rather than solidarity inboth countries. In the high interdependence situa-tion, achievement motivation was negativelyrelated to the differential allocation, but when itwas low, achievement motivation was negativelyrelated to differential allocation only for the HKsubjects,but positively related for US subjects. COLwas negatively related to differential allocation inHK but not the US. Individualistic MBA students inSingapore and the US took slightly more time torecover, and less time to allocate, resources thancollectivists (McLean-Parkset al., 1999). In bothcultures, equity rules were used more when dis-tributing a resource, and equality when recoveringresources. After showing that Mexican MBAsweresignificantly more collectivistic than US students,Gomez et al. (2000) found that collectivists gavemore generous evaluations to ingroup, ratherthanoutgroup, members in a teamwork scenario-basedexperiment. Individualists valued task inputs indetermining evaluations more so than did collecti-vists, whereas both valued equity-based rewards.Individualbehavior relating to group processesandpersonality (16)As some of these studies are only tangentiallyrelated to organizational contexts and are moresocial psychological in focus (e.g., Hui and Villar-eal, 1989; Oyserman, 1993; Yamaguchi et al., 1995;Kwan et al., 1997; Wojciszke, 1997; Oyserman et al.,1998; Grimm et al., 1999; Tafarodi et al., 1999;Carpenter and Radhakrishnan, 2000; Satterwhiteet al., 2000), we highlight only those that have clearlinks to work-related outcomes. For example,Triandis et al. (1988) conducted three studies to



13/37

A quarter century of Culture'sConsequences BradleyKirkmantal296

examine the relationship between IND and COLand self-ingroup relations. Surveys responses fromUS undergraduates were factor-analyzed, and amultifaceted conceptualization of IND emerged.US respondents viewed competition as occurringbetween individuals, whereas Puerto Rican respon-dents viewed competition as occurring morebetween ingroups and outgroups, rather thanwithin ingroups.COL was positively related to positive assess-ments of group processes in a study of under-graduates from 14 countries studying in NewZealand (Thomas, 1999). Cultural distance onCOL (i.e., how different, on average, each indivi-dual is from other group members) was negativelyrelated to group receptiveness. In a sample of USundergraduates,COLwas positively related to self-efficacy for teamwork,need for social approval,andpositive past experience working in teams (EbyandDobbins, 1997). In support of their theoreticalmodel (Kirkmanand Shapiro, 1997), KirkmanandShapiro (2000) found that COL was positivelyrelated to receptivity to team-based rewardsusingemployee surveys in a US insurance company.Using surveys from self-managing work team(SMWT)members in Belgium, Finland, the Philip-pines and the US,COLwas also positively related toteam members' job satisfaction and organizationalcommitment; and resistance to teams mediated therelationships between COL and both satisfactionand commitment, whereas resistance to self-man-agement partially mediated the negative relation-ship between PD and commitment (KirkmanandShapiro, 2001a). Gibson and Zellmer-Bruhn(2001)investigated teamwork metaphor use in interviewsin France,Puerto Rico, the Philippines and the USin five multinational firms. PD was negativelyrelated to using metaphors containing clear rolecontent (e.g., family or military metaphors), andIND was negatively related to metaphors broad inscope (e.g., community metaphor).ResearchaccomplishmentsWhereas Hofstede (1980a) was clear that hisconceptualization and operationalization of cultur-al values was intended only for the country level,researchers have liberally adapted them for Type Istudies at the individual level. Adaptation in thismanner has both strengths and weaknesses. On thepositive side, it has provided a new way to consider,describe and measure culture (Bond, 2002). Studyvariance that might have been attributed to othervariables or explained with post hoc analyses has

been used to derive and test theories about thecultural antecedents to individual outcomes inorganizations. The frameworkdemonstrates multi-method utility at the individual level of analysis,with Type I studies utilizing a variety of methodsincluding experimental, managerial simulation,survey research, in-basket exercises, and scenarios.On the downside, there are clearly disconnectsbetween the theoretical and methodological under-pinning of Hofstede's (1980a) conceptualizationand that of researchers working at the individuallevel. For example, findings that are incongruentwith Hofstede (1980a), such as the studies that haveshown that people in one country can be moreindividualistic and collectivistic, on average, thanpeople from another country (Oyserman, 1993;Cocroft and Ting-Toomey, 1994; Gabrielidiset al.,1997; Oyserman et al., 2002), cannot be used tochallenge his findings because of the conceptualdifferencesinherent in the two levels of analysis (orperhaps methodological artifacts such as acquies-cence bias). As we show in our general discussion,different empirical findings across levels (some-times involving the same cultural values and out-comes) underscore the important theoreticaldifferences between levels.ResearchchallengesOf the 64 TypeI studies at the individual level, only12 included cultural values other than IND-COL,despite the fact that individuals are affected by acomplex set of cultural values (Lytle et al., 1995;Kirkmanand Shapiro, 1997). Indeed, all 12 studiesthat included cultural values in addition to (orbesides) IND-COLfound significant effects. Of thefive studies that included both IND-COL and othercultural values simultaneously(e.g., Earley, 1986;Clugston et al., 2000; Mitchell et al., 2000; Kirkmanand Shapiro, 2001a; Harpaz et al., 2002), allexplained unique variance beyondIND-COL. Suchconsistent findings suggest that including culturalvalues other than IND-COLin the other 52 Type Istudies would have led to important insights.Moreover, in the seven Type I studies that exam-ined relationships involving IND-COL and countryof origin simultaneously, five of the seven (e.g., Huiet al., 1991; Brett and Okumura, 1998; Gomez et al.,2000; Kirkman and Shapiro, 2001a; Tinsley andBrett, 2001) showed that country explained uniquevariance beyond IND-COL (only Tafarodi et al.(1999) and Tinsley (2001) showed full mediationeffects). This 'hidden' variance could be explainedby other cultural values besides IND-COL (Brett



14/37

A quarter century of Culture's Consequences BradleyL Kirkman t al 297

et al., 1997). Incorporating different Hofstede-inspired values (e.g., including UA with PD) orcompeting and complementary conceptualizations,such as Trompenaars' (1993) concept of universal-ism-particularism, which refers to the relativesalience of rules (universals) or exceptions (parti-culars), is warranted. Reward allocation decisionscould be affected by the extent to which subjectsapply rules differently across recipients. Althoughnot specifically discussed above, Kirkman andShapiro (2001a) found that Kluckhohn andStrodtbeck's (1961) cultural value dimensions ofdoing-being orientation and free will-determinismexplained unique variance beyond IND-COLand PD.Further, even though researchers see clearerconnections between IND-COL (compared withother values) and outcomes such as rewardalloca-tion, conflict, negotiation, and individual behaviorin groups (Oyserman et al., 2002), the theoreticalrationale is too limited. For example, in rewardallocation and negotiation studies, the key role ofstatus of the allocator, reward recipient, or nego-tiators (e.g., Leung and Lind, 1986; Leung, 1997)makes PD important, yet it is rarely assessed.Indeed, Fischer and Smith's (2003) meta-analysisshowed that PD accounted for more significantcross-culturaldifferences in rewardallocation thanIND-COL (i.e., higher PD cultures allocate rewardsusing equity, rather than equality, rules). Thusincluding alternative cultural values is clearlywarranted, and future reward allocation studiesmust at a minimum include PD. Unfortunately,when researchers include cultural values besidesIND-COL, they tend to fall back on Hofstede's(1980a) country scores rather than direct, indivi-dual level measures (e.g., Bochner and Hesketh,1994; Mitchell et al., 2000; Steensma et al., 2000a;Harpaz et al., 2002). We debate the validity ofprimaryvs secondary data in our discussion. Thus,much remains to be done on the direct measure-ment of PD, UA, MAS-FEM, and Confuciandynamism at the individual level (e.g., see the TypeII individual level section for examples of thedirect measurement of PD), and we recommendincorporating other value dimensions into thislevel of analysis.Our review also demonstrated the importance offine-grained models when conducting Type Istudies at the individual level. Again, the rewardallocation studies are useful for illustration.Although some researchers found associationsbetween IND-COL and allocator preferences for

equal or equitable distributions (Mann et al., 1985;Leung and Iwawaki, 1988; Hui et al., 1991), otherstudies found no relationship (McLean-Parkst al.,1999), or found that all subjects preferred equitydistributions (Gomez et al., 2000). Includingboundary conditions or moderating variables isthe key to resolving such conflicting findings. Forexample, Toweret al. (1997) found that BritishandRussian allocators differed in their decision ruleonly when allocator performance was low ratherthan high. Similarly, C.C. Chen et al. (1998a,b)found that achievement motivation was signifi-cantly negatively related to differential allocationin both HK and the US when task interdependencewas high; however, when task interdependence waslow, the opposite was true for the US. Similarly,aseries of studies showed that the ingroup-outgroupdistinction played an important role for collecti-vists compared with individualists (Hui et al.,1991; Tower et al., 1997; Gomez et al., 2000).Such findings show that relationships involvingIND-COL(and, by extension, other culturalvalues)exist only when certain contextual conditions arepresent. Leung (1997) drew a similar conclusionwhen reviewing several reward allocation studies,arguing that interactional goals and situationalvariables (i.e., the relationship between allocatorsand recipients and the role assumed by theallocator) interact with culture to affect the alloca-tion rule adopted. Researchers should take theseconditions into account to develop better-specifiedtests of existing theory involving moderators ormediators to explain the contingency relationshipsof cultural values at individual level.Type I studies of culture at the group/organization level of analysisType I studies at the group/organization level arefewer in number and more recent than individuallevel studies. Oetzel (1998) found that individua-listic European-American groups had a greaternumber of conflicts, fewer cooperating tactics,and more competing tactics than collectivisticJapanese-American groups. In two studies alsodiscussed in the individual level section, Pillai andMeindl (1998) found that COL was positivelyrelated to charismatic leadership emergence in afield study of 101 work groups; and Eby andDobbins (1997) found that team collectivisticorientation was positively related to team coopera-tion, and team cooperation mediated the relation-ship between team collectivistic orientation andteam performance in a study of 33 student teams.



15/37

A quarter century of Culture'sConsequences BradleyKirkmantal298

Kirkman and Shapiro (2001b) found that COLwasnegatively related to the level of team memberresistance to the team-related aspects of SMWTs,and that resistancemediated the relationships COLhad with team cooperation, empowerment, andproductivity.Elron (1997) assigned country scores (Hofstede,1980a) to top management team (TMT)members toassess cultural heterogeneity in 121 subsidiaries in34 countries. TMT cultural heterogeneity waspositively relatedto TMTperformanceand conflict;however, conflict was negatively related to TMTperformance. Culturalheterogeneity on both IND-COL and MAS-FEMwas positively related to TMTperformance; and heterogeneity in UA was posi-tively related to conflict. Cox et al. (1991) assessedwhether Asians, Blacks, and Hispanics were morecollectivistic than Anglos based on the number ofcooperative choices individuals made prior togroup discussion. Ethnically diverse groups (i.e.,those that were high, on average, in COL)behavedmore cooperatively than all-Anglo groups (thosethat were low, on average, in COL),and that thesedifferences tended to increase when situationalcues favored cooperation. Of concern was the lowreliability of their survey-based COL measure (amixture of items from Triandis et al., 1986; Hui,1988). Although Hofstede has offereda different setof dimensions for organizational, rather thannational, culture (see Hofstede et al., 1990), twostudies adapted his national culture frameworkforuse at the organizational level. Morris et al. (1993)examined 84 manufacturing firms in the US andfound a curvilinearrelationship between IND-COLand entrepreneurial behavior such that at highlevels of either IND or COL entrepreneurialbeha-vior suffers.Morris et al. (1994) replicatedthe abovefindings using firms in South Africa, but theirfindings did not replicate in Portugal (possiblybecause of relatively high COLin Portugal).ResearchaccomplishmentsEven though Hofstede (1980a) conceptualized andoperationalized the value dimensions at thenational level, it is not surprising that researchershave adapted the values for group and organizationlevel studies as national cultures are 'groups,' andone might reasonably infer that cultural values areequally applicable to smaller groups such as teamsand organizations. To justify aggregation, research-ers commonly demonstrate more variation betweengroups than within using an analysis of variance(ANOVA) procedure, intraclass correlation coeffi-

cient (ICC1)or within-and-between-analysis. Typi-cally, there is evidence of high interrateragreementassessed using tests of within-group reliability (rwg)or an ICC2 (e.g., Kirkman et al., 2001). From apragmaticperspective, as organizations continue touse teams acrosssubsidiariesof multinational firms,we see as very promising research that focuseson how cultural values affect teams and howcultural value diversity plays out at the group/organization level.ResearchchallengesDespite the huge increase in the use of teams inboth the US and in organizations worldwide (Kirk-man and Shapiro, 1997), we found a relative lackofattention to Type I studies at the group/organiza-tion level. However, conceptualizing and operatio-nalizing cultural values at the group/organizationlevel might be problematic for severalreasons. First,researchersneed to grapple with the question ofwhether culture is functionally isomorphic whenmoving from one level to the next (Chan, 1998).Simply put, does the meaning of a cultural valuechange from the individual to the group/organiza-tion level? Clearly,this question is just as relevantfor those researcherswho have adopted Hofstede's(1980a) country level measures for the individuallevel. This issue forms the basis for Hofstede's(1980a) warning about using his measures at theindividual level of analysis; and our review showsthat researchersstudying the group and organiza-tional levels are as curiously silent on this issue asthose working at the individual level. Even ifaggregation of individual data can be statisticallyjustified, the importance of conceptually verifyingthat groupsor organizations can indeed hold valuesthat differentiate them from others in a nationalcontext cannot be overstated.Second, cultural values are assumed to developover time afterrepeatedexposure to multiple facetsof culture(s) (Hofstede, 1980a; Brett et al., 1997). Itmay be unreasonable to assume that because peopleshare membership in a team, there will be morecultural value similarity within, rather thanacross, teams. Simply put, can working in teams(or organizations) alter deeply held, fundamentalcultural values (Triandis, 2004)? Researchers mustconsider when it might be theoretically plausible todevelop group or organizational level modelsinvolving cultural values before research is con-ducted, rather than assume similarity amongindividuals within teams. Similarity withinteams or organizations requires accultural processes

journal of International Business Studies


16/37


and/or attraction-similarity phenomenon, butthese mechanisms are rarely explicitly argued orempirically tested. Further, perhaps the develop-ment of cultural values at the group/organizationlevel depends upon team/organization tenureand task characteristics such as interdependenceor cohesion.In addition, whereas diversity studies have grownexponentially (Milliken and Martins, 1996), cultur-al value diversityhas rarelybeen examined in favorof demographics (e.g., age, gender, ethnicity). Thisexclusion exists despite the finding that deep leveldiversity (i.e., differencesdue to values or beliefs) ismore important for group functioning than surfacelevel (i.e., demographic) diversity, especially overtime (Harrison et al., 1998). Another intriguingquestion for researchers is what matters more, themean level of cultural values in a group or groupdiversity on values? We found no studies thatincluded both facets of culturalvalues at the grouplevel. Further,as SupplementaryAppendix A shows,group level theories were seldom used to buildarguments for the impact of group level culturalvalues. More attention needs to be paid to theunderlying dynamics connecting group level cul-tural values, group processes, and group effective-ness. Theories of collective cognition (e.g., Gibson,2001), borrowing from social information proces-sing, social cognition, or social identity research,might be effectively used.Of the five cultural values, IND-COL wasincluded most frequently in group/organizationlevel studies, perhapsbecause of its close theoreticalties to group behavior. However, links betweenother cultural values and team processes andperformance are equally plausible. For example,PD could be negatively related to willingness of ateam to take autonomous action (Pillaiand Meindl,1998; Kirkman and Shapiro, 2001b). Masculinevalues (e.g., achievement, assertiveness, not caringfor others) may affect cohesiveness (Hofstede,1998). UA could affect group level phenomenasuch as riskyshift or polarization (i.e., the tendencyfor groups to reach more risky or more extremedecisions than individuals). In addition, to assesscultural values at the group level, some studies usedindividual referents (e.g., 'I value my self-interestover my group interest' rather than 'My teammembers value their self-interest over their owninterests') and then aggregated the items to theteam level (e.g., Pillai and Meindl, 1998). Such aprocedure violates commonly accepted practicesfor measuring group level constructs, aggregating

only those items that have 'group' or 'team'referents (Chan, 1998). Thus, scholars shouldstrive for level of analysis alignmentbetween theirtheoretical foundation, hypotheses, operationaliza-tion of constructs, and analyses. If cultural valuesare assessed at the team level, then all itemsassessing cultural values should refer to the team,not individuals within the team, and all statis-tical analyses (e.g., reliability checks, factor ana-lyses, regression) should be conducted usingaggregateddata.Type I studies of culture at the country levelof analysisCulturaldistance (54)Most researchat this level examined the impact ofcultural distance on organizational and countrylevel outcomes. Almost all studies used Kogut andSingh's (1988) index, which comprises the differ-ences between a given (subsidiary)country's scoreon a cultural value and a (home) country's score,with differences summed across Hofstede'sculturalvalues (see Shenkar (2001) and Harzing (2004) forrecent critiques)]. Findings demonstrated that asthe cultural distance between countries increased,the tendency to choose a joint venture (JV)over anacquisition increased (Kogut and Singh, 1988;Chang and Rosenzweig, 2001). However, Brouthersand Brouthers(2001) showed that investment riskmoderated this relationship such that as riskincreases, higher cultural distance is related topreferences for wholly owned entry modes ratherthan JVs. Also, as cultural distance increased,Japanese firms were more likely to choose green-fields (Anand and Delios, 1997) or wholly ownedsubsidiaries (Padmanabhan and Cho, 1996) oversharedownership; the tendency to choose licensingover JVs or wholly owned subsidiaries increased(Kimand Hwang, 1992); the tendency to choose agreenfield over an acquisition increased (Harzing,2002); wholly owned subsidiaries were less pre-ferredthan either shared-equityventures (Barkemaand Vermeulen, 1998; Hennart and Larimo, 1998)or technology licensing (Arora and Fosfuri, 2000);the tendency to choose management-servicecontracts over franchising increased (Erramilliet al., 2002); a greater proportion of incentive-basedcompensation was used for subsidiary managers ofhost-country foreign affiliates (Roth and O'Don-nell, 1996); equity JV partners were more likely toacquire an equal or majority (rather than minority)share (Pan, 1996; Erramilli et al., 1997); greater



17/37

0 A quarter century of Culture's Consequences BradleyLKirkman t300

structural changes in alliance and contracts tookplace (Kashlaket al., 1998); firms engaged in lessR&D(Richardsand De Carolis,2003); and a greaternumber of TMTs departed from US companiesacquired by foreign firms (Krugand Nigh, 1998).In addition, as cultural distance increased, theamount of US foreign direct investment (FDI)decreased (Li and Guisinger, 1992; Loree andGuisinger, 1995); shareholder wealth in thosefirms making cross-border acquisitions decreased(Datta and Puia, 1995); foreign venture longevitydecreased (Barkemaet al., 1996), especially whenJVsor acquisitions were considered (Barkemaet al.,1997); the level of embeddedness and integrationbetween host companies and affiliates decreased(Hakansonand Nobel, 2001); the degreeof personalattachment in international cooperative venturesdecreased (Luo, 2001a), as did the frequency ofexpressive ties in organizational networks (Manevand Stevenson, 2001); and the level of CEO roleconflict and ambiguity (Gong et al., 2001), inter-national expansion performance (Luo and Peng,1999), local responsiveness (Luo,2001b), subsidiaryreturn on assets (Luo and Park,2001), the payoffsfrom JV partner buyouts (Reuer, 2001), IJV sales(Luo, 2002), and the likelihood of success offoreign-owned affiliatesin the US(Liand Guisinger,1991) all decreased. Increasing cultural distancefrom the US was negatively associated with entre-preneurial traits such as internal locus of control,moderate risk-taking,and high energy level at thecountry level (Thomas and Mueller, 2000). Asthe pre-entry international experience of firmsincreased, so did the likelihood of entering cultu-rally dissimilar countries (Erramilli, 1991). Finally,compared with corporate cultural distance,national cultural distance better predicted stress,negative attitudes towards the merger, and actualcooperation (Weberet al., 1996).

Interestingly, others found that cultural distancedid not affect initial or subsequent FDI decisions(Benito and Gripsrud,1992); the type of coopera-tive arrangementsof firms acrossborders (Pan andTse, 1996); the choice between acquisitions, green-fields (Brouthers and Brouthers, 2000), allianceperformance (Glaister and Buckley, 1999), firmperformance (Gomez-Mejia and Palich, 1997),amount of FDI (Habib and Zurawicki, 2002), orabnormal return as a result of US international JV(IJV) announcement (Merchant and Schendel,2000) or after FDIs are made (Kallunki et al.,2001). On the contrary, larger cultural distanceswere related to lower JV dissolution rates (Park and

Ungson, 1997), higher IJV Pothukuchi etal., 2002),and cross-borderacquisition performance (Morosi-ni et al., 1998), decreased preference for greenfields(Nachum, 2003), and increased FDI into Mexico(Thomas and Grosse, 2001). Other research exam-ined relationships using country scores on Hof-stede's (1980a) cultural values while controlling orcultural distance. In these studies, firms preferredFDIover licensing (Shane, 1992, 1994) and soughtmajority ownership in foreign subsidiaries (Erra-milli, 1996) in high PD countries, and UA waspositively linked to majority ownership in foreignsubsidiaries (Erramilli, 1996), a tendency touse JVs or greenfields over acquisitions (Kogutand Singh, 1988), and preferences for greenfieldstart-up ventures over acquisitions (BrouthersandBrouthers,2000).Othercountrylevelstudies (24)Beyond cultural distance, other Type I studiesexamined cultural values at the country level.Regarding HRM, Newman and Nollen (1996)examined the fit between national culture (usingHofstede's country scores) and management prac-tices in 176 Europeanand Asian work units locatedin 18 countries of a US-basedmultinational. Whenmanagersfitted their practicesto a country'svalues,the units had higher return on assets and sales and,in some cases, higher bonuses than those with lessfit. Findings held for all of the cultural valuesseparately except UA. Using a variety of differentworldwide compensation surveys and Hofstede'scountry scores, Schuler and Rogovsky(1998) foundthat IND was positively related to the use of pay-for-performanceand a focus on individual perfor-mance, social benefits such as child care and careerbreaks, and employee stock ownership plans(ESOPs).PDwas negatively relatedto social benefitsand ESOPs.UA was positively related to seniority-and skill-basedpay plans and ESOPsand negativelyrelated to a focus on individual performance. MASwas positively related to individual bonuses andcommissions, career breaks, and maternity leave,and negatively related to flexible benefits andworkplace childcare. From survey data of almost1,000 companies in 20 countries, Ryan et al. (1999)found that, as UA increased, the number used andextent of verification methods in selection pro-cesses decreased (opposite to what was expected).The number of test types, extent of testing, numberof interviews, use of a fixed list of interviews, andnumber of methods of auditing their selectionprocesses all increased as UA increased. As the level



18/37


of PD increased, the overall number of interviewsused in selection and the extent of peer involve-ment in hiring increased while the use of peers asinterviewers decreased.Regardinginnovation championing strategies, asUA increased, preferences for champions to workthrough organizational norms, rules and proce-dures increased. As PD increased, preferences forchampions to focus on gaining the supportof thosein authority before other actions are taken ratherthan building a broad base of support increased;and as COL ncreased,preferences for champions toseek cross-functional support for innovationincreased using survey data from 1,000 employeesin 30 countries (Shaneet al., 1995). In a study usingover 4,000 survey responses in 68 countries,

Shane (1995) found that lower levels of UA wereassociated with preferences for four innovationchampioning roles: the organizational maverick,the network facilitator,the transformationalleader,and the organizational buffer. The greater legiti-macy of these roles suggests that uncertaintyacceptance may be linked to more innovativesocieties. Regardingleadership, PD was negativelyrelated to leader communication, approachability,delegation, and team building; and UA was posi-tively related to leader control and negatively todelegation and approachabilityusing surveydata ofover 400 managers in a single multinational firmrepresenting 39 different countries (OffermanandHellmann, 1997). COL was positively associatedwith team-oriented leadership, and PD and UAwere negatively associated with participative lea-dership in a sample of middle managers represent-ing 54 countries at both societal and organizationallevels (House et al., 1999). The authors note that allof the cultural value items were phrased using a'should-be' orientation rather than a reflection ofactual practices (p 217).In studies of societal outcomes, using data from18 countries common to both Hofstede (1980a) andthe Chinese Values Survey (CVS;Chinese CultureConnection, 1987), Frankeet al. (1991) found thatConfucian dynamism was positively associatedwith economic growth between 1965-1980 and1980-1987. IND was also positively related, butonly in the first time period. IND was positivelycorrelated with national wealth in a study of over13,000 undergraduate students in 31 nations(Diener and Diener, 1995). PD, UA, and MASwere positively related to the level of corruption

in over 40 countries (Husted, 1999). Diener et al.(1995) found that IND was strongly related to the

subjective well-being of nations in a sample ofundergraduate students in 55 nations. UA wasnegatively related to the use of professionalaccounting standards, and positively related touniformity, conservatism, and secrecy in 29 coun-tries (Salter and Niswander, 1995). MAS waspositively related to uniformity and negativelyrelated to conservatism, and IND was negativelyrelated to secrecy. Ronkainen and Guerrero-Cusu-mano (2001) found that countries with higherCOL, PD, and UA had higher rates of intellectualpropertyrights violations in data from 50 countries(using Hofstede's (1980a) country scores).Regardingalliance formation and entry modes (instudies that did not control for cultural distance),firms in high MAScultureswere less likely to pursuetechnology alliances than were firms in high FEMcultures; and firms in high IND cultures were lesslikely to pursue equity ties in their alliance forma-tion in a sample of 494 manufacturing firms inAustralia, Indonesia, Mexico, Norway, and Sweden(Steensma et al., 2000b). Firmsfrom countries withlarge PD prefer subsidiary and equity JV entrymodes whereas firms from countries high in UAprefercontract agreementsand export entry modes,based on country scores from Hofstede (1980a), for10,000 foreign entry activities into China between1979 and 1998 (Pan and Tse, 2000). Firms fromcountries higher in UA preferred higher levels ofequity ownership in 8,078 IJVs n China between1979 and 1996 (Pan, 2002).An interesting set of exchanges regarding rela-tionships between national culture and role con-flict, ambiguity, and overload took place in theAcademy of Management Journal. Peterson et al.(1995) found that high PD and low IND werepositively relatedto high levels of role overload andlow levels of role ambiguity. In a challenge toPeterson et al.'s (1995) findings, Van de Vliert andVan Yperen (1996) found that average daytimetemperature for a country's capital was positivelyrelated to role overload, and that PD scores werenot significant once ambient temperature wasentered. In a reply, Peterson and Smith (1997)added 11 new countries and replaced the capitalcity temperature with the temperature from thecities in which their data were actually collected.Results show that PD was more strongly associatedwith role overload than ambient temperature, andthat the findings extended also to role ambiguity.Finally, the remaining country level studiesexamined various relationships between nationalculture and outcomes. National culture was asso-



19/37


ciated with whether a strategic case issue was seenas a crisis, as stimulating, as a threat, the futurebetter if resolved, as difficult to resolve, quickaction needed to resolve, one correct solution,and as seen as an opportunity based on survey datafrom 303 MBA students and executives in 16countries that were grouped into five countryclusters (Schneider and DeMeyer, 1991). Specifi-cally, LatinEuropeanswere most likely to interpretthe issue as a crisis and a threat. In a 25-countrystudy, those highest in IND exhibited the lowestlevels of information-seeking in individual net-works (Zaheerand Zaheer,1997). IND and PD werepositively (and UAnegatively) related to a country'ssystematic risk in its stock exchange in a 16-countrystudy using Hofstede's (1980a) country scores(Riahi-Belkaoui, 1998). In a 23-country study,Smith et al. (1998) found that PD was negativelyrelated to the frequencyof outgroup disagreements;within collectivistic nations, disagreements weremore frequently handled through reliance on rulesrather than personal experience or training (theopposite was found for individualistic nations); andwithin low PDnations, ingroup disagreementswerehandled more frequently through reliance onsubordinateswhereas outgroup disagreements weremore frequently handled through reliance on peers(compared with reliance on supervisors). Finally,Spector et al. (2001b), using survey data from over5,000 managers in 24 countries, found that INDwas positively related to both internal locus ofcontrol and job satisfaction.

ResearchaccomplishmentsThe research above implies that decisions on howto expand internationally are influenced by howdifferent participating countries are from eachother on cultural value (country) scores. Moregenerally, the findings reinforce Hofstede's (1980a)original contention that values are related to theaggregate management practices and beliefs ofnations. The important findings regarding the fitbetween national culture and management prac-tices (Newman and Nollen, 1996) demonstratedthat being culturally sensitive pays (i.e., with higherreturns on assets, sales, and higher bonuses). These'fit' findings echo recent studies showing positiveaffects for empowerment in Mexico, Poland,and the US, but negative outcomes in India(Robert et al., 2000) and managerial reluctance toempower subordinates in high PD cultures (Aycanet al., 2000).

ResearchchallengesThe most glaring need in Type I studies at thecountry level is to explain the many conflictingfindings regarding cultural distance and decisionssuch as entry mode choice. As mentioned at thebeginning of this section, Shenkar(2001) provideda scathing theoretical and methodological critiqueof the Kogut and Singh (1988) measure of culturaldistance, and speculated that a number of 'hiddenassumptions' could be at the root of the manyconflicting findings with regardto FDIstudies. Oneof these assumptions is the 'illusion of stability.'It is possible that, over time, Hofstede's (1980a)country scores used to create the cultural distanceindices have lost predictive validity (we discuss thispossibility further in the discussion). If no longerapplicable, perhaps an alternative to the countryscore index is individual perceptions of differences(Shenkar (2001); see Kim and Hwang (1992) andLuo et al. (2001) for examples]. Another possibility[not considered by Shenkar (2001)] is that impor-tant moderators have been omitted. For example,Brouthersand Brouthers (2001) found that invest-ment risk in the target market moderated therelationship between cultural distance and entrymode explaining the conflicting findings of pre-vious studies. We urge researchers to continue tolook for theoretically supported moderators of thecultural distance-outcome relationships. Otherstudies showed that examining 'distance' on eachof the cultural values separatelyis important. Forexample, cultural distance on IND-COL and UAwere negatively related to CEO role ambiguity(Shenkar and Zeira, 1992), UA and long-termorientation were negatively related to IJVsurvival(Barkema and Vermeulen, 1997), and IND waspositively related to US research and developmentinvestments abroad (Jones and Teegen, 2001).Pothukuchi et al. (2002) found the positive relation-ship between overall cultural distance and IJVperformance was explained primarily by culturaldistance on MAS. Thus, findings for culturaldistance can be explained by including specificcultural value dimensions independently (Shenkar,2001). As Shenkar (2001) also noted, with regard tointernational expansion choices, perhaps UA ismore important than the other cultural valuesthanks to its theoretical link with attitudes towardsrisk and formalization (e.g., see Barkema et al.(1997) for empirical tests of this possibility). Givenall of these important findings, we take Shenkar's(2001) critique a step further (i.e., he recommendedsimply supplementing the Kogut and Singh (1988)



20/37

A quarter century of Culture'sConsequences BradleyLKirkmant al 303

measure with Confucian dynamism scores), westrongly encourage researchers to avoid furtheruseof the overallculturaldistance ndex.Type IIstudies of culture at the individual level ofanalysisStudies examining the moderating effects ofculture at the individual level of analysis currentlytake three forms (from least to most analyticallyrigorous):(1) testing whether people from various countriesaresignificantly different on culturalvalues andthen, on the basis of these differences, usingcountry as a moderator;(2) testing the moderating effects of culture usingactual assessments of cultural values; or(3) after testing the moderating effects of culturalvalues, determining whether or not countryexplains additional variance.

Clearly, the last approach is the most valuable,because researcherscan show that cultural valuesaccount for country variation. Moderator effectsimply that theories and practices for managingpeople need to be altered based on culturalcontingencies. We organize these 24 moderatingstudies into the traditional management/psychol-ogy categories of negotiation, leadership, indivi-dual behavior relating to group processes, work-related attitudes, motivation, and organizationaljustice.Negotiation (2)Gelfand and Realo (1999) conducted dyad negotia-tion experiments with Caucasian and Asian Amer-ican undergraduatestudents in the US (Asiansweresignificantly higher on COL) and undergraduatestudents in Estonia (who did not significantly differon COL from US subjects). COL moderated therelationship between accountability and profitfrom negotiations such that in high accountabilitynegotiations, the more collectivistic the dyad, thehigher the level of willingness to concede, coopera-tive behavior, profit from the negotiation, andpositive impressions of one's opponent. In lowaccountability negotiations, COL was negativelyassociated with these outcomes. Ng and Van Dyne(2001) found that HIND and HCOL moderated therelationship between minority influence andimprovement in decision quality such that, ingroups with minority influence, individuals higherrather than lower in HCOL were less likely toimprove their decision quality. These results also

held for those higher rather than lower in HIND;and influence targets higher rather than lower inVCOL also benefited more when the influenceagent held a high status position in the group.Leadership 2)Jung and Avolio (1999) examined Asian and USAmericanstudents working on a brainstormingtask(Asians were significantly higher in COL than USAmericans). Country moderated the relationshipbetween leadership style and quantity of ideas suchthat Asian students generated more ideas workingwith a transformational leader than with a transac-tional leader (the opposite was true for USstudents). There were no statistical tests of IND-COL. A second study showed that IND-COLmoderated the relationship between perceiveduncertainty and the odds of using alliances suchthat the relationship was stronger when managerswere collectivistic ratherthan individualistic (Dick-son and Weaver,1997).Individualbehaviorrelating to group processes(4)After confirming that managerial trainees in thePRC were significantly higher on COL than USparticipants, in a three-way interaction, Earley(1989) found that COL moderated the relationshipbetween both accountability and shared responsi-bility and performance such that highly individua-listic people performedpoorest under conditions ofhigh shared responsibility and low accountability,whereas highly collectivistic people performedbetter under conditions of high shared responsi-bility, regardlessof accountability. Country did notexplain any unique variance in performancebeyond COL, and similar results were obtainedwhen country was substituted for COL.In a follow-up experiment conducted with PRC,Israeli,and USmanagers (Earley,1993), IND-COL moderated therelationship between group condition (i.e., workingin an ingroup, outgroup or alone) and individualperformance such that the performance of indi-vidualists who thought they were working in aningroup or an outgroup was lower than theperformance of individualists working alone; theperformance of collectivists was lower in anindividual or outgroup context than in an ingroupcontext; and participant ratings of self or groupefficacy and their anticipated performance out-comes mediated the effects of IND-COL on perfor-mance. Y. Chen et al. (1998b) examined the effectsof individual vs collective primacy (referring towhether people give more weight to their personal



21/37


interests than to their ingroup's interests whenforced to choose between the two) and found thatundergraduates from the PRC (who had morecollective primacy orientation than those fromthe US) exhibited more ingroup favoritism whenthey performedwell individually while their groupperformedpoorly. Results were replicatedsubstitut-ing collective primacy for country and, in somecases, country no longer explained variance aftercollective primacy was entered. Wagner (1995)found that IND-COL moderated the relationshipsbetween both size and identifiability and coopera-tion such that size and identifiability have strongereffects on the cooperation of individualists thancollectivists in a sample of US undergraduates.Work-related ttitudes (5)In a sample of nearly 2,000 managers from 15European and Canadian affiliates of a US multi-national, Palich et al. (1995) did not find moderat-ing effects for IND-COL, PD, UA, or MAS-FEM(using Hofstede's country scores) on the relation-ships between typical organizational commitmentpredictors (i.e., job scope, participative manage-ment, extrinsic rewards,and role clarity)and actualcommitment. Similarly, in a study of EuropeanCommission employees representing 12 countriesand using country scores, no moderating effectswere found for the relationship between variouscomponents of commitment (i.e., affective, norma-tive, and continuance) and

Review of Empirical Research Incorporating Hofstede's Cultural Values Framework

Documents

Transcript of Review of Empirical Research Incorporating Hofstede's Cultural Values Framework