Organizational Research Methods
Edited by
David A. Buchanan and Alan Bryman
®SAGE
RESEARCH ETHICS: REGULATIONS AND RESPONSIBILITIES
79
Research Ethics: Regulations and Responsibilities
Emma Bell and Edward Wray-Bliss
INTRODUCTION: SCIENCE NOT ETHICS?
We wish to begin this chapter with a characterization of academic research, which seems to leave little space, or have little need, for ethical reflexivity.
Proper academic research is a process defined by scholarly detachment from the subject studied. It is a rational, dispassionate, and largely technical endeavour, not emotional, ideological, or personal. It is characterized by value freedom and the eradication of biases. It is properly distanced from political debates about the purposes to which such research might be put. If asked how ethics comes into the frame then, apart from avoiding the obvious issues of falsification of data or blatant abuse of subjects, questions of ethics, like other emotive or subjective issues such as biases, values, and political or religious views, should be largely expunged from the research process lest they taint the dispassionate pursuit of knowledge.
For some readers this characterization might appear to be a 'straw man', a bald caricature of the disengaged academic. For others, given the disparate nature of academic research on organizations, it may approximate an
ideal to aspire to. We present it her.- I i begin what we intend to be a questioning, at times sceptical, examination of research ethics across the paradigmatically divi'-i c field of organizational research: starting \> 'n this view which, perhaps more than any 11! h, ■ seems to pit academic research - presented as 'science' - against or, perhaps more tellingly, above ethics.
Such a view of science can trace its histor. to European eighteenth-century Enlightei merit thinkers, their popularization of fifteenth- to seventeenth-century scientific world view and extension of such views from the areas of natural science to the study o! humanity (Christians, 2005). Located within a political struggle with the landed aristocracy and church hierarchy over the distributions of privileges, inequalities, and social status and an ideological struggle with the Aristotelia 1 Christian ideas, which legitimized these, the Enlighteners promoted reason, observation, facts, and the autonomy of the individual at a direct counter to the church's imposition of an incontestable moral and religious aulhorir In such a context, that which was apprehended by revelation or superstition, or imposed by tradition or domination, was representative
,1 the old order, of the church's and the linded aristocracies' claims to divine rights. To free people from such constraints, reason was to be divided from and elevated above I. iiti. facts from values, the material from ft. spiritual, truth from belief, 'is' from 'ought'. We have here the emergence of a ,1,/v/ of science, the legacy of which we operate wj lh today, where the scientific pursuit i if 'iard facts appears largely separate to ,1'iJ 'divorced from the realm of morals or
However, this apparently clear cut picture nl linlightenment science as separate from, or in .intitliesis to, morality or ethics, is more .■:i,l'ivalent than may at first appear. Woven wllin this view of the liberating force of ti-,->on ' and science is a strong normative, i 'InOal. commitment to the values of progress, truth, and freedom. The freedom of the vi",.ireign individual, acting through reason, i from domination by religious or other traditional moral authority was upheld as the I yltfst ethical end. Thus, though the process nl »aencc was understood to be necessarily
I in 1 from issues of beliefs, values or such, st ence itself was strongly wedded to a liberal humanist ethics. Science and reason were c .1 as key forces through which humankind »( ild move to a progressively less preju-'u-d, less hierarchical, more rational social .■.Jir. Seidman (1998: 21-2), summarizing Ihi 'iews of the eighteenth-century European thinker Condorcet, for example, presents lh1- era as asserting that 'the very nature
II science - its reliance upon facts and observation, its openness to criticism and ic.ision inevitably promotes individualism, '"1-iance, equality, and democracy. Hence the progress of science automatically translates r.t i social progress'.
'1 hough still perhaps an inspiring ideal for many - and, we suggest, the enduring ,,ri *-icitjustification for science and academic ■esi irch more generally - faith in the ' h.'rently liberal, humanist, or progressive i cs of science has been seriously tested ■i more recent times through historical revelations, which have implicated scientists in atrocities against humanity (Punch, 1994).
Neither is social science free from charges of complicity in deeply ethically problematic practices. Prasad (2005: 76-78), for instance, charts how ethnography, considered by many to be the pre-eminent qualitative methodology, has its roots deep in the politics of colonial and American frontier expansion. Urgently needing more information about the 'native' populations so as to subjugate, pacify and/or better exploit them, colonial powers and frontier administrators commissioned and supported ethnographic field work to provide invaluable 'authentic' accounts of the 'natives'. Prasad perceptively highlights how the problematic legacy of colonialism still runs like a seam through contemporary ethnographic research and writing. Such a legacy surfaces, for instance, in quasi-colonial assumptions about the ability of the expert academic researcher to apprehend the native's point of view - to, one might argue, treat the researched as a colonized people or subject race such that the expert researcher can assume to 'know them and what is good for them better than they could possibly know themselves' (Said, 1978: 35; also Wray-Bliss, 2002b: 94).
For some critics, particularly those drawing on feminist and postmodernist traditions, such examples where science has functioned in the sendee of oppression are not merely dismissible as 'bad' science. They also raise deeper ethical questions with the understandings of academic research methods and the authority of academic knowledge, which we have inherited from the Enlightenment. The researcher's detachment and distance - from research subjects, from values or biases, from consideration of the ends that knowledge may be put to - are not, according to these critics, markers of an impartial scientific process that enhances human knowledge and progress. Rather, such detachment represents a discredited, even dangerous, way of apprehending the world that distances the researcher or scientist from an appreciation of their and others' humanity. Glover (2001) provides a compelling, far-reaching account of the consequences of detachment or divorce from humanity during the twentieth century.
80
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
RESEARCH ETHICS: REGULATIONS AND RESPONSIBILITIES
81
He argues that this century's atrocities, wars, and genocides compe! us to re-examine our ethical assumptions and ethicai philosophies. Included amongst the historical experience he considers is the following correspondence between scientists at the I.G. Farben Chemical Trust and officials at the concentration camp, Auschwitz, regarding the procurement of research subjects:
In contemplation of experiments with a new soporific drug, we would appreciate your procuring for us a number of women ... We received your answer but consider the price of 200 marks a woman excessive. We propose to pay not more than 170 marks a head. If agreeable, we will take possession of the women. We need approximately 150 ... Received the order of 150 women. Despite their emaciated condition, they were found satisfactory ... The tests were made. All subjects died. We shall contact you shortly on the subject of a new load. (Glover, 2001: 338-9)
While such horrific examples of the divorce of humanity from the ptirsuit of knowledge are most thankfully rare, a number of other critics have also raised strong ethical objections to more subtle forms of researchers' detachment from the subject and objects of their research. Included amongst such criticisms are arguments linking the treatment of subjects in research as passive objects of study, to the broader objectification and oppression of women and marginalized members of society (Hooks, 1989;Said, 1978; Stanley and Wise, 1993). Critics have also highlighted how conventional practices in academic writing - including writing in the third person, excluding from the text data that is ambivalent or contradictory to the main argument, perfunctory technical sounding summaries of research methods used, and the citing of an author's qualifications and institutional allegiances in biographical notes -push for the elevation of the authority of the 'dispassionate' and 'objective' academic voice (Smith, 1990). Such authority, while useful in making the academic voice heard, may also have the effect of rendering research subject's voices subordinate and subject to that of experts and thereby less able to contest
or contradict academics' and other experts' :" representation or appropriation of their fives    '■ :" (Opie, 1992). For some, the critiques of the    !' ■■ Enlightenment view of science and scientific    -, ■ method are of a magnitude that obliges us lo -"-overturn the historical separation of scieni and ethics and make the explicit articulatioi: and debate about our ethical positions ; intentions central to our research prai (Bauman, 2005; Seidman, 1998).
From the preceding broad brush discussio.i we can perhaps begin to appreciate hov matters of ethics intertwine with researc i both through the process of carrying oul   •'; research and its legitimacy or whether should be conducted at all. In organization! I research, these areas of ethical consideration ■ ■ are no less pertinent. First, in relation to    ' ' the process of research, the dominance of « quantitative research practices which draw on   v; r a positivist epistemology, may serve to redui the status of research populations from that i active, knowledgeable, and diverse subjects I passive, easily categorized objects of knowl edge. The reproduction of the distanced,   -: . detached voice of the author in published   ■. . -texts, in both quantitative and qualitative   - ' research, 'mainstream' and 'critical' stui can obscure the normative and inhei'-i I Incontestable basis of research and push for th. presumption of academic authority over I lives of those studied (Wray-Bliss, 2002a). This may render less-powerful members of organizations, who already have few opportunities to have their voices and view publicly aired, more silent still. ' ■ |
Second, in relation to the legitimacy of organizational research as a worthwhile project, this can be argued to rest, as did ■_■ the Enlightenment view of science before ll. upon largely unexamined assumptions about the goodness of its ends - in this case the 'goodness* of organization. For though each research topic and subdiscipline or field will have its own local rationale - extending Mt" our knowledge of organizational strateg;.. for instance, or optimizing distribution and supplier networks - organizational research in general i s validated by an implicit appeal to the furtherance of organizational effectiveness
This constitutes an assumed ethical end thai serves to legitimize the organizational research arena. However, some critics have
■|.' ied that such assumptions cannot continue .." jnderpin the ethical legitimacy of organizational research. For example, the view that the^ rightful purpose of organizational research is to aid us in a movement toward progressively more rational and efficient Kr>- mization, expressed through notions of ideiice based practice (Learmonth, this
n -k), iii«;y itself be questioned. Critical orga-n ,-ational researchers (Alvesson and Deetz, ■"(K 0); working in such fields as labour pro-L,-;i theory and critical management studies .(iiey and Willmott, 2005) have highlighted ,ir. i critiqued the normative, pro-managerial, di i ■-western (Giiseri, 2002) and masculine .< iherardi, 1995) interests and assumptions j:k erpinning and directing mainstream management and organizational research. Such 11 ■ 'iccrns around the unacknowledged norma-.i vi assumptions of organizational research, the systemic privileging of managerial concerns and voices and the silencing of the interests of a range of other organizational participants and stakeholders are longstanding/Writing in 1974, Braverman, for instance, was highly critical of the failure of industrial mi. idlogists to examine the degradation of Ji.1 working conditions and processes faced by rionmanagerial labour. He attributed this failure to sociologists' own unacknowledged managerial assumptions. By sharing with management 'the conviction that this organi-ziitionof the labora process is "necessary" and
i.icvitable.'", Braverman argued:
This leaves to sociology the function, which it shares with personnei administration, of assaying not the nature of the work but the degree of adjustment "the worker. Clearly for industrial sociology the • oblem does not appear with the degradation work, but only with overt signs of dissatisfaction on the part of the worker. (Braverman, 1974: 29)
I 'ifliculties concerning the ethical legiti-• i icy of organizational research driven by partial, managerial interests may be rendered particularly acute in the contemporary
climate, when the majority of organizational research takes place within the context of business schools which, as Parker (2002: 130) observes, are 'pressed up hard against the utilitarian reality of corporate managerial-ism' , Researchers working in this context may feel, in innumerable material and symbolic ways, the need to be 'relevant to' and 'in the service of business impressed upon them -rendering re-examination of the 'goodness' of commercial organizational ends that much more difficult. Yet some critics have called for a re-examination of largely taken-for-granted assumptions about the goodness of modernist organization per se. Echoing and extending concerns regarding the rationalizing and homogenizing pressures in western society and large scale bureaucratic organization, the postmodern social critic Zygmunt Bauman has powerfully questioned the beneficence of the modern organizational form. His work has linked treasured modernist features of organization - its complex hierarchy, focus upon procedures, instrumental orientation, delineation of responsibilities, and removal of personal sentiment from the efficient performance of organizational roles - with the potentially catastrophic, systematic erosion of moral responsibility in organization (Bauman, 1989/1991). Following Bauman, even the goal of more rational or efficient organization, an end upon which the legitimacy of organizational research at least implicitly relies, cannot be assumed to be an unquestionable good. Furthermore, the nature of goodness itself - or more specifically questions of how we are to decide on matters of ethics - is also contested. The long history of moral philosophy, while aspiring to some universal and eternal understanding of goodness, has achieved the production of multiple well-reasoned but contested and at times conflicting formulations of ethics (Maclntyre, 1998). The modern citizen, including of course the academic, is the inheritor of this legacy, which has become structured into our institutions, laws, customs, and individual moral subjectivities. Such that now, notions including duty, virtue, responsibility, fairness, consequence, and even happiness carry echoes of different
82
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
philosophical ethical traditions that each encourage us to place different weight on different outcomes, intentions, customs, or ' contexts in our deliberations of what is ethical.
From the preceding discussion, several points emerge, which we wish to carry forward in this chapter. First, research ethics is informed by wider philosophical debate and discourse, but it is not an abstract, peripheral, or principally philosophical concern. Instead, it encompasses a set of empirically and theoretically driven issues that permeate the entire research process. Second, organizational research has a cumulative, and sometimes immediate, effect upon participants, which cannot be assumed to be necessarily beneficial. Third, ethical considerations run through all types of organizational research, both quantitative and qualitative, affecting choices of method, degree of detachment from or relationship with the researched, the way that the organization, data, or other's experience is represented in the text and the way that the academic voice is authorized. Finally, the credibility and authority of organizational research rests on either an implicit or explicit ethical warrant, which ensures its ongoing support. One of the principal ethical warrants on which organizational research is founded is the assumed goodness of a managerial agenda and/or the beneficence of modernist organization. These assumptions in turn draw on the historical legacy of the Enlightenment. This legacy is wedded to liberal humanist notions of progress, reason, and the furtherance of humanity and which understands science to be at the heart of this process. It has passed down a view that the researcher's detachment from central aspects of their own and others' humanity is at the core of the scientific method. However, we have suggested that we can no longer, in the current context, make assumptions about the progressive nature of this detached scientific process, of science and research itself, or indeed of the goals of modernist organization. This is not to say that researching organization or organizational research processes are inherently
unethical or harmful - far from it. At a tune when the reach, power, and complexity of organizations risks outpacing our traditional processes of democratic accouiilabil-ity, intellectual comprehension, and mocul imagination, understanding organization is. we would argue, fundamental to wellbeuig and survival. Rather, the implication , the above would seem to be that we, as a community of organizational researche * need to subject our research practices, agendas and assumptions to continual and sustained ethical scrutiny. In the following se-nkm, we explore how the organizational rose, community currently engages with su.h matters.
The status of ethics in organizational research
Within  organizational  research,  we i 11 identify what we might call implicit and explicit engagements with ethics. In-'i n engagements include a myriad of choices -and assumptions, which may not be s .| L itly articulated as ethical issues but wh.J' nevertheless imply ethical choices. These include a researcher's choice of topic tht nonfalsification of research data, avoids i i-il plagiarism, judgements around whose '. ■ ■. i will be represented in the research and i (Wray-Bliss, 2004), even decisions regu Ii 1, where or if to publish the research and how it will be written up imply choices that may impede or enable other's access !■■ I" research, to knowledge about their world or access to how they have been represented, and understood in the wider society (Pinker,, 2002; Grey and Sinclair, 2006). However, (he above issues are seldom explicitly artic uliilcd or engaged with as ethical issues. More nllen . than not, they will be seen as merely mailers of academic convention in the particular nrga- • nizattonal research field. Instead, 'research-ethics' are normally understood more iwr- ' rowly and discretely as concerned with I he need to follow a set of clearly defined |" -cepts concerning appropriate and acceplable. research conduct. These explicit engagement with ethics involve consideration of .i"1"
RESEARCH ETHICS: REGULATIONS AND RESPONSIBILITIES
83
.^hi hers should treat the people who form \      .:1s of their investigation and whether
■ , r-i .'i- certain actions that should or should
,i 11- i.i'.en in relation to them. This approach vi U 1 result in the generation of a set -t iJ principles, which organizational 1 i li'i'S are expected to show that they iMv» Li'iiwlcrcd, usually at the design stage j |;n't ,earch. Failure to adequately address
■ „ .„■ i- -lies may result in censoring by the ,.      m1 tonal research community.
•I |u. main mechanisms for ensuring compli-,„ ,■■!.>- leli ethical precepts have traditionally been self-disciplinary. These have taken the I , i, ,i| .i.iiicreuce to codes of ethical practice i t-,.'l,i| ,i1 by professional associations to uIm .i i •! iianizarional researchers belong. The oncins of Ihcse codes and the ethical precepts m l-i-i lliein can be traced to the military h.hi.i. I known as the 'Nuremburg Trial' • involving-.wenty-three German physicians aril :-in uislralors who were accused of war ciiii.'. im .'rimes against humanity for their r rlk' h'lion in inhuman experimentation on llii-w ikcmcd- unworthy of life during the Si-uliJ World War. The appalling nature pi ii m. atrocities, and the shock of the inlerniirinnat.com munity at the role of science in their perpetration, led the trial judges to 1-n '.iLlc. a universal standard for human ■."(■.■iinn illation in an attempt to pave the v.j, im j reconstituted moral vision (Katz, !>■"'' Argument submitted to the United Si. k . t '■ iiinsel for War Crimes by doctors for Hit pii-^cution sought to define legitimate resi.iiil' nn human subjects and to protect llu' ii lioin potential harm through the princi-I'k-i '. ■! lntary informed consent. In a section ci l.l i, ' Permissible Medical Experiments', the- ■.cfdict:: established ten points, which -i'1 .'ii ■ he.areasof: (i) ensuring the voluntary conseul ofthe subject: (ii) experiments must be for lh.es good of society; (iii) sufficient fti.'i icearch must have been conducted; Uv) unnecessary suffering or harm must be iii'idi.; (v) experiments with the chance f nn;, or death are prohibited; (vi) the ^lould never outweigh the humani-1 1 ' i portance of the problem studied; (vi i) preparation and facilities must be in place
to avoid harm or injury; (viii) only properly qualified persons may conduct the research;
(ix) the subject can withdraw consent; and
(x) the scientist must terminate the experiment if the risk of harm warrants it. These ten points have subsequently become known as the 'Nuremburg Code' and have informed both biomedical and social science ethics codes since.
For nonmedical organizational researchers, professional bodies such as the Academy of Management, or related social science disciplines like the Social Research Association, American Psychological Association, or British Sociological Association set out the main ethical issues that researchers tend to face and provide guidance on how they should respond to them. That there has been relatively little variation in the content of these codes since the 1960s (Bryman and Bell, 2003; Bell and Bryman, 2007) suggests that a broad consensus has emerged surrounding the definition of ethical issues in social science research founded on the principles established by the Nuremburg Code. Such codes universally specify the avoidance or rniniinization of harm or risk of harm, whether physical or psychological, and the need to ensure that research subjects are fully informed about the nature of the research and give their consent to being involved with it. In addition, they stress the necessity to maintain the privacy of research subjects, ensuring the confidentiality and anonymity of data where appropriate, rninimizing the possibility for deception at all stages of the research process, from data collection to dissemination. There may also be an obligation to declare sources of funding and support that have the potential to affect the affiliations of the researcher or cause a potential conflict of interest. In each case, it is the researcher's responsibility to ensure that these ethical issues have been considered and that steps have been taken to address them where necessary.
Despite, or perhaps because of, the convergence of ethics codes there has, in recent years, been an increase in attention and debate concerning ethics in social scientific
84
THE SAGE HANDBOOK Of ORGANIZATIONAL RESEARCH METHODS
RESEARCH ETHICS: REGULATIONS AMD RESPONSIBILITIES
research (Sin, 2005; Truman, 2003; Kent et al., 2002; Nelson, 2004; Wiles et al., 2007). Such sensibilities have informed a terminological shift away from a focus 'on hapless research participants, conventionally termed, subjects, a word reflecting the concept that research participants have things done to them' (Lincoln and Guba, 1989: 222, emphasis in original) towards the use of the term 'participants' (Birch and Miller, 2002). These debates have been most vocal amongst qualitative and reflexive researchers who draw on feminist, postmodern, and postcolonial critiques to critically re-examine the ethics of their own research practices. This may have had the unintentional effect of encouraging more positivist and/or quantitatively minded researchers to assume, mistakenly, that consideration of ethical issues is more relevant to, or only becomes problematic in relation to, qualitative research.
In comparison to the broader social sciences, organizational researchers have been more reticent in reflecting on the implications of these debates for research practice (see Brewis and Wray Bliss, 2008; Bell and Bryman, 2007). Yet, there is little doubt that their impact is likely to be significant, not least because they have been paralleled by a rise in ethical governance, as the agencies and institutions that fund or disseminate organizational research or employ organizational researchers have become more active in seeking to regulate ethical conduct. For example, recent revisions to the code of ethics produced by the Academy of Management allow researchers to be penalized for ethical breaches through a process of formal adjudication, which can result in their exclusion from the Academy (although it should be borne in mind that there is no requirement on organizational researchers to belong to this or any other professional association). In addition, academic institutions that employ researchers, driven in part by a concern to avoid potentially costly litigation, are taking a more active role in overseeing the ethical conduct of their members. Within universities, research ethics committees (RECs) or, in North America, institutional review boards
(IRBs) have been established with respon, r bility forjudging adherence to ethical precej .| -in research proposals submitted by resuu ch f students and academics. These developments 'I signal the establishment of ethical governan structures, which are more formalized. a ii less Tight touch' in comparison to fn i „r voluntarist systems (Kent et al., 2002). St developments may be interpreted on one I a* as a welcome indication that the acadi research community is engaged in a serii and sustained way with ethics. However, a ~t number of concerns have been expn regarding the current trend toward g , formaiization of ethical accountability in the 1 social sciences, which we outline in the ■ •I section that follows. "
Critiques of ethical formalism: Committees and codes :m
Characterizing the historical treatment of -*Jk issues of morality and ethics in civil sock I Bauman (1993) argues that we have I i.li I to reduce matters of ethics - undei n i J as that unending innermost compulsion to . .'I seek to do what is right, to know < i. own heart, to be responsible toward the ~\ Other - to matters of following predi- "ii ; rules, laws or codes, hi so doing, st : i uil! proportion of the population have been ' granted the moral status to reflect, philo' ■> phize, and codify what is ethical while rest have been allotted the roles of me. I) following these rules and norms, socialise into doing so through the mechanism! : organizational and state bureaucracy. For. Bauman, such rule-following, bureaucrat: /i I behaviour does not represent ethics tail -rather its antithesis - the abdication 11 individual ethical agency and its replacement with conformity. Whether we take Jiaimiuu's broad-brush depiction of society at face value or not, we can begin to appreciate that the formalization or bureaucratiza 11 of ethics may be questioned. Such a critique focuses on whether ethical formal in enhances accountability and responsibilil ■ ' limits and constricts our engagement -Mil matters of ethics.
If we apply these considerations to research
■ihics we ciul perhaps begin to appreciate sonic    uie concerns mat have been voiced
■iřárding ethical governance as promoting a nlitl arid inllcxible understanding of ethics,
■ii Hbat encourages the ticking of boxes rather iluin ongoing ethical reflexivity. Critiques of ethical formalism in social scientific research ■i. ědfiven by concerns about the disciplinary .■Meets associated with these mechanisms and hv the possibility that this will lead to the imposition of a hegemonic view of what constitutes 'good' research. Critics seek to
iirue that social science research operates differently from other research involving h.iEiian: subjects (Coomber, 2002) and that tht principles that are used to regulate the community need to reflect this. One .it 'the- criticisms that can be made of the standardization of research governance is that it does not necessarily encourage or support rcsiciiihers in the process of becoming ninn'k-aclive individuals, since it is only through conducting and writing up research Ih? i'..' develop a full awareness of ethical issues A further critique is that they may iv li.nc killing so as to prohibit, or make ii Tif-i inlly more difficult, valuable forms 11 ii .t .irchthat do not readily conform to a njii'ulv defined set of bureaucratic norms,
'lithe ways of ensuring ethical research are heavily reliant on RECs, this may not serve research practice well. This seems to be especially the case where processes are rigid and concerned only with assessing the ethical issues prior to the project start, when the information needed to make the assessment is not available' (Richardson and -- MiMullan, 2007 1127).
IK-sc i oirils can be illustrated in relation to '«■.' |'ii t.iple of informed consent, where the rote of the KI (C lypically is to:
■ .a) make sure that the research proposals passed by the committee are appropriate and safe, and b) that the prospective participants are fully appraised of tne research and of their rights (e.g. to anonymity or to stop the research at any time without prejudu e). The method increasingly used to achieve the second of these concerns is that of the signed consent form, (Coomber, 2002, section 1.1)
The adequacy of form-filling as a method of ensuring that participants understand the implications of consenting to participate in a research investigation has been questioned by several commentators (c.f. Edwards and Mauthner, 2002; Bhattacharya, 2007). Sin (2005) and Wiles et al. (2007) suggest that consent is contingent and situated, varying according to whom one is dealing with and how definitions are operationalized, a process that relies on an ongoing process of negotiation, which cannot be adequately addressed by getting participants to sign a form. This is a particular issue for qualitative researchers and those working with theoretical perspectives that seek to overcome traditional knowledge hierarchies and power relations in the research process through, for example, the adoption of feminist, participatory or decolonizing methodologies (Bhattacharya, 2007; Boser, 2007). RECs tend to be predicated upon a medical model of the research process that is insensitive to more flexible, emergent research strategies (Truman, 2003; Wiles et al., 2007; Lincoln and Tierney, 2004; van den Hoonard, 2001). Methods such as participatory action research (Chapter 2, this volume) make it virtually impossible to anticipate all conventional ethical issues in advance of a study being undertaken (Bell and Bryman, 2007). The problematic nature of informed consent is also apparent in relation to ethnographic research, where the researcher's daily participation and routine appearance in the lives of those she or he researches can encourage participants to forget that their interactions and conversations are the subject of study.
Bhattacharya (2007: I U0) argues that there needs to be space within regulative structures for addressing 'the implications of fluidity and the contingent nature of qualitative research'. In such research, relationships with participants are continually deferred and not necessarily defined purely in terms of a short-term, instrumental academic need to collect data. Although it is also important to note that all types of research can deal with complex ethical issues, which may not become apparent until after the study
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
has begun. The requirement to obtain signed consent also has the potential to adversely affect the participation of particular groups in research, such as those who wish or need to remain anonymous because for example they are involved in committing illegal acts, by increasing rather than reducing their anxiety about participating in a research investigation (Coomber, 2002; Nelson, 2004). Ethical regulation based on principles intended to protect vulnerable research subjects from harm has the potential to make it more difficult to study organizations and individuals who do not want their activities to be studied, using covert methods for example (Galliher, 1982; Hen-era, 2003). Finally, while informed consent protects research participants from harm, it does not allow for the possibility of participants causing harm to others (Koro-Ljungberg et al., 2007).
Ethical formalism has also resulted in the widespread convention that research should be anonymized in publication to protect participants from the risk of potential harm (Cassell, 1978). While this might initially seem to be a clear-cut ethical good, Grinyer (2002) has questioned this assumption, noting that in some cases it can be extremely important to participants that their accounts are not anonymized so as to maintain their authenticity. While in organizational research there are many instances where organizations that authorize research access are not anonymized in publication, increased ethical formalism poses a challenge to this convention. Furthermore, an argumentmay be made not to anonymize research subjects on the grounds of a felt ethical responsibility to expose and document forms of organizational exploitation, abuse or malpractice. Within organizational research, there has been only limited acknowledgement of the implications of unequal power relations in the research process. Relating to participants in organizational research as vulnerable subjects can unwittingly serve to reproduce existing power relations and hierarchies (Wray-Bliss, 2003a). Uncritically adopting, or being compelled to adopt, research protocols that have been specifically designed to protect
vulnerable subjects may institutionalize I oruis of practice that take-for-granted and merely seek to mitigate the effects of assumed pov* ■ i inequalities between 'researcher' and Mhe researched'. In organizational research, the majority of participants are adult, employed, and often highly educated. This potentially enables them to exert considerable influence in shaping the nature, direction, and outcomes of the research to a far greater extent than ethics codes generally allow for.
Such discussions fuel concerns about the rise of an audit culture (Power, 1997) v.:i 11 research ethics where having been si have 'done ethics' becomes more important than meaningful engagement with the issues raised. This encourages a view of ethics as a separate component in the research process that is dealt with at a certain point, potentially having the effect of discouraging researchers from continuing to refie\i- ^L engage with ethical issues througho i 'l. research process (Mason, 1996). As Si i (2005: 289) argues, 'the tick-box appro. ■. 1.1 ensuring compliance in ethical standards ... is an oversimplistic representation of 'I,. "messy" realities surrounding the research process'. Moreover, the formalization of ethical governance is by no means consislcnl. as Richardson and McMullan (2007: 112ft) note, 'RECs vary greatly from institution to institution and there is no consistency i \en about whether the research ethics comi i i.i ■ operates at the level of the discipline, the level of department/school, at faculty level or the level of institution'. The apparent inability of ethical governance structures to deal with certain types of quail;;: .■ research is suggested to have led sonic RECs to reject qualitative research projecls for being unscientific and ungenerah . I k The disciplinary effects associated with such a discourse are, therefore, potentially significant:
Narrow definitions of scientific inquiry - and evidence - serve ultimately to circumscribe painfully and dangerously the range of what is consider.'d useful knowledge and, consequently, to I'mit the kinds of studies supported by a range '
RESEARCH ETHICS: REGULATIONS AND RESPONSIBILITIES
jdinmistrativeand managerial structures, including IkBs funding agencies, foundations, and state novernrrtents and agencies. The long-term results jro iyslematit constriction of academic freedom (lor individual researchers) (Akker, 2002), a seeping 'oss of institutional autonomy (for institutions q' plqliei"-education), and a loss of needed apislcmofogical perspectives on persistent social iustice issues, (fincoln, 2005: 173)
Wi I iii CJK health research, there is evidence that Ihc rigidity of ethical governance mi, n .'s is causing researchers to modify Ik'ii i. 'arch strategies or even to stop doing research m this type of organization altogether iKkI<:iiiIse>ii< and McMullan, 2007). Since c ,ii.-al governance structures in health are increasingly being used as a model for other dhCLpl r is; it is entirely possible that over :| irh. pattern will also affect researchers studying other types of organizations.
V lie preceding discussion illustrates, In il. lining what is acceptable or 'good' iiiJ.Ti-. iional research, ethical governance i ii -liii cs have significant disciplinary effects (Lincoln and (lannella, 2004a, 2004b), their n:"\.-i being derived from having the ability m .h i tie:.flic way that research is con-u.u'il Funding bodies may make grant .iftjiis londirional on the researcher having i.c ii' n I ated that ethical issues have been <. ■!.:, d.1 od while RECs may grant researchers i in.ii ilal approval, requiring modifications lo be made in advance of the study. Such struc-I in ^ k >e the ability to restrict researchers' .n.l l« o follow research questions they ■u Ii m pursue. There are fears that in so doing academic researchers' independence an m hty to speak 'truth to power' may !-■ diiii.nished. For some, this has evoked a .mi.-n about academic research being manipulated to support government, execu-other institutional or organizational - what in the US has come to be s 'Bush Science' (Lather, 2004). As ungberg etal. (2007) argue, the notion
I ' j. •!,
in'. i >,i
I. im.' ,
1! hic-lree science' disguises a subtle form 11 'iii ision, which is based on the notion that those in power act neutrally to defend hie {common good'. Adopting aFoucauldian f-ispective, they argue that IRBs constitute
a type of 'examining apparatus' through which the 'normalization' of researchers and research projects is achieved as a form of control.
Lincoln (2005) suggests that researchers need either to communicate with RECs or to serve on them as a means of representing their interests and getting their voices heard, or even to try to work in partnership to try to understand their worldview (Boser, 2007). Similarly, Bell and Bryman (2007) argue that the organizational research community needs to become more actively engaged in these debates if they are going to be able to iniuence ethical governance structures in a way that ensures that the particular issues it faces are taken into account. These concerns find additional purchase for organizational researchers when we consider the way that ethical codes are being shaped by bodies such as funding councils. For example, the Research Ethics Framework launched in 2005 by the Economic and Social Research Council, the main funding body for UK social science research, is concerned to minimize the risk of harm to not only individual research participants but also their organizations or businesses. This constitutes organizations as well as individuals as the subject of ethical considerations (Bell and Bryman, 2007) and in so doing potentially recasts ethics in a way that restricts the ability of researchers to examine critically certain issues such as forms of oppression within organizational structures, which are subordinated to the organization's right to be protected from critique.
Given that ethics codes originated with the Nuremburg Trial's concern to give justice to human subjects who suffered at the hands of those who were integrated into the organized atrocity of the Nazi holocaust (Bauman, 1989/1991), the extension of definitions of harm to include protection of organizations might seem a perverse departure. However, it must be acknowledged that the definitions of goodness upon which ethical oversight regimes are based are not neutral - they are founded upon a modernist, Enlightenment view of science outlined at the start of this
88
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
RESEARCH ETHICS: REGULATIONS AND RESPONSIBILITIES
89
chapter. Ethical formalism can encourage a view of ethics as a series of abstract, neutral, objectively measurable principles, which can be universally applied, rather than a series of politicized judgements about the legitimacy of research as a worthwhile project based on assumptions concerning the goodness of academic knowledge about organization. We end this chapter by considering the implications of this final point and suggesting some possible responses to the issues we have raised in this chapter so far. As part of the organizational research community ourselves, these proposals are of course informed by our own theoretical and epistemological affiliations. We, therefore, offer them for debate and consideration, rather than in any attempt to set an agenda for others to follow.
Relating to and representing the other
Throughout this chapter we have sought to emphasize the importance of ethics, not only in relation to the process of doing organizational research, but also in terms of what is perceived to be the legitimacy of its purpose. How we as an academic community respond to the rise of ethical governance structures is in large part detenrrkied by the answers to these broader questions. Addressing them involves evaluating the ethical intentions that undeipin organizational research; in other words, why we do it and the purposes it serves. This not only entails sustained examination of the relationships between the researcher and individuals and organizations who become participants in organizational research - the traditional focus of discussions about research ethics. It also involves exploration of the relationships that organizational researchers have with other groups, including audiences of managers and students who read and apply organizational research, universities and funding bodies that provide financial support and other researchers whose work informs it or who continue to develop it.
Some initial insights into these issues can be gained from considering how organizational
researchers have responded to ethical ovei-sight regimes and critiques of research eilucs, One potential response has been simply u> push for the reduction or even the rejection, of ethical formalism. This is justified on the grounds that either there is a commonality of interests between organizational researches and those individuals and organizations tlrat constitute the focus of their research, i ,| that the professional integrity of resear.-hi i. is such that no such external disciplinary mechanisms are required. This latter point would seem to evoke a deontological ethical position, where the researcher searches .I:;-1 own heart andreason to determine their ethical obligations. If this assumption is cone.-. n may seem to follow that formal guidelines* on research ethics, developed in large pail to protect the interests of the researched, may be dispensed with. Alternatively, we might argue, drawing implicitly upon a utilitarian ethical formulation and weighing up the moral worth of consequences for the majority, that the goals of scientific research have an impoi . r.i that overrides issues of research ethics. However, in the context of the postposilivist questioning of the modernist faith in sci.- :■ e truth and ideology in recent decades discussed earlier in this chapter, we suggest the political naivety of the first position and arrogance of the second are dangerous bases upon win f to assume a commonality of interests u i override ethical concerns.
If we accept that the formalization it research ethics is a trend that cannot be reverted then another response would be to engage directly with ethical over-ii: i regimes to ensure that the concen ■■ "I organizational researchers are effectively represented. This response introduces niivi dynamics since it potentially lends ethical oversight regimes greater authority through which to require researchers' conlo 1.1 \ to predefined ethical codes. Furthermoic, while the representation of different i ll ical issues or academic disciplines may mitigate the narrowness of definitions a™ applications of ethics, the constitution and function of such bodies tends to suggest pies-sures toward universalising and normalizi is.
.. ■ .■ i ethics. This generates the same n .i-U-nis of exclusion, the application of '« i ntially inappropriate ethical criteria and ■i ^(ioris of how to apply such criteria ' > it-is discipfinarily, methodologically, and
pcl'tically diverse fields and subfields of = iizational research. In short, we remain
, ii.-i Kvinced that such processes are the i.n   or indeed the best way of encouraging nizational researchers to actively and
„.....itcly engage with the complex range of
ell ual issues that doing research generates, further strategy would be for the iiizational research community to move iiu.iv from the ethical universality implied - I v compliance-oriented ethics codes and
X "      instead to embrace ethical pluralism. This ,wni saeh would involve different communi-# •  "   ties oi organizational researchers attempting = 11 j ticrl itc ethical practices for their spe-
\ ,-ilc lumi nullities; positivist organizational
isearchers could have one set of ethical ! p.^c i-i    interpretive researchers another,
i .-n.sal researchers yet another (a situation
i '.Lii h some may suspect already pertains
! Ihi-i.gh is rarely explicitly articulated). How-
the adoption of a strategy of ethical I II i ilism introduces other challenges, for follow the logic that ethical respon-| sihil l.i . vary according to the research
i .MiiimuniLy studied, we can envisage argu-
.' nrn s: for responsibilities not only to be
1 .-ill. inccd when, for instance, vulnerable or
j l:.s(.Hcally silenced organizational subjects
studied but also diminished when the 1 •    irch subjects constitute a community
in some way, has forfeited our normal t ."tlii. alregard. For organizational researchers
} .1 in'.ing upon postmodernism and Critical
j 'Mi. niiy;: acutely conscious of the political
i fqiialitics of organization and the dangers i 1  iniversal truth claims, for example, a
| -i'.'. lized or selective research ethics, one
j r| 11 could vary according to the political
j of the community studied, might
j nut he; inconceivable. Such selective ethics,
l ■lo'scyer, raises difficult issues about the
f 'w i ie of ethics and politics of academic
reh, which have yet to be systematically >red (Collins and Wray-Bliss, 2005).
Any movement toward ethical pluralism will have to engage with the historical context of the attempt to universalize research ethics. The Nuremburg Code was devised as a universal code precisely so that groups of individuals - be they Jews, the disabled, or indeed those whose politics one vehemently disagrees with - could not be removed from the status of moral beings. Movements toward dismantling the aspiration to universality - the desire to write a code that says to all humanity for all time that the potential atrocities and abuses of science are not to be tolerated in sciences' name - will need to think carefully about what protections or prohibitions have the gravity to take its place.
While we find ourselves having some sympathy with each of these arguments we believe that there is still a need for greater transparency and reflexivity than these approaches invite and that this necessitates a process of cultivating greater ongoing critical self awareness in relation to ethics. Our final suggestion, therefore, entails shifting the emphasis away from approaches that are informed by teleological and deontological theories and concerned with 'what to do', toward an approach which instead seeks to develop dispositions such as honesty, sensitivity, respectfulness, reciprocity, and reflexivity, driven by a concern with how the researcher should 'be' (Wax, 1982; Lather, 1986; Lincoln, 1995; Wray-Bliss, 2003b; Bell and Bryman, 2007; Richardson and McMullan, 2007; Brewis and Wray-Bliss, 2008). This invites conceptualization of the ethical relationship between the researcher, their topic, and those whom they research as an ongoing, active, personal, and relational responsibility. In addition, it encourages a shift away from a controlling orientation to ethics as primarily involving formalized compliance to bureaucratic rules, toward one that is a more enabling and aspirational. It entails a focus on how organizational researchers choose to conduct themselves, casting ethics as a matter of beliefs, thoughts, and values as well as actions. For critical management researchers who are concerned with exposing and redressing the political
90
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
RESEARCH ETHICS: REGULATIONS AND RESPONSIBILITIES
91
inequalities of organization an approach to research ethics, which takes into account the relative power and influence of the groups represented in the research might seem particularly appropriate (Bell and Bry-man, 2007; Brewis and Wray-Bliss, 2008; Wray-Bliss 2003b). Such a position also evokes feminist and postmodern approaches to ethics that emphasize responsibility to the Other (Bauman, 1993; Benhabib, 1987) through valuing connectedness with the actual researched populations who may be particularly vulnerable or have historical reasons for being particularly suspicious of the intentions of academic researchers (McLaughlin and Tierney, 1993; Bishop, 2005). This approach might further necessitate organizational researchers devoting more attention to ethics. For example, by particular journals becoming more proactive in specifying ethical standards to which their prospective contributors must attend and encouraging greater discussion of ethical issues in published articles, or through the education of organizational researchers in a way which casts ethics as an ongoing aspect of their professional development. Most fundamentally, however, rather than only being relevant to certain methodological approaches, ethics would be treated as a central aspect of the conduct of all research involving organizations.
We have, in this final section, outlined a number of possible responses to ethical formalization in relation to organizational research. We do not regard any of these responses as ideal. Nor do we envisage the possibility of an ideal response, either for the communities of organizational researchers or for individual researchers. This is because matters of ethics centre on the fundamental question of how we should live. Moreover, while much moral philosophy has aspired to universal ethical pronouncement on the good life, we are wary of the potentially subordinating power effects of such claims. For us then, research ethics involves explicit consideration of how we should relate to research participants, other researchers, institutions that support research, and research
audiences, including those who read and apply our research. Such questions, as we ha11 sought to emphasize throughout this chapter, do not lend themselves to easy solutions. Our position here has been, therefore, to suggest that ethical issues permeate the. nl i L research process for all researchers whether positivist, interpretivist, postmodernist, and whether dealing with quantitative and/or qualitative data. Furthermore, while c-u 1 formalization is a trend which in our view cannot be easily avoided, the way that i I of us, as individual organizational researchers, engages with such challenges has the pol .-:r i I to vary enormously. We can all, of o ■ i fall back on the path of least resistance at' I minimally comply with whatever res-.- iri i ethics principles are prescribed to us I. our funding bodies, journals, supervisor1. . university ethics committees. Alternatively, we can seek to embrace the possibililies that ethics introduces for reinvigorating and redefining the way that we think of and do organizational research.
REFERENCES
Akker, J. (2002) 'Protecting academic freedom wnr'd-
wide', Academe, 88{3): 44-5. Alvesson, M. and Deetz, S. (2000) Doing CnUcal
Management Research, London: Sage. Bauman, Z. (1989/1991) Modernity and the Holocaust,
Cambridge: Polity Press. Bauman, Z. (1993) Postmodern Ethics, 0xrard:
Blackwell.
Bauman, Z. (2005) 'Afterthought: on writi ic, on writing sociology', in N, Denzin and Y, Lincoln (eds) Sage Handbook of Qualitative Research (third rdn), London: Sage, pp. 1089-98.
Bell, E. and Bryman, A. (2007) 'Ethical cedes and management research: a comparative analysis', British Journal of Management, 18(1).
Benhabib, S, (1987) 'The generalised and the cenrrete other', in E. Frazer, J. Hornsby and S. Lovibond (eds), Ethics: A Feminist Reader, Oxford: Blackwell.
Bhattacharya, K. (2007) 'Consenting to the rorisct form: what are the fixed and fluid understuiďrgs between the researcher and the reseaithori', Qualitative Inquiry, 13(8): 1095-15.
Birch, M. and Miller, T. (2002) 'Encouraging participation: ethics and responsibilities', in M. Mautfr.rr,
[/■:Birch, 1 Jessop and T. Miller (eds), Ethics in Qualitative Research, London: Sage, pp.91—106. Bishop;1" R- (2005' 'Freeing Ourselves from neocolonial dorriih'atiwi in research: a kaupapa Maori approach tocreating knowledge', in N.K. Denzin and y S:; Lincoln (eds), Sage Handbook of Qualitative Research, (third edn.), fhousand Oaks, CA: Sage, pp.109-38.
Boscr,:S: (2007) 'Power, ethics and the IRB: dissonance over1.-human participant review of participatory „;..(•.,-.-( |-, Qualitative Inquiry, 13(8): 1060-4.
Bravetrfiarc, H. (1974) labour and Monopoly Capitalism: flfei:Degradation of Work in the Twentieth Century, laridon: Monthly Review Press.
Brewis, J: and Wray-Bliss, E. (2009) 'Re-searching ethics-towards a more reflexive critical management studies'!-OrganizationStudies, 29(11): 1-20.
Bryrnan, :'A. and Bell, E. (2003) Business Research Melhods, Oxford: Oxford University Press.
Cassell, J,-(1978) 'Risk and Benefit to Subjects of Fioldwork', American Sociologist, 13: 134-43.
Christians^.G. (2005) 'Ethics and Politics in Qualitative Research', in N. Denzin and Y. Lincoln (eds), Sage Handbook of Qualitative Research (third edn.), I ondon: Sage.
Collins, H. and Wray-Bliss, E. (2005) 'Discriminating EthicsV'Wuman Relations, 58(6): 799-824.
Si.-an bet. P. (2002) 'Signing Your Life Away', SociologicalResearch Online,7'(1). <http://www.socresonline. orcj.nk/7/1/coomber.html>.
Edwards E. and Mauthner, M. (2002) 'Ethics and feminist research: theory and practice', in M. Mauthner, M. Birch, J. Jessop and T. Miller (rds), Ethics in Qualitative Research, London: Sage, pp.14-31.
Galliher, 1-:F. (1982) 'The protection of human subjects: a re-examination of the professional code of ethics', in M. -Bulmer (ed.), Social Research Ethics: An txamination of the Merits of Covert Participant Observation, London: Macmillan.
Gherarrii, s. (1995) Gender, Symbolism and Organizational Cultures, London: Sage.
Glides, V.- and Alldred, P. (2002) 'The ethics ot intention: research as a political tool', in "V Mauthner, M. Birch, 1. Jessop and T. Miller >'t,;-, £."„...• in Qualitative Research, London: Sage, PP.32-52.
Glover, j.:(2001) Humanity: A Moral History of the
twentieth Century, London: Pimlico. Grey, C. and Sinclair, A. (2006) 'Writing differently',
Organisation, 13(3): 443-53. Grey, C. and Willmott, H. (eds) (2005) Critical
Management Studies: A Reader, Oxford: Oxford
-"liversity Press.
Grinyer, A. (2002) 'The anonymity of research participants: assumptions, ethics and practicalities', Social Research Update, issue 36.
Griseri, P. (2002) Management Knowledge: A Critical View, Basingstoke: Palgrave.
Hearn, J. (2004) 'Personal resistance through persistence to organizational resistance through distance' in R. Thomas, A. Mills and J.H. Mills (eds), Identity Politics at Work: Resisting Gender, Gendering Resistance, Abingdon: Routledge, pp.40-63.
Herrera, CD. (2003) 'A clash of methodology and ethics in "undercover" social science', Philosophy of the Social Sciences, 33(3): 351-62.
Hooks, B. (1989) Talking Back: Thinking Feminist, Thinking Black, London: Sheba Feminist Publishers.
Katz, J. (1992) 'The consent principle of the nuremberg code: its significance then and now' in G.J. Annas and MA Grodin (eds), The Nazi Doctors and the Nuremberg Code: Human Rights in Human Experimentation, New York: Oxford University Press.
Ktiit. I .Williamson,E., Goodenough.T.andAshcroft, R. (2002) 'Social Science Gets the Ethics Treatment: Research Governance and Ethical Review', Sociological Research Online,7(4). <http://www.socreson!ine. org.uk/7/4/wil!iamson.html>.
Koro-Ljungberg, M., Gemignani, M„ Winton Brodeur, C. and Kmiec, C. (2007) 'The technologies of normalization and self: thinking about IRBs and extrinsic research ethics with foucaulť, Qualitative Inquiry, 13(8): 1075-94.
Lather, P. (1986) 'Research as praxis', Harvard Educational Review, 56(3): 257-77.
Lather, P. (2004) 'This is your father's paradigm: government intrusion and the case of qualitative research in education', Qualitative Inquiry, 10(1): 15-34.
Lincoln, Y.S. (1995) 'Emerging qualitative criteria',
Qualitative Inquiry, 1: 275-89. Lincoln, Y.S. (2005) 'Institutional review boards and
methodological conservatism', in N.K. Denzin and
Y.S. Lincoln (eds), Sage Handbook of Qualitative
Research, (third edn.), London: Sage. Lincoln, Y.S. and Cannella, G.S. (2004a) 'Dangerous
discourses:   methodological   conservatism and
governmental regimes of truth', Qualitative Inquiry,
10(1): 5-14.
Lincoln, Y.S. and Cannella, G.S. (2004b) 'Qualitative
research, power and the radical right', Qualitative
Inquiry, 10(2): 175-201. Lincoln, Y.S. and Guba, E.G. (1989) 'Ethics: the failure
of positivist science', The Review of Higher Education,
12(3): 221-40.
92
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
Lincoln, Y.S. and Tierney, W.G. (2004) 'Qualitative research and institutional review boards', Qualitative Inquiry, 10(2): 219-34.
Maclntyre, A. (1998) A Short History of Ethics, London: Routledge.
Mason, J. (1996) Qualitative Researching, London: Sage. McLaughlin, D. and Tierney, W. (eds) (1993) Naming
Silenced Lives: Personal Narratives and Processes of
Educational Change, London: Routledge. Nelson, C. (2004) 'The brave new world of research
surveillance', Qualitative Inquiry, 10(2): 207-18. Opie, A. (1992) 'Qualitative research, appropriation of
the "other" and empowerment', Feminist Review,
40(Spring): 52-69. Parker, M. (2002) Against Management, Cambridge:
Polity Press.
Power, (VI. (1997) The Audit Society: Rituals of Verification, Oxford: Oxford University Press.
Prasad, P. (2005) Crafting Qualitative Research: Working in the Postpositivist Traditions, London: M.E. Sharp.
Punch, M. (1994) 'Politics and ethics in qualitative
research', in It Denzin and Y. Lincoln (eds),
Sage Handbook of Qualitative Research (first edn.),
London: Sage, 83-97. Richardson, S. and McMullan, M. (2007) 'Research
Ethics in the UK: what can sociology learn from
health?', Sociology, 41(6): 1115-32. Said, E. (1978) Orientalism: Western Conceptions of the
Orient, London: Penguin. Seidman, S. (1998) Contested Knowledge: Social Theory
in the Postmodern Era, (second edn.), Oxford:
Blackwell.
Sin, C.H. (2005) 'Seeking informed consent: reflections on research practice', Sociology, 39(2): 277-94.
Smith, D (1990) Texts, Facts and Femininity: Exploring the Relations of Ruling, London: Routledge.
Stanley, L. and Wise, S. (1993) Breaking Out Again: Feminist Ontology and Epistemology, Londor Routledge.
Truman, C. (2003) 'Ethics and the Ruling Relations r Research Production', Sociological Research Online, 8(1) <http://www.socresonline.org.Uk/8/1/trumar html>.
van den Hoonard, W.C. (2001) 'Is research-ethics reviev a moral panic?', Canadian Review of Sociology and Anthropology, 38(1): 19-36.
Wax, Mi. (1982) 'Research reciprocity rather than informed consent in fieldwork', in J.E. Siebc (ed,), The Ethics of Social Research: Fieldwon Regulation and Publication, New York: Springei Verlag.
Weems, L. (2006) 'Unsettling politics, locating ethic representations of reciprocity in postpositivist inquiry', Qualitative Inquiry, 12(5): 994-1011.
Wiles, R., Crow, G., Charles, V. and Heath, S (2007) 'Informed Consent and the Research Process: Following Rules or Striking Balances?', Sociologic, Research Online, 12 (2). <http://www.socresonline. org.uk/12/2/wiles.html>.
Wray-Bliss, E. (2002a)'Abstractethics, embodied ethi the strange marriage of Foucault and positivism in LPT', Organization, 9(1): 5-39.
Wray-Bliss, E. (2002b) 'Interpretation - appropriation: (making) an example of labour process theory', Organizational Research Methods, 5("jl 80-103.
Wray-Bliss, E. (2003a) 'Research subjects/reseaKh subjections: the politics and ethics of critical research', Organization, 10(2): 307-25.
Wray-Bliss, E, (2003b) 'Ethical discriminations? representing the reprehensible', Tamara, 2(3): 7-22,
Wray-Bliss, E. (2004) 'The right to respond? iV. monopolisation of voice in CMS', Ephemera, 4(2): 101-20.
6
Rhetoric and Evidence: The Case of Evidence-Based
Management
lark Learmonth
INTRODUCTION
Evidence-based management has become a IMMiciilaily liol topic, now, in the early years i-l ll le twenty-first century. And this chapter i" ii. .line with some of its ambitions; for example, it is intended, in part, to be a i" -II for academics (and practitioners) to use
«e forms of evidence to inform practice. ;:i".\ever, the chapter is also concerned to
vide a critical exploration of some of the ideas about 'evidence' and 'management' that have become associated with evidence-bii^-a management. Or at least, evidence-basedmanagement in its most prominent ■«'•' which is to say the version that piomises applications of science to manage-ui i through which we can: 'hear and act on the facts [in order] to make more informed Md intelligent decisions' (Pfeffer and Sutton, 2006;:T4).
Curiously, however, for a movement that is so 'now', this promise is based on a version of science and evidence that is somewhat dated. For in today's intellectual environment, it is only rather traditional versions of science that fail to question the idea that scientific facts are the products of impersonal, disinterested, and rigorous observations concerning the evidence. This traditional view of science and evidence was most influentially propounded by Robert Merton (1973/1942) for whom science's 'tmth-claims, whatever their source, are to be subjected to preestablished impersonal criteria; [t]he acceptance or rejection of claims entering the list of science is not to depend on the personal or social attributes of their protagonist; [o]bjectivity precludes particularism' (1973/1942: 270; italics in original). Evidence, in this traditional view, can be misunderstood, it can even be falsified or invented; nevertheless, the only judge we
JOB SATISFACTION IN ORGANIZATIONAL RESEARCH 197
12
Job Satisfaction in Organizational Research
Alannah E. Rafferty and Mark A. Griffin
INTRODUCTION
Job satisfaction is one of the most frequently studied aspects of organizational life in the psychological sciences (Cranny et al., 1992; Fisher, 2000; Landy, 1989; Locke, 1969, 1983). Psychologists have argued that the way that an individual feels about work and the meaning that work holds in his or her life is a critical element of the employment experience (e.g., Judge et al., 2001a; Tait et al„ 1989). The degree of interest in job satisfaction is reflected in the number of studies that have been published in the area. Entry of the search term 'job satisfaction' in the psycLIT database reveals well over 5,500 published studies in refereed journals. Importantly, however, as reflected in our analysis of articles published in the Journal of Applied Psychology and the Journal of Organizational Behavior since 2000, relatively few studies have focused on the job satisfaction construct itself.
As noted by Locke (1983), it is impossible to conduct a comprehensive review of the job satisfaction literature in a single chapter. We adopt a psychological perspective on
job satisfaction and focus on methoi issues that have emerged in the h 11 i n i i the last few decades. We begin our i i i 11, exploring why job satisfaction has   ^ i tl subject of such interest. Next, wc l . i<m. classic studies, including the HaU i< Studies (Roethlisberger and Dickson; 1939), J \ Hoppock's (1935) and Schafter's i I'.S study, and Herzberg and colleagues' l1'1" 1959) work. We then define the construct *»\ of job satisfaction, noting some: methodological debates in the area. Next, *J i we discuss the distinction betw< 11 lIi Ml and facet job satisfaction. We then i .in u -the major methodological approi measurement instruments that have I v.1: used when studying satisfaction. Our then turns to recent empirical re: conduct a review of articles published • c 2000 in two major organizational psy< I ■ • -journals, the Journal of Applied * and the Journal of Organizational Finally, we identify some emerging i in the job satisfaction literature, in< u li i' dispositional approaches to job sati: I k i 111 and multilevel research.
IVA„ itudyjob satisfaction?
are:: a number of perspectives that h the: study of job satisfaction. While 'n" cm-w of recent empirical studies reveals 'J| , j, . satisfaction is often treated as k-pinJent variable, some authors argue jj j,ii> satisfaction is a crucial construct, se it mediates relationships between * ikir conditions and organizational and nji. i J 1 1 Oiitco ines (e. g., Dormann and Zapf, •'li;; li.1 .'irehers have been particularly "n L.r«.; cl in exploring relationships between i, [v| :tetion and individual performance. 1 c:. !i ■> action-performance relationship has , vn i it subject of hundreds of studies, and a number of meta-analyses have summarized CijIk of these studies (Bowling, 2007; jrl.ld i ) and Muchinsky, 1985; Judge etal., j;ui-. : city et al., 1984). Landy (1989) 1.-st.. I-' the ongoing search for confirmation i| ire i ik between job satisfaction and job ul 'it usee as the 'Holy Grail' of Indus-irul aii.l Organizational Psychology. Indeed, ist ni.lii.rs have persevered despite evidence ili« l i-i is only a weak relationship between rhc-e •.' i (stmcts. In the most recent meta-analytic review, Bowling (2007) concluded ill,' u satisfaction-performance relationship is largely spurious and is influenced by individuals' dispositional characteristics, 'fcscarchers have also studied other outcomes, m lii I ii* voluntary turnover (e.g., Horn et al., I->U2, 'Alight and Bonett, 2002), absenteeism l' j;, lh. rcnou, 1993), and counterproductive lul i i ii - (e.g., Gottfredson and Holland, 11'1!', "name just a few.
A v.,■ hkI perspective that has motivated ikm ni-1 on job satisfaction is interest in the link between job and life satisfaction. Some researchers have argued that since work is acentral life activity it is likely to mil j individuals' satisfaction with life i'-. Run et al., 1991; Rode, 2004; Tait 1 il. 989). Empirical research indicates that thcie is moderate positive relationship '-'I* -i ii |ob and life satisfaction (r=.44; Tait etal., 19X9). However, as with research looking at satisfaction-performance relationships, ■'-it iocs have suggested that the job-life
satisfaction relationship is influenced by dispositional factors, which are related to both constructs (e.g., Heller et al., 2002; Judge and Hulin, 1993). We examine research exploring the relationship between dispositional characteristics and job satisfaction in greater detail later in this chapter. In the following section, we examine some classic studies that have had an important impact on the methodologies and approaches used when studying job satisfaction.
Historical perspectives
Research exploring job satisfaction began in the 1920s and 1930s with the Hawthorne Studies (Roethlisberger and Dickson, 1939) and work by Hoppock (1935). Although these two sets of studies were conducted at roughly the same time, the Hawthorne Studies have had the most enduring influence on job satisfaction research (Locke, 1983). We also describe studies conducted by Schaffer (1953) and Herzberg and his colleagues (1957,1959).
The Hawthorne Studies
The Hawthorne Studies were a series of studies conducted from 1924 to 1932, which provided the impetus for research focused on psychological and social factors in the workplace and are generally acknowledged as one of the starting points of the Human Relations school of thought (Judge and Larsen, 2001; Locke, 1969,1983). This series of studies was initially designed to study the effects of work conditions, such as the timing of rest breaks, the length of the working day and pay rates, and on worker fatigue. However, the emphasis soon shifted to the study of 'attitudes,' as workers did not respond in expected ways to systematic changes in working conditions. Roethlisberger and Dickson's (1939) use of the term 'attitudes' encompasses operators' attitudes to management, their moods in the workplace, and the reactions to their supervisor and the experimental team.
We draw on one of the early studies, the Relay Assembly Room study, to provide an example of how 'attitudes' were assessed in the Hawthorne Studies. The Relay
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
Assembly Room study involved a small group of female workers who were isolated in a room from the rest of employees. In one of the 13 experimental periods, a survey consisting of eight questions was administered to operators. This initial questionnaire assessed whether operators had experienced any changes in their 'mental health habits and mental attitudes.' An additional questionnaire was then administered that focused on specific issues in the test room that could be responsible for these positive responses to the test room. Twenty-one questions were asked of operators, who suggested that they liked the smaller group in the test room, liked having 'no bosses,' enjoyed the 'freedom' available in the test room, and 'the way they were treated.' In summary, in the Relay Assembly Test Room study researchers were beginning to focus on individuals' cognitive and affective responses to their job and were interested in determining the particular aspects or facets of work that were responsible for these responses.
The Hoppock Study
Hoppock (1935) published the first intensive study of job satisfaction, based on interviews with adults in a small town and 500 schoolteachers. Hoppock's research was driven by two underlying questions (Landy, 1989). First, are workers happy? Second, are some workers happier than others are? Hoppock used verbal self-reports collected via interviews to assess job satisfaction. When summarizing his findings, he identified many factors that could influence job satisfaction, including worker fatigue, job conditions such as monotony, supervision, and autonomy. Results indicated that only 12% of workers could be classified as dissatisfied and that occupational groupings influenced satisfaction with higher status occupational groups reporting higher satisfaction. In summary, Hoppock began to systematically analyze job satisfaction and identified a range of factors contributing to job satisfaction that are still studied today (e.g., Rose, 2003).
Schaffer (1953)
Schaffer (1953) argued that whatever ps, j logical mechanisms make people satisfied or' dissatisfied in life also make them : n.hl or dissatisfied in their work. He su;'.n_.|Lr that the amount of dissatisfaction expeiiencc*-by an individual is determined by; a) the strength of an individual's needs or drives-and b) the extent to which an individual cay perceive and use opportunities in the situation for the satisfaction of those needs. Schm ilc, identified 12 needs that need to be met dnd overall satisfaction was said to be prcui j by the two most important needs held by a person. If the individual's two most nupouant needs are metby the job then they a
Based on his analysis, Schaffer (1953) developed a questionnaire measure ■■ :■—„ the strength of each of the 12 iv-il.. hi-degree to which each of the needs v,,-ii satisfied by the individual's job, ami h individual's overall satisfaction. The ovoial,,. satisfaction scale consisted of three items, tweu of which were based on the work 11 ll< (1935), while the third item was an open-ended question. In summary, Schlatter's woik acknowledged that individual differences arc an important antecedent of job sail ...--ii i ,. topic that has recently received considc.. ult attention.
Herzberg: The two-factor theory By the late 1950s, Herzberg and colleagues (1957) were able to review the avarlab .. il satisfaction literature in a systei i. ■!■ ii, They concluded that there was sufficient evidence to confirm a relationship bel t satisfaction and outcomes such as performance, turnover, absence, and psychosoi i. m illnesses. In addition, Herzberg et al. (i"""'-concluded that satisfaction and dissatisf.ii i « were two distinctly different phenomena Herzberg et al. (1959) conducted a follow-up study to explore this idea, and they interviewed 203 accountants and engineers. Herzberg et al. (1959) concluded that specific factors were related to feeling positive ahuiil the job, including: achievement and recognition, the work itself, responsibility, adv. 'i-1 ment or promotion opportunities, and s j ■■ ■
I
- t
m
If
11
i
tOB SATISFACTION IN ORGANIZATIONAL RESEARCH
j contrast, dissatisfaction with a job was ,,i ii another set of factors, including: ■ rnpwi'y P°''°'es an^ administration, tech-, j. if supervision, salary, relationships with
!.....[s, and working conditions.
On the basis of this study, Herzberg and his coIltt>gues (1959) proposed the two-factor .in. ii-, >i the motivator-hygiene theory and a ..m.ihi of empirical tests of the two-factor iirjil >'Bre conducted (e.g., Ewen et al., pi! ii'aen, 1966). Ewen et al. tested the theory using a sample of 793 male employees hiii i. i.tfige of different jobs and failed to suppud Ihe theory. Subsequent research also failed to support the two-factor theory and n-.-j'Ji on- this topic declined dramatically. Un .Ii11 389) suggests that while this theory i\ ii.■! i )ly reasonable at the descriptive jgvcl, it is- not an explanation of how job ^jt .1 il ' on: develops and of its relationships a-ii(i nth t.'conslructs. Despite the limitations of the theory, it stimulated research into job k, i-r i tion and prompted new directions of investigation.
Defining job satisfaction
The definition of job satisfaction has U : . abated throughout its history. Many researchers have, defined job satisfaction as ai emotional reaction to the job (e.g., Cranny
"et ah, 1992; Locke, 1969; Spector, 1997). Cm ii yand colleagues (p. 1) state that 'there ls ;,c icral agreement that job satisfaction ■s . r affective (that is, emotional) reaction to a job.that results from the incumbent's comparison of actual outcomes with those that ■ li--.ned (expected, deserved, and so on).' Locke; (1969: 316) defined job satisfaction
-as 'the:pleasurable emotional state resulting lii.ii the appraisal of one's job as achieving "r I Militating the achievement of one's job '■ iliies,' Although job satisfaction is usually described as an affective response, it is usually i n- sirred as an evaluative assessment of job j Ii ides compared to an individual's internal ' - -Sternal standards (Fisher, 2000; Niklas and I1' ■minn. 2005: Pekrun and Frese, 1992). 1 re distinction between affect and evalu-- n.ihas been addressed by research in the
social psychology of attitudes, and recently, scholars have begun to draw explicitly on the attitudes literature when defining job satisfaction (e.g., Fisher, 2000). Fisher identified two components of job attitudes, an affective or emotional component and a cognitive component. Weiss (2002) identified three distinct, but related aspects, of job satisfaction. He argued that job satisfaction incorporates an individual's evaluation of their job, their beliefs about their job, and their affective experiences on the job. Importantly, however, Weiss identifies evaluation as the central component of job satisfaction and in line with this statement, he defined job satisfaction as 'a positive (or negative) evaluative judgment one makes about one's job or situation' (p, 175). An important consequence of this perspective is that affect and beliefs are understood to be different causes of job satisfaction.
The definitional debates in the job satisfaction literature have important implications for how we measure job satisfaction and the most appropriate research designs to use. Weiss (2002) argues that current job satisfaction measures are limited, because they do not enable researchers to assess the three component model of job satisfaction that he identified. He argues that since affective states and beliefs structures have important behavioral and cognitive consequences, then they are likely to have added value over the cognitive evaluations, which have been the focus of measurement efforts to date.
Some empirical research has explored the extent to which existing job satisfaction measures capture cognitive versus affective information. Brief and Roberson (1989) conducted a study with 144 students, and explored the extent to which the Job Descriptive Index (JD1; Smith et al., 1969), the Minnesota Satisfaction Questionnaire (MSQ; Weiss et al., 1967, and the female Faces Scale (Dunham and Herman, 1975) captured cognitions and positive and negative affect. The different measures of job satisfaction varied in the extent to which they captured the affective and cognitive aspects of job satisfaction. Brief and Roberson suggest their
200
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
JOB SATISFACTION IN ORGANIZATIONAL RESEARCH
201
results as suggesting that the Faces measure is the 'most balanced' of the scales as it displayed relationships with cognitions and positive and negative affect.
Similarly, Fisher (2000) argued that it is useful to think of existing job satisfaction measures as lying on a continuum assessing the cognitive component of job attitudes to assessing a combination of cognitive and affective components. Based on an empirical study that used an experience sampling methodology to collect data from 121 employees over a two-week period, Fisher concluded thatmostjob satisfaction measures only assess affect to a limited degree. In contrast to other researchers. Judge and Larsen (2001) argued that it is not productive to classify job satisfaction measures as either cognitive or affective. They argued that there is no need to measure new, affectively laden job satisfaction measures or to supplement or replace job satisfaction measures with measures of affect. Judge and Larsen argue that while cognition and affect can help us better understand the nature of job satisfaction, separate measures of these aspects of job satisfaction are not likely to prove useful.
Niklas and Dormann (2005) argue that findings suggesting that existing job satisfaction measures primarily tap the cognitive components of satisfaction have important implications. While cognitions are relatively stable over time, affective reactions are unstable, so different methodologies are required to assess cognitions and affect. In addition, while the issue of time frame is relatively unimportant when measuring cognitions, this issue is critical when assessing affect, as retrospective reports of affect tend to be biased. As such, there is a need to assess affect on a time-contingent basis. To date, however, researchers studying relationships between trait and state affect and job satisfaction have retrospectively measured affect over the last weeks, months, or years (Niklas and Dormann, 2005). This is a critical limitation of existing approaches, because people tend to overestimate the intensity and frequency of positive and negative affect in retrospective measures, and retrospectively reported affect
is likely to be affected by, current (Fisher, 2002). In addition, negative a more easily recalled than positive affei Baumeister et al., 2001).
In summary, there is still consii debate concerning how to define jol faction. This debate has important rr ological implications. Our review si that the different definitions of job satis require authors to develop different mi of satisfaction. For example, if we Lh , job satisfaction as an attitude fhatc of cognitive evaluations, affect, and: then theorists argue that there mi to develop measures of affect and-. »■ iL| independent of evaluations. Niklas an mann's (2005) work highlight the Tail r[,j| affective reactions are unstable, which nnvh, that different methodologies are requi 11 |<j assess cognitions and affect. Researche -recently begun to address the methodol iein, issues associated with assessing affec ,„i, studying relationships between di.>tpt» iiitnu' factors and job satisfaction.
Facet satisfaction versus overall job satisfaction
Most scholars recognize that job satisf is a global concept that also con various facets (Ironson et al., 1989;; h:\p. and Larsen, 2001; Smith, 1992). Rosea have examined relationships between a ■■'u job satisfaction and facet satisfaction (Vii'h' *i 1992; Weiss, 2002), the impetus forwhitft* lies with Locke (1969). He suggests lli.il   < ■ overall job satisfaction is the sumc evaluations of the elements of whk 1   . r-isfaction is composed. Global or: i   i II job satisfaction scales ask an indivk combine his or her reactions to a job 11 •» overall or integrated response (e.g., all things considered, how satisfied are you wit! job?). In contrast, facet satisfaction. scales are designed to examine specific aspects of . job within the more general domain (Iroiisott et al., 1989). The most typical calegor . of facets was proposed by Smith et al. (l1''1'0 who identified five facets, including M'i _ . , promotion, co-workers, supervision, and the
If. Another common distinction is , ! ej[frinsic and intrinsic satisfaction. 1,1 li f>i on with pay and promotions has been V, len'd forms of extrinsic satisfaction. "' ist, satisfaction with co-workers, jp.,. i.. irSi and the work itself are considered 1  ™ fi i ■■■intrinsic satisfaction (Judge and
jV™ ^01). 11 nil ical studies have indicated that global ie isfi."^ «fJ0'1 satisfaction and facet satisfac-ion riLi'tircs are only weakly to moderately
eil.-< l--g- AldaS and Brief> 1978)- These Ibid us 'ave insulted in a debate as to why clol 'I i leasures of job satisfaction are not
•it to the sum of facet satisfactions
,f c Jli.-hhouse and Becker, 1993; Scarpello niil ('.iii'l'bcll, 1983). Scarpello and Campbell ■. fwiii 'i- d relationships between single item 5 „ki!\ I in sasures of job satisfaction and facet nf,iM * s of satisfaction taken from the short-form of me MSQ (Weiss et al., 1967). These j,.,,,,, oacluded that it is not appropriate to i«e .he sum of the facet measures as admeasure of overall job satisfaction as global measures of job satisfaction appear [,, i priii- information relating to satisfaction vuli .'C npalional choice, life off the job, in ailJit i n to capturing information about job ijtisl.i I on.As such, global evaluations of job «.a! sl-i.tion arc more complex than the sum otitic job satisfaction facets that are typically measiued. In summary, empirical research linili%'s suggest that global satisfaction and "'■'lacf s ilisl'aclion measures assess distinctly '^liferent issues and are not equivalent mea-iun^ .'I pb satisfaction.
I'n - facet satisfaction and global job satis-la.lu 11 istinction also becomes problematic il >'..- . i insider Weiss's (2002) discussion 'ir latisfaction. In particular, if job xli-lu ion is a cognitive evaluation then Wri",, jues that the facets of job satisfaction ?ie -m distinguishable from the overall job satisfaction evaluation. Rather, Weiss that there 'are only discriminate <*ix,- 11 the work environment that we evaluate' (p. 186). The basic argument being ■Top "i. -I by Weiss is that while we may c,'i'i i'.e issues such as how satisfied we ■n-'.With our supervisor or our pay, these
evaluations combine to create an overall evaluation of how satisfied we are in the workplace. This argument suggests that what have been called facets of job satisfaction are simply evaluations that individuals make about their work environment. In summary, an analysis of existing research suggests that global measures of job satisfaction appear to tap a broad domain of satisfaction that incorporates life satisfaction. In contrast, while facet job satisfaction measures are designed to assess a broad domain they appear to assess a more limited and focused aspect of job satisfaction involving specific evaluations. In addition, the recent emphasis on job satisfaction as an evaluation that is caused by affect and beliefs also calls into question die facet versus global satisfaction distinction.
Research design issues in the study of job satisfaction
The predominant methodological approach that has been used when studying job satisfaction is the field study, with authors using both cross-sectional and longitudinal approaches. While early research on job satisfaction used qualitative data collection approaches such as interviews, very little recent research has adopted diis approach. In addition, a number of quasi-experimental field studies have also been conducted that assess job satisfaction. However, job satisfaction has not been the key variable of interest in these studies. More recently, with increasing interest in the role of affect in job satisfaction, a number of studies have focused on intraindividual processes. These studies have used experience sampling methodologies. In the following section, we focus on exploring early research that used interviews, quasi-experimental studies and studies using experience methodologies to study intraindividual processes.
Qualitative studies
Qualitative studies are typically conducted in applied settings are and are generally broadly focused studies that are exploratory in that they are not driven by explicit hypotheses.
JOB SATISFACTION IN ORGANIZATIONAL RESEARCH
203
These studies are designed to gather data on individual experience and meaning, and many different methods such as interviews, conversations, observations, personal experiences, and textual analysis canbeusedto collect data. Qualitative studies emphasize depth of information and analysis as opposed to breadth of information and analysis (Kidd, 2002). Next, we explore studies that have adopted qualitative approaches when exploring job satisfaction.
A number of early researchers used techniques such as interviews in order to develop an understanding of the nature and antecedents of job satisfaction (e.g., Hoppock, 1935; Herzberg et al., 1959). For example, Hoppock interviewed adults in a small town and 500 schoolteachers in order to explore whether workers were happy in their jobs and whether some individuals were happier than others. However, a search of recent studies using PsycLfT revealed that very few authors have used qualitative approaches to study job satisfaction. In those studies that we did identify that used some type of qualitative approach, interviews were generally used as a supplement to quantitative analyses rather than the primary data collection approach (e.g., Lyons and O'Brien, 2006). For example, Lyons and O'Brien conducted a study in order to explore the determinants of job satisfaction for a sample of African-American employees. These authors collected data using surveys and also collected data using an interview to administer an open-ended question ('what makes you or would make you a satisfied employee?'). Lyons and O'Brien concluded that the interview data was consistent with the quantitative results.
We identified one study that did use an interview as the primary data collection tool. Bussing and Bissels (1998) conducted an exploratory study to test a model of work satisfaction and conducted semistruc-tured interviews with 46 German nurses. Participants were asked to give a detailed description of issues such as their working situation and their overall satisfaction or dissatisfaction in the workplace. Bussing and Bissels argue that the use of qualitative
analysis allowed them to identify snliv.mi,, M different qualities of work satisfaction |> addition, the authors argue that they v ■„ i, to identify apparently inconsistent that reflected the contradictory t . ii 11. n and pressures placed on nurs     n .: workplace. In summary, we iden h, ■ ... few recent studies that have used :i 11 methods when studying job satisfaction. Onr potential explanation for the lack of such studies is that the amount of qualil in,i ,Vl, published in APA journals in the last   ■ ■,,.,„. is quite small (Marchel and Ow< i . ju; / Despite this, however, Bussing and I.: ,[-.. (1998) work suggests that there an   i).. f „, benefits to be gained from using i. i ih methods when studying job satisfai
Quasi-experimental field studies
A number of researchers have couducted quasi-experimental studies where job satisfaction has been included as a s interest (e.g., Lam and Schaubrocck, 2()O0<i, 2000b; Logan and Ganster, 200-., S.- v, 1999; Workman and Bommer, 2004). A quasi-experiment is an 'experirr.-.: '| i\i hafvejs treatments, outcome mesiMircs, and experimental units, butdo[es] not ut.' i.n I h , assignment to create the compari .■■ i> I n which treatment-caused change is infi-.i.-i' (Cook and Campbell, 1979: 6). Kallier, comparisons occur between a nonequivatcnt control group and an experimental group or groups who differ in many ways 4 1 from the treatment of interest. < '1 r I-.. challenges of using a quasi-experin ■ I ■ approach involves deterrrrfriing what a li -are attributable to the treatment of interest versus those effects due to the initial noncomparability of the experimental and control group. However, quasi-experinicnHl studies are important, because they allow researchers to study causal relationshi| 1 a complex field setting. In addition, !hes.e studies also allow us to determine whether results obtained in the laboratory generalise to real world environments. In the case of job satisfaction and many other organizational constructs, it is essential that we study these variables in situ using rigorous methods that
II     i- to consider issues of causality. In next' section, we examine one quasi-1) Liital study that incorporated job
'.11
Seiner conducted   a pretest-
.. nonequivalent    control group
,i-L>;jerimental   design,   which was ■iito test the effectiveness of a formal ut nil 'i gpr°gTam >n a Fortune 100 company iv    irie year period. Job satisfaction was ^ hi 1 come variable in this study. Results ,,..1.;,;t-.(| th.:t  the facilitated mentoring „1 ihad a significant effect on job s irlN ... ion. One year after the initiation 1 In vvigraiu, employees who had been ™-i r, 1, .reported   significantly higher ..,1- Mlisfaction  than their nonmentored .111 it.1 parts. In summary, we could only in 11. a small number of quasi-experimental ,li ,n . 1 lat assessed job satisfaction as one of many outcomes of interest. Wbile quasi-jv>-i Mentation has been discussed as a pi,...- 111itool for drawing casual inferences p. nr.i ud settings (Cook and Campbell, ■ "'■I ( iok et al., 1992), researchers do not jj 1 in have fully utilized this approach v.i i.'-i - udying job satisfaction. We argue thai one issue that researchers can focus on .«-.-1 -udying job satisfaction is the use of -quasi-experimental research.
'i-i'.ji'i bvidual processes The majority of job satisfaction research has > -L •». I in differences between individuals. In recem years, however, increasing attention iiji k-.i paid to the way individuals' job 1.'i-huon changes over time. Although e 11 ■ 1 search had investigated the stability "I sj si action (e.g., the genetic basis of ■ih - in faction; see Arvey et al., 1989), little attention had been paid to the fluctuations in satisfaction from month to month, dr. 11 lay, or within even shorter time 1 'ii-uK The development of diary studies J 11 l.i experience sampling methodologies has encouraged a better understanding of 1 1 11 ctuations within individuals. The traditional cross-sectional, between-subjects •ipLtoaen used in job satisfaction research assumes that constructs are stable over time
and that variations around the average level of a variable are random (flies and Judge, 2002). Experience sampling methods allowed the possibility that this wi thin-person variance is systematic and meaningful. From this perspective, job satisfaction might develop over time, change in response to transient events, be moderated by more stable factors, and might itself influence diverse within-person factors such as insomnia for example. We discuss one study that has explored intraindividual processes in the following section.
As described by flies and Judge (2004) experience sampling methodologies ask individuals to report on their momentary feelings or their subjective feeling states, and/or to report on their current physiological state. This approach has been said to be beneficial, because it eliminates error associated with selective recall processes. An example of what experience methodology involves can be seen in a study conducted by Hies and Judge. These researchers conducted a study involving 33 participants from two state universities in the United States. In the first phase of data collection, participants completed a measure of trait affectivity and asked a significant other to complete the same measure. In Phase 2, one week later, participants reported their momentary mood and job satisfaction three times a day, for two weeks, using a diary method. That is, participants were asked to record their responses to survey questions assessing mood and job satisfaction at regular intervals throughout the day. The measure of job satisfaction collected through this approach was called 'experience-sampled'job satisfaction. In the final phase of the study, Phase 3, which occurred two months after Phase 2, participants responded to an overall job satisfaction survey and reported on their beliefs about their job. The results of this study revealed that experience-sampled job satisfaction was associated with the overall job satisfaction measure collected in Phase 3. In summary, a number of research studies have studied intraindividual processes and job satisfaction. This methodology offers an
204
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
JOB SATISFACTION IN ORGANIZATIONAL RESEARCH
205
important means by which to understand the impact of affect on job satisfaction.
Measurement of job satisfaction
A number of different techniques have been adopted when studying job satisfaction. By far the most common approach involves the use of questionnaires. A variety of questionnaire measures have been developed overthe last 70 years. Some of the most commonly used and well-validated measures are the JDI (Smith et a!., 1969) and the MSQ (Weiss et al., 1967). Two other measures that are commonly used when assessing job satisfaction include the Michigan Organizational Assessment Questionnaire (MOA; Cammann, Fichman, Jenkins, and Klesh, 1983), and Brayfield and Rothe's (1951) Overall Job Satisfaction Measure. In the next section, we discuss these measures in some detail.
Questionnaire measures Job Descriptive Index The JDI was developed by Smith et al. (1969) and the original scale consisted of 72 items that assessed five facets of job satisfaction, including: satisfaction with work itself, pay, promotion opportunities, supervision, and co-workers. Roznowski (1989) assessed the JDI and made a number of changes to the measure. In particular, modifications were made to reflect issues such as changes in technology that have occurred since the original scale was developed. Roznowski demonstrated that the items loaded on five factors and that the scales have good reliability, which has been supported by other researchers (e.g., Judge andHuiin, 1993).
Minnesota Satisfaction Questionnaire The 'long-form' of the MSQ consists of 100 items that assess 20 subscales, such as individuals' satisfaction with the extent to which they get to use their abilities on the job, achievement, advancement, moral values, and so on. Twenty of the items in the MSQ have been used to assess general job satisfaction and form the 'short-form' of the MSQ. The short-form consists of 12 questions assessing
intrinsic satisfaction. Intrinsic sali-i,^ items measure issues such as accompli 1 obtained from one's job and opp» r M] ( to use abilities on one's job. I   i,[ (, |if assess extrinsic satisfaction. Es.inpi^ issues assessed in the extrinsic sc i   m_| ■ satisfaction with promotion opporti ui i , _ ,. satisfaction with pay. The short-1 ,,| MSQ has been shown to have gooi11 i, iul • (Mathieu, 1991; Wong et al„ 1998),.:
Michigan Organizational Assessment Qur. tionnaire   Cammannetal. (1983)< a measure of overall job satisfai I m they developed the Michigan Org m , t„ , i Assessment Questionnaire. The : i . ■ u. ■ ■ scale consists of three items such a: all, I am satisfied with my job.' '!h had demonstrated adequate reliabihii McFarlin and Sweeney, 1992).
Brayfield and Rothe's (1951) Oveicl Job Satisfaction Measure Brayfield m. k, (1951) developed an 18-item mi job satisfaction. However, a i iviln ■ i researchers have adopted a 6-item 1 .isim of this measure and this measure h >. ;l, played adequate reliability (e.g., Ai\i ■ I rUi and Luk, 1999). The 18-item mi.-iv lis also demonstrated adequate reliabi I :; ilMI , Schriesheim and Williams, 1999).
Graphic measures
Faces Scales Kunin (1955) dew | .--I 'it-General Faces Scale because he slated lhat attitude surveys, which translate Vil is feelings into another man's words' ( ' result in distortion of meaning. To address this issue, Kunin conducted research to establish scale values for two series of faces that varied on a continuum from happy to unhappy. Respondents are asked to circle the h.'t that reflects the way that they arc Jew i ie about their job. Dunham and Herman (I1 developed a female version of the Im-scale, and demonstrated that the male j iJ female versions can be used interchangeably to measure job satisfaction. The Faces S-.uV-have been used by a number of researi I i (e.g., Dunham, Smith, and Blackburn. I"7
,11 iMitnann, 2005; Rousseau, 1978). 1^ .uch'as Fisher (2000) and Niklas p   , j in argue that the faces scales are '"     \iv iaden measures, as they found that ltle^      . us h as assessed by the General i'^ s |    ; strongly determined by state
o^t^rm'tel research
i -j- lie range and focus of recent . ,t s ■." .in research, we conducted a ^ v ,,fJ studies published since 2000 in ![,. . ial of Applied Psychology {JAP) ■ | ,1 ui' limrnal of Organizational Behavior (/(//(■ Wc identified 84 studies assessing „ h s n-.li- ■■■ n wi'n n:fty of mese empirical .li d „- published in JAP, while 34 studies 'wf pdlished in JOB. In the following c tjun ■'.-'' oul line some of the method-,1^.^,1 i r.u.icteristics of these studies. Job Ml.stJs lion was treated as a dependent vj.mI'' 'it the majority of these studies. Fpc-.i i • Hy.i» 41 of the 50 studies published ii- li'" acd 29 °f me 34 studies published u Ji 'IK job satisfaction was treated as a dependent variable. Only nine JAP studies and four JOB studies treated job satisfaction at, .in independent variable. Studies where 'nl si iJ i m. acted as a mediator or moderator ■stie i ii luded in the independent variable sit Jy ,ni.nt. Researchers in two of the major
erf ii i.....onal psychology journals appear to
ii ■ i satis taction primarily as a broad uiils- i il measure and have paid less attention to the meaning of the construct and its impact en ui i-i outcomes.
N.-i, we summarized the measures that ■sii." i ,-d to assess job satisfaction in JAP .iiij /('A since 2000. Analysis revealed that !im i-l i -fused either the Michigan Organizational Assessment Questionnaire (Cammann fl ■'   -H3) or Brayfield and Rothe's (1951) ^i ■ The next most commonly used nv. ni . was the JDI (Smith et al., 1969), ".in.a -v s used six times. The MSQ (Weiss >-'l   1.   967) was used in four of the t-iLN   Importantly, however, 13 studies ii--.I in.-.sures specifically designed for the i^scattii: or did not clearly identify where " job satisfaction measure was obtained
from. Another notable feature of the measures used was that over 40 separate measures of satisfaction were used in the 84 studies. This result indicates that there is little consistency in the use of job satisfaction measures.
Finally, we reviewed the research methodology used when conducting job satisfaction studies in JAP and JOB. Our analysis suggests that both cross-sectional and longitudinal field studies were used. Of the studies published in JOB, 26 adopted a cross-sectional survey approach while seven studies used a longitudinal survey methodology. One study (Workman and Bommer, 2004) used a quasi-experimental methodology when collecting data in the field. There was a very even split between cross-sectional survey studies and longitudinal survey studies in JAP.
Emerging issues
Our review of the job satisfaction literature reveals that interest in this construct has not waned in the 80 years since its inception. Indeed, researchers have continued to study questions concerning the relationship between job satisfaction and performance relationship despite discouraging results (e.g., Bowling, 2007). Recent theoretical developments regarding the dispositional influences on job satisfaction and methodological advancements concerning the measurement of the affective aspects of job satisfaction, suggest that interest in satisfaction will continue to grow. Our review of recent job satisfaction literature reveals a number of emerging trends. In particular, increasing attention has been devoted to exploring the role of dispositional factors as antecedents of job satisfaction and as third variables responsible for job satisfaction-outcome relationships. A second theoretical issue that has emerged in recent research are multilevel studies of job satisfaction. Next, we explore these issues in some detail.
The role of dispositional factors
In the past, cognitive judgement approaches
(e.g., Locke's (1976) needs-values conflicts
206
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
JOE SATISFACTION IN ORGANIZATIONAL RESEARCH
207
approach) and social influence approaches (e.g., Salancik and Pfeffer, 1978) have dominated the field as explanations for the causes of job satisfaction. In the past two decades, however, increasing attention has focused on the dispositional bases of job satisfaction (e.g., Staw et al., 1986; Staw and Ross, 1985; Weiss, 2002). Staw and Ross stimulated interest in this perspective when they argued that previous research had overestimated the influence of the situation on job satisfaction and underestimated the impact of an individual's dispositional characteristics. The dispositional approach suggests that a person's job satisfaction reflects their tendency to feel good or bad about life in general. This tendency to feel good or bad is independent of the job itself and its positive or negative features (Weiss and Cropanzano, 1996).
Researchers interested in the satisfaction-performance relationship have begun to explore the role of dispositional characteristics as an explanation for the relationship between these two constructs. Bowling (2007) concluded that the relationship between satisfaction and performance is largely spurious. Results of this meta-analysis revealed path coefficients of 0.09 to 0.23 between satisfaction and performance after personality was controlled. Judge et al. (2002) conducted a meta-analytic review of relationships between job satisfaction and the 'Big 5' personality traits (Goldberg, 1990). This review indicated that neuroticism, extraversion, and conscientiousness, displayed moderate relationships with job satisfaction. Neuroticism was most strongly associated (negatively) with job satisfaction, followed by conscientiousness, which displayed a positive association with job satisfaction. However, the 80% credibility interval for conscientiousness included zero, suggesting that this result may not be replicable. Other dispositional characteristics that have been the subject of considerable attention in recent times include trait affec-tivity (Thorensen et al., 2003), and core self-evaluations (Judge et al., 2005; Judge and Bono, 2001). Below, we examine research on positive and negative affectivity and then we
examine the research addressing the role of core self-evaluations.
Positive affectivity and negative affectivity
A great deal of attention has been dii... v. towards examining relationships belwceii positive affectivity (PA) and negative aflcctiv-ity (NA) and job satisfaction (e.g., Thoi.-et al., 2003). People high on positive affectivity tend to be sociable, lively. iU1(j are often in a positive mood. In contrast people who are high on negative affei I > i, tend to be more distressed, unhappy, am] irritable (Watson et al., 1988). \ 11 n of theoretical models have been di-> sloped to explain the relationship belv.ri job satisfaction and trait affectivity.:. and Cropanzano (1996) developed Affective Events Theory (AET), which fo(.u- s u affective experiences at work as prr \ i i.il drivers of affective reactions, whk I 'i ilui n. attitudes and behaviors. This model in | iL. that job events mediate relationships bel^ ■ .n dispositional affectivity and job ,< i ■ i inn Judge and Larsen (2001) proposed the stimulus-organism-response (SOR) mm I The basis of this model is that PA and NA influence individuals' responses and judgements by moderating relationships beli'i 'n a stimulus and the organism. In addition, this model argues that affect mediate; i I tionships between an organism and tuck responses. Finally, Forgas and George (2001) proposed the affect infusion model (AM). This approach suggests that indivii kli attitudes to their job are partially a function of the affect that 'infuses' their cognitive processing when they are forming evalu; i n nia of objects. As such, the AIM is a direct £ I i - t, model.
A number of empirical studies have explored relationships between trait and -I 'f affect and job satisfaction. These s n involve intraindividual processes desu I J previously in this chapter. For example I '<■' and Judge (2002) examined the withi n- pc " ■■ relationship between mood (state affecl; n" ' job satisfaction, and also examined the]-:
ct,«>nality in moderating these relationships. " miity-seven employees completed mood "|. | job satisfaction surveys at four different 11 ,;s a day for four weeks. The results of i,is study indicate that 36% of the total ,.. uin job satisfaction was due to within-nersorl factors and mood explained 29% of I is within-person variance. In addition, mood and job satisfaction were related both within and acToss individuals. Results also indicated that within-person variability in mood was x'.r.'d 1° within-person variability in job satisfaction and variability in mood and job satisfaction was predicted by neuroticism. |. ii illy, individuals who scored higher on ii'uroticism were more likely to have their mood affect their job satisfaction.
■\ 11 mber of meta-analyses have summa-n.,-tl i.-tationships between trait affectivity nil sue ■■affect and job satisfaction (e.g., ■| ii i -.i'.en et al., 2003). In the most recent review, Thorensen et al. reported an estimated in I.'1 t.'il mean population correlation of iLlween trait PA and satisfaction, while state I'A bad an estimated mean population ui ii* .iTion of 0.38 with satisfaction. TraitNA .idanestimatedmean population correlation nt ('.''7 with job satisfaction while state NA had an estimated mean population correlation of -0.3 lv- However, Niklas and Dormann (2005) argue that the conclusion that there is a relationship between trait affectivity and ■■i it: affect and job satisfaction is problematic i . ■ imber of reasons. While results have I"j n iilcrpreted as suggesting that state affect l? a result: of job events, which then causes ■I' ci'i^f action, these relationships are open to interpretation, because trait affectivity was ■"■r ■ in Irolled when estimating relationships 'Mi' CLniState affect and job satisfaction. As m. . h, I he correlation between state affect and job satisfaction may in fact be spurious. In ■ 'lilifion, Niklas and Dormann argue that there ' ■ ■ need to determine whether state affect only "ii ueiiccs job satisfaction at the time when 1 ■'-.measured or whether state affect has a I •■I ng;inHuence on satisfaction. Mklas and Dormann (2005) conducted I'vo week diary study with 91 employ-a" ■.'« address these concerns. Participants
completed questionnaires four times a day. Facet job satisfaction was assessed as was general satisfaction using the Faces Scales. Trait job satisfaction and trait affect were measured and participants were asked to indicate how satisfied they usually are with work and how they usually feel. When state job satisfaction and state affect were assessed, participants were asked to indicate their current attitude toward their job and their current affect. The results of this study revealed that within-person aggregated state job satisfaction was strongly correlated with generalized job satisfaction. In addition, state affect and state job satisfaction were related even when trait affect and generalized satisfaction are controlled. Niklas and Dormann suggest that this study provides evidence that the relationship between state affect and job satisfaction is not due to the influence of a third factor, trait affectivity. Rather, they conclude that positive or negative work events lead to state affect, which has an impact on job satisfaction above and beyond that of affective disposition. Importantly, Niklas and Dormann state that job satisfaction depended more on affect elicited by current affairs than on personality dispositions. Because trait affectivity was controlled, Niklas and Dormann argue that state affect does not merely represent a mediator between trait affect and satisfaction. Rather, there must be situational factors that elicit state affect, which lead to changes in job satisfaction.
Core self-evaluations (Judge et al., 2005) defined core self-evaluations as fundamental, subconscious conclusions that individuals reach about themselves, other people, and the world. Four core self-evaluations - self-esteem, generalized self-efficacy, neuroticism, and locus of control - have been identified. Core self-evaluations can help elucidate the psychological processes underlying the dispositional source of job satisfaction. A growing number of empirical studies have explored the influence of core self-evaluations on job satisfaction. For example, Judge et al. (1998) conducted an empirical study and reported that core self-evaluations have consistent effects on job satisfaction,
208
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
JOE SATISFACTION SN ORGANIZATIONAL RESEARCH
209
independent of the attributes of the job itself. Judge and his colleagues state that results indicated that core self-evaluations had an indirect influence on job satisfaction, via their impact on individuals' perceptions of their job's attributes. These perceptions then influence job satisfaction.
Despite the progress that has been made in accounting for dispositional influences on job satisfaction, Johnson, Rosen, and Levy (in press) identified a number of theoretical and measurement issues with the core-self evaluation construct that are of particular concern. We focus on just two of the issues identified. First, Johnson et al. argue that, to date, core self-evaluation has been discussed as a superordinate construct with effects indicators, such that this construct is responsible for levels of self-esteem, generalized self-efficacy, locus of control, and neuroticism. In such models, causality flows from the construct to the traits, which are manifestations of the construct. However, Johnson et al. argue that core self-evaluation is more accurately modeled as an aggregate construct. In an aggregate model, core self-evaluation would not cause the traits but be caused by them. This is a critical distinction, because if core self-evaluation is modeled as a superordinate construct, then the likelihood of uncovering true effects is reduced. Second, the measurement of core self-evaluation does not align with how it is conceptualized. The use of summed scores is inappropriate, as an aggregate construct cannot be adequately represented by a summed score.
Multi-level studies of job satisfaction Job satisfaction has also been incorporated in multilevel processes involving teams and organizations. Many permutations of relationship are possible when complex mediational processes are considered at multiple levels of analysis. We consider two broad and important types of multilevel problems that involve job satisfaction. The first multilevel problem involves top down processes whereby team and organizational processes shape the experience of job satisfaction of individuals. In these studies, predictors at
the team and organization level of , influence individual experiences of job •, j, faction. For example, Jex and Bliese (|i' nj~ found that the collective self-efficacy  , u US army companies positively influence''* individual experiences of job sat , ■   ( , i addition, collective self-efficacy the relationship between work o job satisfaction such that work overload was ' less negatively related to job satisfaction in units with higher levels of collective' self-efficacy. Other multilevel studies i,., investigated the way individual        ^ 11, has been shaped by aggregate pii. such as leadership (Griffin et nL.">ni> organizational justice (Mossho d. i ,. 1998) group racial composition (Mi „■ |,.r et al., 1999), and demographic diversity,, (Wharton et al., 2000).
Second, researchers have incorporated job ' satisfaction as an aggregate construct v il'u , systems of other variables that are assessed beyond the individual level. This apuroadt seeks to extend theoretical relations i obtained at the individual level to higher levels of analysis (Chen et al., 2005) Ostroff (1992) invigorated research ol ill -kind when she proposed that aggregate job
satisfaction would be related to o.....
zational performance. Her study exit id;:i the satisfaction-performance debate to the organizational level of analysis and pre uLu a theoretical framework for undcrsta l ii if how individual attitudes might mfli i*'iir organizational outcomes. Patterson, Warr ..ni. West (2004) further extended this thinking, to show that aggregate job satisfaction mediated the fink between a conin. i organizational climate and objective mea-i r\ of company performance. They argued Ma" aggregate job satisfaction captured affective experiences that arose from the pml.t.-and procedures enacted in the company. In summary, researchers are increas i'a'\ concerned with studying multilevel re) i'i n ships between job satisfaction and team and organizational variables. This researc i Ii. * opened up a whole new realm of posstb lilii * concerning the role and the importance of job satisfaction.
CONCLUSION
Jcb
satisfaction has been an active and ,,e area of research since the 1930s, 'lin i'iew nns reveaied mat there is still ,,, l i.ible interest in job satisfaction and "hjl11 'towing number of methods are being sfj (> jfiidy satisfaction. In particular, we d studies using methodologies such as '.. i _ ,i:Sj questionnaires, and diary studies. ro. „| Jevelopments in the methods used to lady the affective aspects of job satisfaction ■ fat i is continuing growth and sophistica-,„ | research methods and designs in the jjci t )\.;rall, our review suggests that while wiiil.' i- ro'e,s t''liinSe f°r individuals and the--i icrspectives fade or rise for scholars, .ill? <-\ ii rience of job satisfaction continues '*% play a central role in understanding life ,t \Mirl   We expect that this interest will i-ontinuc across the spectrum of issues raised iitihis chapter. Debate regarding the definition itJ i.inire of job satisfaction will continue .i insights grow from new methodologies, ui)"- • ultural perspectives, and changing theoretical paradigms. This debate is important and will generate new methodologies as researchers seek to explore the implications of ne\rdefinitions for theory, measurement, and jiiactice. Recent advances in intraindividual . iiu multilevel designs suggest that enormous scope remains for embedding the notion of n<j satisfaction within broader theories and jr.ictices. We conclude that the study of job satisfaction is a healthy topic of research that will continue to enhance understanding of hcw.people interact with organizations.
REFERENCES
Ald.:n, R... and Brief, A.P. (1978) 'Examination of .Iternative models of job satisfaction', Human Relations, 31(1): 91-8.
Arvcy, R.D., Bouchard, T.J., Segal, N.L and Abraham, L.M. (1989) 'Job satisfaction: environmental and genetic components', Journal of ipplied Psychology, 74(2): 187-92.
Aryee, S., Fields, D. and Luk, V. (1999) 'A cross-cultural 'test of a model of the work-family interface', Journal of Management, 25(4): 491-511.
Baumeister, R.F., Bratslavsky, E., Finkenauer, C. and Vohs, K.D. (2001)'Bad is stronger than good', Review ol General Psychology, 5(4): 323-70.
Bowling, N.A. (2007) 'Is the job satisfaction-job performance relationship spurious? a meta-analytic examination', journal of Vocational Behavior, 71(2): 167-85.
Brayfield, A.H. and Rothe, H.F. (1951) 'An index of job satisfaction', Journal of Applied Psychology, 35(5): 307-11.
Brief, A.P. and Roberson, L. (1989) 'Job attitude organization: an exploratory study', Journal of Applied Social Psychology, 19(9): 717-27.
Bussing, A. and Bissels, T. (1998) 'Different forms of work satisfaction: concept and qualitative research', European Psychologist, 3(3): 209-18.
Cammann, C, Fichman, M., Jenkins, D. and Klesh, J. (1983) 'Assessing the attitudes and perceptions of organizational members', in S. Seashore, E. Lawler, P. Mirvis and C. Cammann (eds), assessing Organizational Change: A Guide to Methods, Measures, and Practices, New York: John Wiley,
Chen, 6., Bliese, P.D. and Mathieu, J.E. (2005) 'Conceptual framework and statistical procedures for delineating and testing multilevel theories of homology', Organizational Research Methods, 8(4): 375-409.
Cook, T.D. and Campbell, D.T. (1979) Quasi-experimentation: design and analysis issues for field settings, Chicago: Rand McNally College Publishing Company.
Cook, T.D., Campbell, D.T. and Perracchio, L. (1992) 'Quasi-experimentation', in M.D. Dunnette and L.M. Hough (eds), Handbook of Industrial and Organizational Psychology (Second edn., Vol 1), CA: Consulting Psychologists Press.
Cranny, C.J., Smith, P.C. and Stone, E.F. (1992) 'The construct of job satisfaction', in C.J. Cranny, P.C. Smith and E.F. Stone (eds), Job Satisfaction: How People Feel About Their Jobs and How It Affects Their Performance, New York: Lexington Books.
Dormann, C. and Zapf, D. (2001) 'Job satisfaction: a meta-analysis of stabilities', Journal of Organizational Behavior, 22(6): 483-504.
Dunham, R.B. and Herman, J.B. (1975) 'Development of a female faces scale for measuring job satisfaction', Journal of Applied Psychology, 60(5): 629-51.
Dunham, R.B., Smith, F.J. and Blackburn, R.S. (1977) 'Validation of the index of organizational reactions with the JDI, the MSQ, and Faces Scales', Academy of Management Journal, 20(3): 420-32.
Ewen, R.B., Smith, P.C, Hulin, C.L and Locke, E.A. (1966) 'An empirical test ot the Herzberg two-factor
210
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
theory', Journal of Applied Psychology, 50(6): 544-50.
Fisher, CD. (2000) 'Mood and emotions while working: missing pieces of job satisfaction', Journal of Organizational Behavior, 21 (Spec Issue): 185-202.
Fisher, CD. (2002) 'Antecedents and consequences of real-time affective reactions at work', Motivation and Emotion, 26(1): 3-30.
Forgas, J.P. and George, J.M. (2001) 'Affective influences on judgements and behavior in organizations: an information processing perspective', Organizational Behavior and Human Decision Processes, 86(1): 3-34.
Goldberg, L.R. (1990) 'An alternative "description of personality": the big-five factor structure', Journal of Personality and Social Psychology, 59(6): 1216-29.
Gottfredson, G.D. and Holland, J.L (1990) 'A longitudinal test of the influence of congruence: Job satisfaction, competency utilization, and counterproductive behavior'. Journal of Counseling Psychology, 37(4): 389-98.
Graen, G.B. (1966) 'Addendum to "an empirical test of the Herzberg two-factor theory"', Journal of Applied Psychology, 50(6): 551-55.
Griffin, M.A., Patterson, M.G. and West, M.A. (2001) 'Job satisfaction and teamwork: the role of supervisor support', Journal of Organizational Behavior, 22(5): 537-50.
Heller, D., Judge, T.A. and Watson, D. (2002) 'The confounding role of personality and trait affectivity in the relationship between job and life satisfaction', Journal of Organizational Behavior, 23(7): 815-35.
Herzberg, F., Mausner, B., Peterson, R.O. and Capwell, D.F. (1957) Job Attitudes: Review of Research and Opinion, Pittsburgh, Psychological Service of Pittsburgh.
Herzberg, F„ Mausner, B. and Snyderman, B.B. (1959) The Motivation to Work, New York: John Wiley and Sons.
Highhouse, S. and Becker, A.S. (1993) 'Facet measures and global job satisfaction', Journal of Business and Psychology, 8(1): 117-27.
Horn, P.W., Caranikas-Walker, F., Prussia, G.E. and Griffeth, R.W. (1992) 'A meta-analytica! structural equations analysis of a model of employee turnover', Journal of Applied Psychology, 77(6): 890-909.
Hoppock, R. (1935) Job Satisfaction, New York: Harper.
laffaldano, M.T. and Muchinsky, P.M. (1985) 'Job satisfaction and job performance: a meta-analysis', Psychological Bulletin, 97(2): 251-73.
Hies, R. and Judge, T.A. (2002) 'Understanding dynamic relationships among personality, mood, and job satisfaction: a field experience sampling
study'. Organizational Behavior and Hun„n Processes, 89(2): 1119-39.        ' °"
Hies, R. and Judge, T.A. (2004) 'An experieni e sample measure of job satisfaction and its reLiionsh p wjtjj affectivity, mood at work job beliefs, era general job satisfaction', European Journal of VJork ani Organizational Psychology, 13(3): 367-89.
Ironson, G.H., Smith, P.C., Brannick, M.T., G bsoi.yy^ and Paul, K.B. (1989)'Construction of a job in i|^e!a| scale: a comparison of global, composite, and spptfe measures', Journal of Applied Psychology, 7401 193-200.
Jex, S.M. and Bliese, P.D. (1999) 'Efficacy beliefs as a moderator of the impact of work-related stressors a multilevel study', Journal of Applied Psycliologf' 84(3): 349-361. ' '
Johnson, R.E., Rosen, C.C. and Levy, P.E. (in press} 'Getting to the core of core self-evaluation: a review and recommendations', Journal of V-qanirauonal Behavior.
Judge, T.A. and Bono, J.E. (2001) 'Rel.i.jonship 0| core self evaluations traits-self-esteem, ■ret sa'mS self-efficacy, locus of control, and emotional SuiHitif-with job satisfaction and job performance: 3 metaanalysis', Journal of Applied Psychology, 86(1) 80-92.
Judge, T.A., Bono, J.E., Erez, M. and Locke, E.A. P005} 'Core self-evaluations and job and life satisfaction' the role of self-concordance and goal attainment', Journal of Applied Psychology, 90(2): 2 1/ 68
Judge, T.A., Bono, J.E., Hies, R. and Gerhaidl, M.W (2002) 'Personality and leadership: a qna.itRtiwaml quantitative review', Journal of Applied Psychology, 87(4): 765-80.
Judge, T.A. and Hulin, C.L. (1993) 'Job sjlislaction as a reflection of disposition: a multiple causa, analysis', Organizational Behavior and Human Dewion Processes, 56(3): 388-421.
Judge, T.A. and Hies, R. (2004) 'Affect and joo satisfaction: a study of their relationship a: work and home', Journal of Applied Psychology, 89(4) 661-73.
Judge, T.A. and Larsen, R.J. (2001) 'Dispositional affect and job satisfaction: a review and theore'icai extension', Organizational Behavior and Human Decision Processes, 86(1): 67-98.
Judge, T.A., Locke, E.A., Durham, C.C. and Kluger, A N. (1998) 'Dispositional effects on job ani satisfaction: the role of core evaluations', Journal of Applied Psychology, 83(1): 17-34.
Judge, T.A., Parker, S., Colbert, A.E., Heller, D. and Hies, R. (2001a) 'Job satisfaction: a cross-cultural review', in N. Anderson, D.S. Ones, H.K. Singang" and C Viswesvaran, (eds) Handbook of Industrial
JOB SATISFACTION IN ORGANIZATIONAL RESEARCH
-\, Thorensen, C.J., Bono, J.E, and Patton, 5 k (200*b) 'The job satisfaction-job performance
rgJruzBtional Psychology, London: Sage ions.
Hp
I no'ishii' a qualitative and quantitative review, y,cal Bulletin, 127(3): 376-407. rrfr' SA Uol)2) 'The role of qualitative research in Kfical journals', Psychological Methods, 7(1):
r (1lJ55) 'Tne construction of a new type of measure', Personnel Psychology, 8: 65-77. and Schaubroeck, J. (2000a) 'A field •.Testing frontline opinion leaders as ' charge agents', Journal of Applied Psychology, 85(6):
Larn S.S <■ <lnd Schaubroeck, J. (2000b) 'The role of locus of control in reactions to being promoted and to being passed over: a quasi experiment', Academy K/ement Journal, 43(1): 66-78.
Landy, H (1989) 'Job satisfaction: the meaning of work', Psychology of Work Behavior (fourth ed.),
. Patrfir Grove, CA: Brooks/Cole Publishing Company.
lockc, E.A (1969) 'What is job satisfaction?', Organizational Behavior and Human Performance, 4(4) 309S36.
to'eke, E.A. (1976) 'The nature and causes of job - satisfaction', in M. Dunnette (ed.), The Handbook of
Organizational Psychology, Chicago, Rand McNally. Locke, I.A. (1983) 'The nature and causes of job
satisfaction', in M. Dunnette (ed.), Handbook of
Industrial and Organizational Psychology New York:
John Wiley and Sons, togan, M S.andGanster, D.C (2005) 'An experimental
evaluation of a control intervention to alleviate
job related stress', Journal of Management, 31(1):
90-107.-:
Lyons, H.Z. and O'Brien, K.M. (2006) 'The role of person-em/irowient fit in the job satisfaction and tenure imertions of African American employees', Journal of Counseling Psychology, 53(4): 387-96.
Maichel, C and Owens, S. (2007) 'Qualitative arch in psychology', History of Psychology, 10(4): 301^24.
Mnlhi'eu, J.E. (1991) 'A cross-level nonrecursive model of the antecedents of organizational commitment and satisfaction', Journal of Applied Psychology, 76(5): 60 »18.
Md-drin, D.B. and Sweeney, P.D. (1992) 'Distributive and.; procedural justice as predictors of satisfaction: with personal and organizational outcomes', Academy of Management Journal, 35(3): 626-37.
Mwsholder, K.W., Bennett, N. and Martin, C.L. (1998) A multilevel analysis of procedural justice context', Journal of Organizational Behavior, 19(2): 131-41.
Mueller, C.W., Finley, A, Iverson, R.D. and Price, J.L. (1999) 'The effects of group racial composition on job satisfaction, organizational commitment, and career commitment: the case of teachers', Work and Occupations, 26(2): 187-219.
Niklas, CD, and Dormann, C. (2005) 'The impact of state affect on job satisfaction', European Journal of Work and Organizational Psychology, 14(4): 367-88.
Ostroff, C (1992)'The relationship between satisfaction, attitudes, and performance: an organizational level analysis', Journal of Applied Psychology, 77(6): 963-74.
Patterson, M. Warr, P. and West, M. (2004) 'Organizational climate and company productivity: the role of employee affect and employee level', Journal of Occupational and Organizational Psychology, 77(2): 193-216.
Pekrun, R. and Frese, M. (1992) 'Emotions in work and achievement', in CL. Cooper and IT. Robertson, (eds), International Review of Industrial and Organizational Psychology, Chichester: John Wiley and Sons.
Petty, M.M., McGee, G.W. and Cavender, J,W, (1984) 'A meta-analysis of the relationships between individual job satisfaction and individual performance', Academy of Management Review, 9(4): 712-21.
Pillai, R., Schriesheim, CA. and Williams, E.S. (1999) 'Fairness perceptions and trust as mediators for transformational and transactional leadership: a two-sample study', Journal of Management, 25(6): 897-933.
Quinn, R.P. and Staines, G.L. (1979) The 1979 Quality of Employment Survey, Institute for Social Research, University of Michigan, Ann Arbour, Michigan.
Rain, J.S., Lane, I.M. and Steiner, D.D. (1991) 'A current look at the job satisfaction/life satisfaction relationship: review and future considerations'. Human Relations, 44(3): 287-306.
Rode, J.C. (2004) 'Job satisfaction and life satisfaction revisited: a longitudinal test of an integrated model', Human Relations, 57(9): 1205-30,
Roethlisberger, F.J. and Dickson, W.J. (1939) Management and the worker: an account of a research program conducted by the Western Company, Hawthorne Works, Chicago, Cambridge, MA: Harvard University Press.
Rose, M. (2003) 'Good deal, bad deal? job satisfaction in occupations', Work, Employment, and Society, 17(3): 503-30.
Rousseau, D. (1978) 'Relationship of work tononwork', Journal of Applied Psychology, 63(4): 513-17.
Roznowski, M.0989) 'Examination of the measurement properties of the Job Descriptive Index with
212
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
experimental items', Journal of Applied Psychology,
74(5): 805-14. Salancik, G.R. and Pfeffer, j. (1978) 'Asocial information
processing approach to job attitudes and task design',
Administrative Science Quarterly, 23(2): 224-53. Scarpello, V. and Campbell, J,P. (1983) 'Job satisfaction:
are all the parts there?', Personnel Psychology, 36(3):
577-600.
Schaffer, R.H.(1953) 'Job satisfaction as related to need satisfaction in work', Psychological Monographs: General and Applied, 67(14): 1-29.
Seibert, S.E. (1999) 'The effectiveness of facilitated mentoring: a longitudinal quasi-experiment', Journal of Vocational Behavior, 54(3): 483-502.
Smith, P.C. (1992) 'In pursuit of happiness', in C.J. Cranny, P.C. Smith and E.F, Stone (eds), Job Satisfaction, New York: Lexington Books.
Smith, P.C, Kendall, L.M. and Hulin, C.L (1969) The Measurement of Satisfaction in Work and Retirement, Chicago: Rand McNally.
Spector, P.E. (1997) Job Satisfaction: Application, Assessment, Causes, and Consequences, Thousand Oaks: Sage Publications.
Staw, B.M., Bell, N.E. and Clausen, J.A. (1986) 'The dispositional approach to job attitudes: a lifetime longitudinal test', Administrative Science Quarterly, 31(1): 56-77.
Staw, B.M. and Ross, J. (1985) 'Stability in the midst of change: a dispositional approach to job attitudes', Journal of Applied Psychology, 70(3): 469-80,
Tait, M„ Padgett, M.Y. and Baldwin, T.T. (1989) 'Job and life satisfaction: a reevaluation of the strength of the relationship and gender effects as a function of the date of the study', Journal of Applied Psychology, 74(3): 502-7.
Tharenou, P. (1993) 'A test of reciprocal causality for absenteeism', Journal of Organizational Behavior, 14(3): 269-87.
Thorensen, C.J., Kaplan S.A., Barsky, A.P. and de Chermont, K. (2003) 'The affective underpinnings of job perceptions and attitudes: a meta-analytic review and integration', Psychological Bulletin, 129(6): 914-45.
Watson, D., Clark, L.A. and Teliegen, A 'Development and validation of brief measures gf" positive and negative affect: the PANAS sa|es' ' Journal of Personality and Social Psychology, 5*0 1063-70.
Weiss, D.J., Dawis, R.V., England, G.W. and Lofquisi LB. (1967) Manual for the Minnesota Questionnaire: Minnesota Studies in Vocation Rehabilitation, University of Minnesota, Vocational Psychology Research.
Weiss, H.M. (2002) 'Deconstructing job satisfaction separating evaluations, beliefs and affective experiences', Human Resource Management Review, 12(2) 173-94.
Weiss, H.M. and Ctopanzano, R.(1996) 'Affective events-" theory: a theoretical discussion of the stucluie causes and consequences of affective experiences at work', Research in Organizational Behavior, 18-1-74.
Weiss, H.M., Nicholas, J.P. and Daus, C.S. (1999) 'An examination of the joint effects of a'Tective experiences and job beliefs on job satisfaction and variations in effective experiences over time', Organizational Behavior and Human Decision Processes,, . 78(1): 1-24.
Wharton, A.S., Rotolo, T. and Bird, S.R. (2000) 'Social context at work: a multilevel • analysis of job satisfaction', Sociological Forum, 15(1)^, 65-90.
Wong, C.S., Hui, C. and Law, K.S. (1998) 'A longitudinal' study of the job perception-job satisfaction relation.-' ship: a test of three alternate specifications', Journal of Occupational and Organizational Psychology, 71(2): 127-46.
Workman, M. and Bommer, W.H. (2004) 'Redesigning computer call center work: a longitudinal field experiment', Journal of Organizational Behavior, 25(3): 317-37.
Wright, T.A, and Bonett, D.G. (2002) 'The moderating effects of employee tenure on the relation between organizational commitment and job petformance: \ a meta-analysis', Journal of Applied Psychi -■ 87(6): 1183-90.
13
Studying Organizationa Populations Over Time
Glenn R. Carroll, Mi Feng, Gael Le Mens, and David G. McKendrick
INTRODUCTION
\ir.st social scientists agree that formal ■ ■■ jmitzutiuiis occupy a central role in modem in ietyin that most collective action occurs within lliem. There is also an emerging cji ->-n.iUS, evidenced by the research designs organizational scientists typically use, that .1 1« 1 crful way of studying organizations . 1. ■ I. .'s the collection and analysis of data '■"re .tune. Yet, issues about how to study organizations over time sometimes engender confused debate. In this chapter, we aim to r irify some of the main issues involved, ".tin a special focus on studying organiza-1: ral populations over time. Throughout the ' 11 ii.1 , we illustrate our general points with speEific examples drawn from our ongoing si uly of the historical worldwide population °C '<apc drive producers.
What is an organizational population?
'-■|Tl5'; studies of organizational populations were often simply called an 'industry study.'
In a typical study of this kind, analysts study a single industry over time by collecting information on all the firms or other organizations operating in a market. For instance, Epstein (1927) conducted a study of failure among American automobile producers. Such studies continue today (see Klepper, 2002; Mitchell, 1994; Tripsas, 1997). The definition of industry or market used in industry studies may be explicit and clear. However, it is usually not theoretically reasoned in any depth and often simply follows common sense or convention, as incorporated in directories, reports, or archives (used as sources for data). As a result, what is regarded as the relevant organizational population varies significantly from one study to the next - e.g., in some cases, it implies legal incorporation, in others, the production of goods or services, in yet others, the sale of goods or services possibly produced by other parties.
Attention to the conceptual foundations of the population-based design can be traced to the population ecology perspective on organization. Specifically, Hannan and Freeman (1977: 935-6) reasoned that,
RESEARCHING WORK AND INSTITUTIONS THROUGH ETHNOGRAPHIC DOCUMENTARIES 271
16
Researching Work and
ten?
Institutions through Ethnographic Documentaries
John 5. Hassard
INTRODUCTION
The aim of this chapter - which complements the discussion of feature films in Chapter 35 - is to describe and analyse the methods and methodology of ethnographic documentary for organizational research. This is achieved predominantly through an historical appreciation of the development of styles and schools. In so doing, we note how, in the British tradition, the ethnographic documentary has been deployed to provide grounded sociological insights into the 'real world' of institutions and occupations. The dynamic of this method is one of the attempts to gain evermore direct and unmediated images of organizations and work.
The chapter adopts the following structure. Initially, we introduce the methods and methodology of ethnographic documentary in order to prepare the way for an historical review of anthropological studies of work, occupations, and organizations. Subsequently, we document how such ethnographic evidence has been realized under four main
filmic styles - the world-of-labour film. the free cinema film, the modern television documentary, and the video diary/hlog and analyse each as a case example of i •> i -i verite, or film-truth. Finally, we discuss" the representational nature of ethnogi. > it documentary and examine the extent in which the film-truth method achieves its ifv^i m of offering privileged access to the eve , i;» reality of work and organizations.
History and method of ethnographic documentary
In The Cinema and Social Science, dc Heuich (1962: 13) writes, 'When Lumiere dclinitdy established a technique and sent his c.i i ■■ amen all over the world, the ethnographic and sociological cinema was born'. Else'.' ici1 Morin (1956: 14) notes how the precuisors of the cinema intended to develop the cinematograph as a 'research' instrunM l order to study the 'phenomena of natme'. While the ingenuity of Lumiere's techniques made it likely that the cinematogiapli
jd potentially, become a methodologically ľ i, 11   ;' instrument with which to research
, i u d organizational behaviour, Morin ii.'fii I i-16) notes, however, that it was
i iims: before the 'illusionists' had taken ,i s lit" íyp6°f 'microscope' away from the , ( ,,ni| s' and transformed it into a 'toy', (hereby reviving the 'tradition of the magic jj i,, i Mid the shadow play'.
i |> . 'äinst this backdrop of methodolog-,,,, ,i id technical development in which, as Mi i i  1 ggests' me cinematograph becomes
i, ,[, I away" from science, that we find .,ii-.|i truth' - for Morin, one of the ,j>ai,ic aims of early film-making - being
fMl 11 luced into the film-maker's art. Over •|. ■ ľ.i i-lcs the century, documentary film-
i Lii-i in particular, have addressed the ,„ .i| i   of how to make anthropological
i ii, iiľ re 'real' and 'truthful'. The history
)f ■! m. ľ raphic documentary filrrrmaking has . vi -i i ange of styles emerge in this quest. For '|i. mi -Apart these have been subsumed under one generic lille - cinema verite (film-truth).
rh!i vphy and method
\a wling to its proponents, the simof cinema
■•id is simply to present the 'truth'. In
ii. " nch Cinema since 1946, Roy Armes ■ I'M : 125), for example, argues that cinema : ii'i ffers a 'rejection of the whole aesthetic u jt'l kit ihe art of the cinema is based. An n .i.'-ling visual style and striking beautiful effects are rejected as a hindrance to the poilruyalof the vital truth'.
Oil i. the years, however, critics have pointed to confusion about the basic aims of cinema verite, and suggested that under its banner all kinds of films have been ňu dc, many ot which are unrelated to the ■Mizina ľ philosophy. Debate over the value aiKtjpurpo.se of cinema verite was particularly
v ted in the early 1960s, and notably ■• "undthe time of the 1963 Lyon 'summit' conference on cinema verite, organized by '■i-s/io Television Francaise and attended by suchnotables of the ethnographic film as Jean Ruuch, Edgar Morin, Robert Drew, Richard Leaeock, Michel Brault, and Albert and David Missies. While proponents such as Armes
heralded cinema verite as the 'true art of cinema', critics branded it as 'anti-art'. Among the critics, James Lipscombe (1964: 62), for example, argued that the term cinema verite had become used so loosely that most of the films produced in its name 'have absolutely nothing in common except celluloid'. In a similar vein, Charles Ford claimed that cinema verite was the 'biggest hoax of the century', and that 'nothing is more fabricated, more prepared, more licked into shape, than the so-called improvisation of cinema verite' (Natta, 1963, quoted in Issari and Paul, 1979). Such diversity of views has made cinema verite one of the most controversial methods in film studies, with this controversy enduring to the present day.
While we explain later how, as a philosophy and method, cinema verite ostensibly starts with Dziga Vertov's work on the concept of kino-eye in 1919, it was during the 1960s that the style became widely adopted and discussed in film-making circles. When it was adopted, it was in fact as a tribute to Vertov's work that the term cinema verite was first used in a stylistic sense, in fact to describe Chronique d'un Ete (1962), a film collaboration between the anthropologist Jean Rouch and the sociologist Edgar Morin. According to Issari and Paul (1979: 6), such was the impact of Chronique d 'an Ete that the term 'spread like wildfire', one result being a proliferation of definitions in which cinema verite:
...acquired different names from its founders, practitioners and critics in accordance with their understanding of the style. To Jean Rouch and Edgar Morin it was cinema verite; Richard Leacock called it 'living camera'; Mario Ruspoli, the Maysles brothers, and Louis Marcorelles were amongst those who favoured 'direct cinema'. William Bleum designated the style as 'mobile camera', but William Jersey selected the term 'realistic cinema'. Italians named it'film inquiry'. To Armondo Plebe it became 'synchronous cinema'; to Colin Young, 'cinema of common sense'; to Jean Claude Bringuier, 'cinema of behaviour'; to Norman Swallow, 'personal documentary' or 'tele-verite'. Others dubbed it 'film journalism', 'truth film', 'direct shooting' and 'free cinema' (Issari and Paul, 1979: 7).
272 THE SAGE HANDBOOK OF ORG;
Of all the terms coined, however, two have proved the most enduring - cinema verite and 'direct-cinema'. Although methodologically they stem from largely similar historical roots, and while their associated films have many similarities, their proponents often testify to separate philosophies, the main difference residing in the function of the film-maker (and by extension 'film-researcher'). While the practitioner of cinema verite (in the Rouch and Morin style) deliberately intervenes, hoping that greater spontaneity and 'truth' will be stimulated by the 'subjective' participation of the film-maker in the event filmed, the 'direct' film-maker claims to take an 'objective' stance, merely standing by in the hope that a situation of anthropological or sociological interest will resolve itself in a dramatic way (see Barnouw, 1975; Nelmes, 2003).
In terms of stylistic influence, it is 'direct-cinema' that has become the exemplar for the film-truth method generally. In this style, the film-maker tries above all to avoid sociological judgement and subjectivity. The aim is to convey the phenomenology of a social situation; to film people as they live their everyday lives. The philosophy suggests that the filmmaker's role is to reveal the most essential parts of this reality, not to create a new one. The 'direct' filmmaker should dispense with those elements of method which serve to distort 'truth'; that is, with the professional 'camouflage' of conventional film-making, such as scripts, scenes, studios, actors and so forth. Films are to be made of ordinary people in real situations, the only acceptable aids being portable cameras and sound equipment. As the filmmaker is only interested in events as they naturally unfold, no direction is permitted. In the more 'pure' forms, there is no plot, no preconceived dialogue, and no questions are either posed or answered. The philosophy states that, in shooting a film, the crew's task should be confined to following subjects and filming their moments of personal drama. Decisions concerning action should rest with the subject - the filmmaker merely decides whether or not to
IIZATIONAL RESEARCH METHODS
film a particular piece of action (I   , . Paul, 1979; Nelmes, 2003). m-
Pioneers and schools
We have noted how a range of el i in,, documentary approaches have bee . .h ., under the general heading of cir We will now attempt to place si 'Mi, the development of these appro by discussing the ideas and wt of its founders, Dziga Vertov ainl |<, Flaherty, and second, by descn'i'i n,,. cinema verite has been shaped in what' we will call the French, Ami ih in British schools of ethnographic d(   n tll> We will show how each of these schools offers a distinctive approach to el 11 « , n" filmmaking and, especially in the Uriiish tradition, withregard to work, insti i buns ,ir occupations.
Dziqa Vertov and kino-pravda As noted, it was in the hands of thi J iw film-maker Dziga Vertov (1896-   ,s I name Denis Arkadievitch Kaufman i i| n ethnographic documentary style 'film-truth' or kino-pravda is first developed, in theory and practice (Petric, 197^ i is associated with what has been I.  i r 11 . 'extremist theory' of the ethnogi i:hi. nr the view that the cinema should rid itself, as Sadoul(1940:172)says, 'ofeveryll pij ,<l u. has not been taken from life'.
In line with the prevailing phi!" ■■ i, .■" socialist realism in early Soviet film making, Vertov argued that kino-pravda n-qi i'i ll-nonparticipation of the filmmaker as a fundamental condition of attaining ethnographic authenticity. If the movie camera .,, - In I. an effective tool in influencing the -n:i.--.-. it was necessary to present realistic images of their everyday lives. Instead ■■: ■■ jja' sets, real locations would be used, ui -.In1 peasants played peasants and workers played workers. If filmmaking was to become, a truly proletarian art, Vertov believed its I unction was to portray the life of the people in
all its intense and intimate detail il i.....»
1964: 73). Given the technology available, however, Vertov was perhaps somewhat
(.r-EARCHING WORK AND INSTITUTIONS 1
,ii ifious in his claims to be providing l'l '.. •r.'i.'i.' approach for anything more 3 short sequences of film. In order to ''in' , tuthful' images with cumbersome * inni '-y- ms ear'y ^ms saw ^ use °^
J'+ i ,.'vSand telephoto lenses, as in his '|(V |. iirhiiig of a family weeping atagrave. jn the early 1920s, at a time when filui.n1.iiig technology was in its infancy, . ,. llM_ nal ideas on the ethnographic film ' fn. prophecy more than practice. Vertov, remains the father figure of cinema verite in that his philosophies reflect the core pj.,,,1 I l ic movement.
lb ten Flaherty's Nanook of the North n, ul ii r founding father of the cinema ... /, c. .umentary was the American trapper-■ n- .i or Robert Flaherty, who in 1920 ,u I spent fifteen months making a
pij.ii, film in Hudson's Bay for the French fur company Revillon, a film screened in I "22 as Nanook of the North. In il. ii-" lal version, Nanook is a silent Iv 111 M -■ portrait of the world of the t'sktmn 'I he story develops through scenes, iv ikh iternate between descriptions of tl ^ ciivduy straggle for existence and ll - i "* idem ess of family life. The first tiuunii i film in the cinema verite style, [\ n ml lemains highly acclaimed and widely ..i-m i, (Nelmes, 2003).
The high quality of Nanook, however, is ltd n il ibuted to the large amount of prepara-|i 11 vn HI undertaken by Flaherty (as also with In- .'.'i i if Aran, 1934: see Calder-Marshall, 1963) Unlike in 'purist' forms of direct obsci vat ton, where subjects receive a minimal b ii I'11 Flaherty studied Nanook and his family in detail before filming. As de Fieusch (1962. 3s!) notes, 'Flaherty does not follow Nanook about in order to take pictures: he enters into conversation with him, he asks i in t ooperate closely in the sociological portrait'. Because of the elaborate manner " "ii . i the scenes are reconstructed, the extent to which Nanook represents film-truth ls 1'.' i isly questionable. The well-known scene where Nanook and his family wake up in •l: i gloo, for example, is achieved by Flaherty
OUGH ETHNOGRAPHIC DOCUMENTARIES 273
removing the roof in order to have enough space and light to shoot (Winston, 1979). In Flaherty's hands, the documentary becomes 'a work of art imbued with rationality and truth' (de Heusch, 1962: 35-6). Nanook is constructed like a fictional film in which the main character acts the part of himself within the framework of a prepared story. The story is conveyed in accordance with the demands of the director.
The French School: Rouch and Morin's Chronique d'un Ete
Historians of cinema verite commonly claim that the development of increasingly unobtrusive film-making technologies will bring the notion of the 'cinema eye' closer to reality (see Barnouw, 1975; fssari and Paul, 1979; Nelmes, 2003). It is claimed, for example, that a classic example of such technological progress was witnessed in the development of the 16 mm camera with portable sound equipment, notably in the work of the French anthropologist turned filmmaker Jean Rouch. Rouch is often considered the founder of ethnographic cinema, because he overcame a number of traditional filmmaking handicaps simply by using lighter technology. Synonymous with the so-called 'French School' of the ethnographic documentary film, Rouch is the figure most readily associated with the term cinema verite (Issari and Paul, 1979; Aitken, 2005).
As noted earlier, Rouch is best known for a film made in collaboration with Edgar Morin (and the French-Canadian film-maker Michel Brault) on the lives of Parisians, Chronique d'un Ete (Chronicle of a Summer, 1961). According to Rouch, Chronique d'un Ete was an attempt to combine 'Vertov's theory' and 'Flaherty's method' (Armes, 1966). Initially twenty-one hours long, but later cut to one hour and forty five minutes, the film begins by showing the reactions of passersby to the question, 'Are you happy?', and continues by interviewing them in detail about, for example, their likes, dislikes, hopes, and anxieties. In the final scenes, we witness the participants themselves viewing the film in a screening room, the footage
274
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
RESEARCHING WORK AND INSTITUTIONS THROUGH ETHNOGRAPHIC DOCUMENTARIES 275
describing their reactions to the interview process.
Although technically Chronique d'un Ete benefits from the development of realtime (rather than post-synchronization) sound recording, it appears to suffer methodologically from the fact that most of its subjects appear uneasy in front of the camera. More than any other landmark documentary, Chronique d'un Ete highlights the issue of whether the camera should be used to 'record' action or to 'stimulate' it. In hands of Rouch and Morin, the cinema verite film becomes not so much a recording of what would happen if the camera was not there, as what happens because the camera is there.
The American School: Drew Associates' Primary
In contrast, the American tradition of cinema verite, based on the 'direct-cinema' approach, adopts a more Cartesian orientation. The exemplar of this is found in the collaborative work of Robert Drew, Richard Leacock and colleagues (including filmmakers Albert Maysles, Donn Pennebaker, and Gregory Shuker) - or Drew Associates - and their classic documentary in the 'living-camera' style, Primary (1960).
Providing funds to develop sufficiently mobile 16 mm equipment, the living-camera project was commissioned by lime-Life, for whom Drew Associates were to make a news-film equivalent of the magazine's well known news-spreads. To this end, Drew and Leacock negotiated an agreement with Senators John F. Kennedy and Hubert Humphrey to record their political campaigns in the 1960 Winsconsin primary election. The plan was to follow the candidates and film their activities 'without intruding'. No lighting was to be used; the crew would film only what was in front of them; they would go into situations 'cold'.
By employing this method, Drew felt that not only would the facts of a situation be revealed, but also the essence or feeling. The 'living' camera would report the emotional as well as the intellectual truth. 'Internal forces'
would be revealed by studying -i ,iL forces' (Leacock, 1992). Like Ve (.- | Associates presumed that the es situation would be captured when : u",j,. ( were too tired, relaxed, or involved lo pruhy themselves from the scrutiny of. nl. (l f The film is commonly considered a break", through in that, with relatively sm.ill, cmiei and portable technology, the camera « could now 'walk in and out of buildings m and down stairs, film in a taxi or limousine' and all the time 'get sound and picfti events occurred' (Leacock, 1992). 1'iequentlv screened, Primary is a film with, i <, i „ Vertovian flavour, in which the suhjeil matter is treated in a largely disinterested :l .i -11 n , | action following no plot other than u u i, | events which leads to Kennedy w in 11 ii election.
Work and institutions in the ethnographic documentary
Having described the basic philosophy and method of the ethnographic docurr offered a review of landmark co;i ill i.i n> we now offer a case account i i-i I -i researching work and organizatio i i| it >l the ethnographic description of occupations and institutions. From the 1930s to the present, and especially since the advent of I.-!.-, -mi large audiences have been found in Britain for documentary films, which cc:i-u i| hi the relationship between individuals and their workplaces. We document how such work has been developed under four main the world-of-labour film, the free cinrirttt film. the modem television documentary, i i;J Ir video diary/blog - and discuss each as a case example.
World-of-labour films (1930s/1940s) The origins of work-related ethnn.!. > " documentary in Britain lie in the early world-of-labour films of producer-theorisl .Iota Grierson (seeManvell, 1946; Bamom,. "■"^ Beveridge, 1979; Forsyth, 1979; Miui 1990 and 1998; Nelmes, 2003). In his 1929 silent film on the herring jiidmlry, Drifters, Grierson effectively seres i -■' I i!i
, i pean occupational documentary, 'i''   i- .as revolutionary in European film-' y N , i that for the first time the focus for ''      i,m entertainment was nothing more "' "|, i i ni the work of fishermen. In irnpor-1 I , ^1 els, Grierson's work reflects the I ii i * of both Vertov and Flaherty. While -   u style is evident in Grierson's con-I   social problems, Flaherty's romantic i Ii rpins the quest to record the heroic i ordinary existence (Aitken, 1990 .    i'M. Holder of a doctorate in social I i'om Glasgow University, Grierson's riven by a concern for the ethical I M     ogical purpose of the documentary nliii i- ^&e' (1953; 45) notes, Grierson's , „ i 'to exalt human toil and civic virtues' . ■ i all 'to magnify the unconscious ■ i- the physical effort involved in
i -i.i i'
ijiu"Mil's work subsequently influenced ;ial and industrial documentaries of i  otha, whose work represents an early it to demystify the structure of British ti' i-uial society, especially the role of the M11   toong Rotha's best known films are A/i, ,rd (1935), concerning the construc-l,i i  'fa ship and the reactions of the (in. i liinity in which the shipyard is based, i j Using Tide (1935, later retitled Great
• :s), which deals with extensions to the mpton docks and the interdependence
iI .a ous commercial activities (see Rotha, I 'in, Aitken, 1990, 1998 and 2005). In his iIh.-i.l- ical treatise, Documentary Film Art i ',. Rotha explains his approach to the i-1 iin^raphic film through praising Flaherty's Ljiiiquo and commitment, but criticizing
• is .1- er-romantic and idyllic conception of
■ ' i' i five struggle with nature, an image far '.ni'i ed from, and with little to say about, the
■ i> i ling realities of the modern industrial ■I- I. Rotha's view was that serious ethno-r' r 1 c documentaries should focus on the
I . social realism of industrial civilization; they % should concentrate on issues that confront i large masses of the world's population (Rotha, J 1967). For Rotha, the idyllic documentary 1 largely avoids any conscious, critical social * analysis.
Other quality documentaries set in industrial or other work-related settings find Grierson in collaboration with Alberto Cavalcanti or Harry Watt. For example, it was Grierson who produced Cavalcanti's Coal Face (1936), a study of the lives of miners, while Cavalcanti in turn produced Watt's North Sea (1938), a study of seafarers. The latter film departs, however, from the early requirements laid down by Grierson, in that it employs actors and tells a scripted story. This was also the case with Watt's classic of the period, Night Mail (1936), which tells the story of the night run by the postal train between London and Glasgow, and centrally the work of postal workers receiving, sorting, and dispatching mail during the run. A collaboration with Grierson and Basil Wright, like North Sea, the film was produced by the esteemed Film Unit of the General Post Office (see Aitken 1998 and 2005).
Free cinema movement (1950s/1960s)
In the post-World War Two era, the British ethnographic documentary escaped from its role in creating a largely propagandist view of British society through what came to be known as the 'free cinema' movement, whose aim was to return to a more direct approach to the, notably working-class, documentary. The origins of free cinema lie in the pages of the short-lived Oxford University magazine, Sequence (1949-52: 14 issues). The editors of Sequence, Lindsay Anderson and Gavin Lambert (and later Karel Reisz), severely criticized British filmmaking for presenting a view of the nation from the perspective of the pre-1914 bourgeoisie. It was Anderson's view, in particular, that British filmmaking was devoid of critical analysis, especially regarding the class system. According to Robinson (1973:292), the Sequence view was that British film-makers had characterized 'the function of the working class [as] to provide "comic relief" to the sufferings of their social superiors'.
The term 'free cinema' was first coined in February 1956 as the headline for a National Film Theatre programme of short films. The notes for the programme explain how the
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
movement's films are 'free in the sense that their statements are entirely personal. Though their moods and subjects differ, the concern of each of them is with some aspect of life as it is lived in this country today'. The attitude is one of 'a belief in freedom, in the importance of people and in the significance of the everyday' (quoted in Robinson, 1973: 294). The Sequence philosophy was to show the way 'ordinary people' enact their 'everyday lives'. The objective was for these films to possess a 'feeling of freshness' in documenting the 'real environment in which people work and play' (Robinson, 1973: 294).
The classic of the Free Cinema movement is undoubtedly Anderson and Reisz's Every Day Except Christmas (1957), a film mainly about night workers at London's Covent Garden Market. The film begins with the image of a shed somewhere far out of London, in which a lorry is being loaded with food and flowers at the very moment when the BBC announcer is saying 'goodnight' to listeners. From every corner of the country, similar vehicles speed towards the capital and Covent Garden Market. While the lorries are unloaded, dealers prepare their stands: boxes are opened; apples are polished; workers pause for tea and a chat. A large well-dressed man gives orders; the major retailers arrive and bargain; and a small trader loads a handcart. Housewives and itinerant merchants buy after the retailers have gone; an old woman sells flowers; tramps pick up any damaged fruit. Finally, the lorries are loaded with empty packing cases and depart for the four corners of Britain.
In recording these scenes, the camera carefully studies the faces of traders, customers and the like and catches their subtle yet commonplace gestures. In so doing, Every Day Except Christmas achieves an ethnographic montage in which each face contributes to the sociological effect. As in much of Anderson and Reisz's work, it is the class structure of Britain that is reflected primarily in scenes which document the ebb and flow of work and rest. The philosophy leans firmly upon Vertov's theories in that
the goal is to bring to the audiein reflections on the world in which it I
Modern television documentary (1960s-present)
From the 1960s onwards, television r the short or feature length filfi main medium for disseminating eth
'Ml
■■I
ii
'""-"rV
documentaries in Britain. Indeei-Family (1974) made nationally [a i j. Wilkins' of Reading, British tele\i,i,i y played host to a wealth of film-fnii _n. The advent of mass television, lm' ,. did not alter the trend for eiln m<i n „ filmmakers to be preoccupied w 11 i r |tr standing the nature of work, occtip i-nn ll( institutions.
The 1990s, in particular. sa\, i.nnj audiences bombarded with cir. i studies of work and organizations. < n i |,i audiences being found for series   i I Flying Squad (police officers. bug, li i,,, The Duty Men (customs officers, BBC, 1991), Fire (firefighters, Thames, 1991), DMA. (drug enforcement officers, BBC, 1992),, Town Hall (town councillors, BBC, M;i ■ The Ark (zoo keepers, BBC, .-if.i //„ Adventurers (venturecapitalists, BBC. I" i-j Turning the Screws (prison officers, Ch.i-i.n' 4, 1993), Skipper (trawlermen, BBC. . ."!•! Coal (coalminers, BBC, 1994), >. ,' , (factory workers and managers, Cham..1 1 1995), and When Rover Met /."Ml  'i i workers and managers, BBC, 1996), for example.
Stylistically, British television-based ethnographic documentaries have been influenced considerably by the work of li.i American film-maker Frederick Wiseman, who in the mid-1960s made a series of investigations for US television inl-i m relationships between individuals and institutions. As much as John Griersri, it has been Wiseman's technique of using documentary film for observing social relationships that has influenced a gcm-ruli-ir of high profile British cinema verite filmmakers, including Roger Graef, Nick Broomfield, Diane Tammes, Chris Oxley, and Molly Dineen.
'■
SEARCHING WORK AND INSTITUTIONS THROUGH ETHNOGRAPHIC DOCUMENTARIES 277
, t jly expression of Wiseman's influence .iiid in Graef's series, TTie Space J,n Words (BBC, 1972), which looked at fljniiiiication' jn five institutional settings; t  pmily. school, work, diplomacy, and i Vl-tJ The series was innovatory for British i  i,ion m tnat ^ focused on 'ordinary tin lions' and presented the images with the ,i in um of introduction. Graef's idea was |, I viewers develop their own theories ,1 t,imiiiunicalion'. In filming the series, (i irl ? p<)ucy was 'never to use lights, or -"i p. I scenes, to keep the equipment out of .,  and lo have the minimum of people in ihe room'. To remain as anonymous as rai ;ibie, Graef even had a rule of 'never li i' nig any of the subjects being filmed ii I hi. eye'. A radical venture at the time, (,i,i 1 has noted, with irony, how the response )l 'I ■ media was to interview academics nd critics in order to obtain 'professional' e\ji nations of the issues raised by the films, rr, iiice completely anathema to the project i .■ b.-I".1992). Oraef is better known,  however, for " divi'ii ifntaries which examine the relation-1 n between organizational practices and ij. itutional politics. Examples here are his 'in s on investment in new technology at Mi lish Steel, and the allocation of spare beds in he Nalional Health Service. His most ntnnous work in this style, however, was . --lies documenting the lives and work of members of the Thames Valley Police Force l/'u. i-,T9H2, with Charles Stewart). In this seiieSf Graef wished to combine the apparent (■ i "i I ivity of direct-cinema methods with the analysis of contemporary socialissues, which, in the case of one episode, 'A Complaint rl Rape', led to widespread public debate. The Police series laid new ground for British ethnographic documentary in that it opened up issues which the subjects themselves were i "i aware of, thus instigating a self-discovery process.
During the 1990s, the institutional docu-i". i itaries of Diane Tammes also reflected a direct-cinema style. In her series, Casualty 'Channel 4, 1991), for example, an institution ■■• described with relatively few edits or
guiding commentaries. The film reflects that variant of the direct-cinema philosophy associated with the British National Film School, which suggests that the longer a sequence is allowed to run the more 'truthful' the images are likely to be. Tammes, however, has admitted to becoming 'less strict' with herself as her career as a documentary filmmaker has progressed. While she suggests her early work was 'purist', in that, 'the camera was there simply to see how people interacted', subsequent work has often involved dialogue with subjects from behind the camera. 'Truth', she argues 'is about what people actually come away with at the end of the film'. The problem is that 'everyone who makes a film is just putting their own truth on the screen' (Tammes, 1992).
An interventionist style was also evident during the 1990s in the work of Molly Dineen, and in particular her highly popular series on zookeepers, The Ark (BBC, 1993). Originally aiming to make 'a film about a zoo', Dineen ultimately told 'a story about a British institution undergoing radical change'. Dineen explains how, as financial problems loomed at London Zoo, it became a case of 'suits versus beards' - in other words, 'the management men and the financiers against the scientists and academics' (The Guardian, 1993: 53). As the crisis came to a head, she documents how problems of 'human resource' management began to take priority over 'animal' management. Consequently, her style of filmmaking became increasingly investigative, critical, and interactive, with Dineen, from behind the camera, probing to uncover the layers of political intrigue underlying, for example, decisions on redundancy and restructuring.
From the mid-1980s, a further step away from the philosophy and practice of 'objective' or 'purist' ethnographic documentary was signalled in works which actually made overt the construction of the cinema verite film. This is nowhere better represented than in works by the celebrated film-maker Nick Broomfield. When in 1990, for example, Broomfield (with Joan Churchill) made a follow up for Channel 4 of his earlier
278
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
,EARTHING WORK AND INSTITUTIONS THROUGH ETHNOGRAPHIC DOCUMENTARIES 279
institutional film Juvenile Liaison (1975), he brought to the fore his own problems as a filmmaker, especially that of tracing the original participants and gaining their renewed consent. This style of focusing on what goes on behind the scenes was subsequently used by Broomfield to great effect in bis film, The Leader, The Driver and the Driver's Wife (Channel 4, 1991) where, exploiting the off-screen drama, Broomfield documented his difficulties in trying to secure an interview with the South African right-wing politician Eugene Terreblanche, the result being a black comedy of trial and tribulation - one somewhat reminiscent of Michael Moore's attempt to hold an interview with General Motors CEO Roger Smith in Roger and Me (1989). This style of filmmaking could not be further removed from Broomfield's earlier work, for here the camera is anything but anonymous. Broomfield has remarked that, as his filmmaking style has 'matured', he has felt increasingly hide-bound by having to keep to a 'pure' cinema verite approach, for in the filmmaking process, 'things don't happen when they are supposed to happen'. Broomfield argues that the associations you make throughout a film are often much richer than simply documenting what happened. For Broomfield, therefore, 'the notion of objectivity is no longer tenable' in ethnographic filmmaking. By keeping the 'pretence of objectivity' and thus 'failing to share personal experiences', he argues that the audience is 'less able to evaluate your final product' (Broomfield, 1992).
Video diaries and video blogs (1990s-present)
The important film makers of the future will be amateurs (attributed to Robert Flaherty, c.1925, by Jean Rouch, 1992)
Since the mid-1990s, one of the most popular forms of television cinema verite has seen both professional film-makers and also members of the public use 8 mm video camera technology to record personal 'diaries'. It can be argued that the personal documentaries which accrue from these diaries reflect many
of the ethnographic qualities I
verite pioneers sought in terms dit
authenticity.
In work and organizational setti
of the video diary form has bee i i u
employed in the Channel 4 sen    i,,,,
Edge. Films in this series frequen'b r il
form of covert ethnographies broa
the title Undercover. An early i , i,
*-'1'jc nj
a programme employing this method, on that aroused considerable media nl i „ was Low Paid Work  (1993), i,    lt. ' former BBC film-maker Sima three months undertaking covert < ■ ■ 11 < ■([,. in shops, factories, restaurants, i homes throughout Lancashire. Com.ciilina a camcorder in a shoulder bag, Ray filmed the situations she encountered as ,i >,n ,, Asian woman in search of ei m i-'-m.-t and accommodation. During the    i in ,, observation, all the jobs she n i i « i j secure were remunerated at less ihan the legal minimum rate, employers showing scant regard for the law. One of the lowe1- \r. id j >h Ray obtained was, in fact, obtain, .i I i uu-h the local Job Centre, a State-run ij-. . Through covert observation of the relationship between employers and emj method that would, perhaps, today fan ,■ gain permission from a university research ethics committee -Ray's film portray •■ mi 1' the feelings of hopelessness and depicssion experienced by low paid workers, in incase during a period of high unemployment. The film drew a sharp contrast be ■ .■ .-l :i m low paid worker's struggle for ex:--, n.i iJ the luxurious lifestyles enjoyed by many employers, these contrasting images being silhouetted against a background in \- \\ i i * State frequently failed to prosecute i r'f1"' 1 who broke the law, and where the Govi -111111 was concerned to veto European C    nil 1 ,J legislation on minimum wages.
The most ambitious British video diaries project, however, has undoubtedly bee 1 1 * BBC's Video Nation series, a television project in social anthropology and aud i-1.r interactivity. Starting in 1993, the HI-' has encouraged people to record their h"-on video, with a subset of these v 1 1
, DCjjig shown on television. Originally CtJt[ ! 1:1 fh>' Community Programme Unit lf" u>r  2. and founded and coproduced 1" "  , 1993 and 2001 by Chris Mohr and 'Vii^i K*«i llu-' series initially involved , ,t gioup of fifty people, selected '' ,  , usstlie UK, being trained in the rncorders and invited to record /.i-K of' their everyday fives during the IS'"'i.  ul a year. The BBC team selected 1  is I lin these recordings and presented \\ ■[ 1 ,1   short films immediately before a,  ,i! /,'.;: More than 10,000 tapes were 1 in ..l.-ly ■ submitted to the BBC, from i| roximately 1,300 films were edited j ,| «n 011 television. Viewing figures were high (1 to 9 million) and led to i|,d scries of Video Nation films, such Js W..i v'Shorts, Hong Kong Shorts, Com-11, • 1 % h , Bitesize Britain, and others. In ■>[ 111 P.IIC 2 decommissioned the series limit: a website from the archive at it n i.'v vw.bbc.co.uk/videonation/. As the ,«(■ isi r shows, a significant proportion of the Video-Nation contributions concern either ,|. i-.-l itionship between people and their n—>l.-» inent or personal commentaries on the organizations they work for. Such films il 1 iiHIect extremely intimate and emotional impressions of the relationships between people, work, and organizations.
I mil mound 2003, a Web-based continuation and expansion of video diaries has t-11 ji».-i in the form of video biogging or Ifjj ng'. During the mid-1990s, when the to.'io Nation project was being developed, the bandwidth and processing power required 1 . >erate video online made it an extremely ■li 1 Mlt operation for most amateurs. With the advent of greater computer and connection speeds, however, vlogging has witnessed gnificant increase in popularity. For •■i. nple, the Yahoo! Videoblogging Group
I i' sits membership increase dramatically in 2005, while the most expansive video-shanng site to date, YouTube, founded
ii February 2005, was launched publicly "■l'"een August and November 2005. In "nr :thc ISBC launched its first official ■ id. o.blog in October 2006. Currently, many
open source content management systems, like WordPress or Drupal, enable posting of video content, thus allowing bloggers to host and administer their own video biogging sites. Furthermore, the convergence of mobile phones with digital cameras now allows publishing of video content on the Web almost as it is recorded. While the potential of vlogging for academic research into work and organizations appears exciting and significant, in comparison to earlier forms of ethnographic filming, a formal methodological philosophy is yet to be established. However, in terms of practical impact, The Economist on 15 March 2008 - in a side-bar titled 'On strike, virtually' - cited examples of management being under pressure from 'virtual workforce protests' in a piece that highlighted the use of video biogging for orchestrating protest and dissent.
Ethnographic documentary, realism, and representation
It has become more difficult to think of ethnographic films as definitive representations of events, independent of the process that produced them, and ethnographic film-makers have begun to look upon their work as more tentative forays into cultural complexity [and as] parts of a continuing enquiry (MacDougal, 1978:415).
Having described the philosophies and aims of ethnographic documentary, and also the forms it has taken in producing studies of work, institutions, and organizations, we finally examine the integrity of the approach in relation to methodological constructs of 'representation' and 'reality'.
In contrast to the philosophies of Vertov and Grierson, a key argument of recent theoretical writing on film-truth is to regard the 'truth' claims of the genre as in some way 'fictional'. In contrast to the early 'purist' assumptions of direct-cinema, which rest upon the Cartesian belief that reality enjoys an independent existence 'out there', current methodological assessments of the professional ethnographic documentary suggest that its products are in many ways as artificial as those of popular cinema. It is argued, for example, that
280
THE sage handbook of organizational research methods
through editing and associated processes, both genres come to contain similar elements of plot, character, situation, and event (Renov, 1986; Hall, 1991; Nicholls, 1991; Nelmes, 2003). As such, both offer cultural images, which present the viewer with challenges or dilemmas, build heightened tensions and dramatic conflicts, and ultimately terminate with resolution and closure. In sum, it is claimed that ethnographic documentaries operate with reference to a 'reality' that is constructed, one that is, above all, the product of signifying systems.
In being so constructed, we find the signification process of ethnographic documentary, like that for other fictional accounts, being subject to scrutiny over its role in (re)producing hegemony. The claim of film-truth to be accessing some privileged reality 'out there' becomes treated itself as an ideological effect (Nicholls, 1991). Acknowledging this constructed basis serves to undermine, for example, the claims to moral superiority of, for example, Vertov, Flaherty, the British world-of-labour filmmakers, and the proponents of direct-cinema. It serves also to suggest that cinema verite offers access to a world rather than to the world. While cinema verite films appear to direct us toward the world, it can be argued that they remain constructed texts, offering representations rather than replications. Such representations reflect truth claims not only of what we discover in the social world, but also of what interpretations, meanings and explanations can be assigned. The social world is made manifest through agencies of external authority, through representatives. What we experience is an historical world made available through picture windows in which social practices are telescoped, dramatized, and reconstructed.
One of the main hegemonic products of this process is our impression of encountering the cinema verite world as if it were being presented for the first time. The world made available to us is one we perceive as Newtonian and Cartesian; a world comprised of reason, realism, and common sense; a world waiting to be discovered through advances
in technology. We discover that
take to be the only thing we ca
We are presented with a universe   ■ (-|,
within the framework of its repn mi, u ,
In this representational process, clem™,.
of style, structure, and perspei li t |. ,i mix. When the filmic argument i i1 ■ .[ we move beyond the factual wi <|.| ,,,
of the construction of meaning t i .,.....
system of significations. In so < hi. never experience a pure con between evidence and perspectivi become fashioned by arguments.'- . i j turn rely on strategies and convi i l-m | their accomplishment (Shorter, 19'
The so-called methodological 'nli i ,« of the ethnographic documentary c n | u fore, be read as an ideological re] which operates in the guise of •niu:i.,l sense'. Barthes (1974) calls this insti 111 nr ||, enforced nature of representation I ■ degree style', for it adopts a p  [■„ ,| innocence and neutrality in the systems which provide the found; turn  11[ which institutional perspectives   ir -■.ii themselves. The direct-cinema docu i nentancs of, for example, Wiseman and (iiiul ir never as neutral as they appear, '.•< -lit-, embody a distinctive view of i ■ i ilti i-like hospitals and prisons in a way which reflects strategies of resistance over u i u j anistic and bureaucratic logic.  I fie   I i-dimension' (Lukes,  1974) of this   u y monic process is that which is aci o!i . In without overt commentary. In ell m-i.nK" documentary, we tend to perceive nr I -particular arrangements of sound and image. Objectivity is constrained by decisions of what should, and should not, merit hcuig commented upon.
In sum, these representational and ideological arguments suggest that ethnographic documentaries always present a truth i n k ■ than the truth. Cinema verite films o I J ' authorial voice and, as such, what we experience is less the world 'reproduced' as 'represented'. This is so, 'even I It-evidence they recruit bears the authenticating trace of the historical world itself (Nichulk, 1991: 118).
"éarching work and institutions through ethnographic documentaries 281
CONCLUSION
j, tnl£ objective-subjective stuff is a lot of Itshi* E don't see how a film can be anything (subjective (Frederick Wiseman, quoted in Levin,
■y, .|   basis for considering its potential
sr 11 (   inizational research, this chapter has
■ Un. 1 -I scrrption and analysis of a range of ,„i; .| hie documentary filmmaking styles, i, |v ii Hi these styles have followed the ■, nlii osophics of Dziga Vertov and John (ei,. i. i i advocating aprogressively 'direct'
ni; ■, tui al^ approach to filmmaking in which "everyday events are recorded without the ., |, i ly: cinema-based props of scripts, staging, lighting, and so on. Proponents j Ii i Hy explain their development by rccotn se to n dcterrainist theory in which tech-ijln-i . I innovations (e.g., mobile 16 mm
."I luilugy with synchronized sound in the L .ii i i60s and 8 mm video technology in the | .1,-    provide for the capture of increasingly
nn'iiisive images through the eyes of a
[ifi ipl oric fly-on-the-wall.
I Li1 chapter has addressed this technical . -.I uylistic history at two main levels, research methodology and sociological anal-, > s I ,rr the former, film-truth may appear in i,11 ■ appealing to organizational research in that it appears to promise progressive r. iii. i on in levels of obtrusiveness and reactivity in the observation process. Also iK'lliodologically, cinema verite seems to offer::: a textual layer additional to, yet different from, those employed in traditional organizational research investigations. The iH";estion is that a film record can add to t ■ diversity and richness of data collected on, lor example, work, occupations and institutions. As such, there appears potential fur ethnographic documentary to contribute '■ spirit of triangulation through providing i i e 'immediate' images than can be obtained other research methods. 1 his methodological review, however, was m turn complemented by a critical sociological analysis of the relationship between
i' lism and representation' in ethnographic
documentary. This led to an awareness of significant shortcomings in the technological determinist argument, which underpins much of the methodological argument for cinema verite. Initially from the standpoint of method, but increasingly from that of deconstruction, it became apparent that an equally appropriate metaphor for fly-on-the-wall documentaries could be 'fly-in-the-soup'! Critical reviews of the ethnographic documentary, together with work on authority and knowledge, proved influential in directing the argument toward an appreciation of the ideologically enforced nature of representation. In the process, earlier methodological concerns with overcoming obtrusiveness became defused by discussions of the relativity of authorial presence and the ideology of 'common-sense'. Awareness of the ideological effects of cinema verite helped deconstruct many of the methodological assumptions upon which 'purist' approaches were founded. It was argued, accordingly, that cinema verite documentaries operate with reference to a reality that is essentially the product of signifying systems.
REFERENCES
Age!, H. (1953) 'Esthetique du cinema', Positif, 7, 45-52.
Aitken, I. (1990) Film and Reform; lohn Griersonandthe Documentary Film Movement, London: Routledge.
Aitken, !. (1998) 'The Documentary Film Movement', in J Hassard and R Holliday (eds) Organization/Representation: Work and Organizations in Popular Culture, London: Sage.
Aitken, I. (ed.) (2005) Encyclopedia of the Documentary Film, London: Routledge.
Armes, R. (1966) French Cinema Since 1946: Vol II: The Personal Style, Amsterdam: Ellerman Harms.
Barnouw, F. (1975) Documentary: A History of the Non-Fiction Film, New York: Oxford University Press.
Barthes, R. (1974) 5/Z, New York: Hill and Wang.
BBC (British Broadcasting Corporation) (1983) Arena -'The GPO'.
Beveridge, J. (1979) John Grierson: Film Maker, London: Macmillan.
Broomfield, N. (1992) (interview with) 'The Late Show',
BBC Television, March. Calder-Marshall, A. (1963) The Innocent Eye: The Life of
Robert J Flaherty London: W H Allen.
282
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
Cameron, I. and Shivas, M. (1963) 'Interviews', Movie, 8: 15-22.
Flaherty, F, (1972) The Odyssey of a film Maker,
New York; Arno Press. Forsyth, H. (1979) John Grierson: A Documentary
Biography, London: Faber and Faber. Graef, R. (1992) (interview with), 'The Late Show', BBC
Television.
Guardian, The (1993) 'The Ark in the Storm', 17 January. Hall, J. (1991) 'Realism as a style in cinema
verite: critical analysis of primary', Cinema Journal,
30, 424-46.
Heusch, L. de (1962) The Cinema and Social Science,
Paris: UNESCO. Issari, M. and Paul, D. (1979) What is Cinema Verite?,
Methuen, NJ: The Scarecrow Press. Lawson, J. (1964) Film: The Creative Process, New York:
Hill and Wang. Leacock, R. (1992) (interview with), 'The Late Show',
BBC Television, March. Levin, R. (1971) Documentary Explorations, New York:
Doubleday.
Lipscombe, J. (1964) 'Correspondence and controversy',
Film Quarterly, 18(2): 60-8. Lukes, S. (1974) Power: A Radical View, London:
Macmillan.
MacDougal, D. (1978) 'Ethnographic film: failure and promise', Annual Review of Anthropology, 7: 402-25.
Manvell, R. (1946) Film, London: Macmillan.
Morin, E. (1956) Le Cinema ou L'Homme Imaginaire, Paris.
Natta, E. (1963) 'Pro e Contro il Cinema-verite'-/
del Cinematogralo, November, 415-20. Nelmes, J. (ed.) (2003) An Introduction to Film
London: Routedge. Nicholls, B. (1991) Representing Reality: I
Concepts in Documentary, Bloomington,
University Press. Petric, V. (1978) 'Dziga Vertov as Theon'
Journal, 1: 41-2. Renov, M. (1986) 'Rethinking documentary: t
a taxonomy of mediation', Wide Angle,
71-7.
Robinson, D. (1973) The History of Work
New York: Doubleday. Rollins, J. (1985) Between Women: Domrsln
Their Employers, Philadelphia: Temple hn
Press.
Rotha, P. (1936) Documentary Film Art, I nHon and Faber.
Rotha, P. (1967) The Film Till Now: A Survey ol
Cinema, London: Spring Books. Rouch, J. (1992) (interview with), 'The Late Show
Television, March. Sadoul, G. (1940) Histoire d'un Art: Le (
Flammarion.
Shorter, J. (1993) Cultural Politics of Everyday We,
Buckingham: Open University Press. Tammes, D. (1992) (interview with), 'The Late Show',-
BBC Television, March. Winston, B. (1979) 'Documentary: I think v;c dre
in trouble', sight and sound', International Film '
Quarterly. 48(1): 2-7.
PART III
Strategies: Approaches to Organizational Research
Focusing on approaches to achieving research aims, illustrating links between topic, aims, strategy, analytical framework, and theoretical development, and also demonstrating the range of choice and degree of creativity as well as technical knowledge underpinning research strategies
Cross-Cultural Comparative Studies and Issues in International Research Collaboration
CROSS CULTURAL COMPARATIVE STUDIES
329
Mark F. Peterso
INTRODUCTION
The methods issues in cross-cultural organizational research are linked with the distinctive characteristics of the phenomena studied, the theoretical and metatheoretical perspectives taken by researchers, and the social process of collaborating across culture and distance. Whereas most organizational research can be simplified by assuming that the phenomena encountered in one's own part of the world are all that matter, cross-cultural research draws attention to phenomena that are sometimes different in nuance and at other times profoundly different. The methods for studying culture, the choice of theories and constructs, and the intellectual traditions in which researchers understand them vary not only with personal idiosyncrasies, but also with the societies in which researchers are geographically located. The need to access diverse populations to do global research
means that researchers increasing ml advantage in forming collaborate j"iL|>», which are challenging to participate in and' to manage due to communication < variability in preferred methods, t traditions, and global differences in:■* ir-situations and personal lives.
The aim of this chapter is to id< 11 ■1 n, r controversies in cross-cultural or^. n i'.i mp research methods, particularly those on questionnaires, as well as in the social process of doing such research. I will firs < ■■ i sider basic technical issues that prodm -1 n„ ' ing controversies and common misiiu. c standings. Since even researchers who si uf basic methodological perspective rarely .i- ■' about all of these issues, international teams "-struggle with technical controversies. A.I.a tion then turns to methodological implicit. r.-of perspectives on culture and controversies in cross-cultural research. Third, the inl. of ideals about how to conduct cm—-al" 1J|
1 m
: with the researcher's cultural ti.onal situation will be considered, I ,ij>s to work successfully with a Inl i ofialresearch team will be suggested. P ,(-i pace constraints, readers are referred III jr studies of cultural differences nil infor example, hierarchical moder-tLii i - r'" ^on> mu'ltgr0UP structural models, i ■ i deal linear modelling, and also for using qualitative methods (Easterby-Siilh >iid Malina, 1999; Raudenbush „, II *, 2002; Van de Vijver and |..iri„ I9?7)-
Technical issues
I, i, t nit-: equivalence and level of analysis mi "Mil l,e greatest technical challenges in Uiral research. Equivalence issues p.t ii researchers who wish to compare one ,,| 'i. i tion, say an attitude like commit-iin.ii, ...toss several nations based on some II lj i: > ally meaningful aspect of national Culture, such as individualism (Gelade et al., J i ir 11 ere, we find that we are also compar-ngiiages, thought patterns, social norms, n:i Jimy environments, social structures, in, pi ■ iably other issues of which we are ji .'..in. Consequently, the ability to control hi In ors that confound the substantive '" compatison of interest with the degree of ■lui'si ■ i that is typical in domestic research is only possible when comparing a few select phenomena in small numbers of nations ■ i in - 2008).
ri .'hellion equivalence The technical literature concerning equiva-I i -l a irts from careful translation checking (Bmhii et al., 1973). For example, back " ' ■'.'. mrequires having one party translate ■'i .i", another translate it back to the source language, and then both to engage "Ifi a piocess of comparison and correction. r'-ri'i , translation is achieved by having *■ ' I'-'iple separately translate a survey ilto a target language, and then either the 'I I i.n or others compare the translations '"i make, corrections. Back translation is ''■•re consistent with current norms in most
parts of the world, but the decision concerning which to use is an issue of power as much as research norms. In practice, back translation puts power for making final judgments about translation adequacy in the hands of the party who designed the survey, usually in English. Parallel translation puts this power in Lite hands of the parti es who are fl uent in the target language.
The translation procedure used is important, but is now seen as secondary to quantitative indicators of translation quality. In the early years of comparative research, documenting the careful conduct of translation was considered state of the art for achieving equivalence. However, greater confidence is now placed in indicators of equivalence in measurement structures. Equivalence is assessed by comparing reliability coefficients or by using the measurement portions of structural equation modelling programs (Van de Vijver and Leung, 1997). The availability of quantitative indicators means that the results of a pilot study to evaluate measure equivalence is likely to be more meaningful than are debates about the virtues of back translation or parallel translation.
Most discussions of translation equivalence in comparative survey research assume that the structures of different languages are similar, leaving translation adequacy as the main explanation for nonequivalence. However, languages have long been known to differ, often in surprising ways (Nisbett et al., 2001; Sapir, 1921). Many controversies about the implications of phrasing questionnaire items in the generic (English) organizational literature concern the distinctive characteristics of English, such as whether pronouns are singular or plural. For example, how my superior treats me may be different from how my superior treats us. In comparative studies of leadership widi cofieagues in Japan, we faced the problem of communicating such differences in Japanese, where the subject of a sentence is often implicit, especially when the subject is personal like I, me, we, or you (Peterson, Maiya and Herreid, 1993). The awkwardness of explicitly representing
330
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
CROSS-CULTURAL COMPARATIVE STUDIES
331
the 'me' versus 'us' distinction in Japanese is likely to evoke responses different from those one would expect from survey research in English. Researchers who collaborate in a multiple-language project also need to discuss whether their goal is to extend work that is established in one language into another. In that case, the success of a translation might be evaluated relative to the language of origin. Where a measure is being more substantially redesigned, as in a study that redesigned role stress measures (Peterson et al., 1995), or a new measure is created for global use (Leung et al., 2002), overall similarity among the translations being studied may be more important.
A full treatment of language structures is outside the scope of the present chapter. However, it would be surprising if any survey measure that has been carefully tailored to promote both reliability and validity in one language will have precisely the same meaning in languages from radically different language groups. Still, languages tend to be sufficiently analogous that effective communication is typically possible (Pike, 1982, pp. 132-133). A host of comparative survey-based projects show that measures which were designed to have optimal psychometric properties in one language can be effectively translated into many others with only a modest loss of precision. The reduction in areliability coefficient from 0.80 for a measure in one language to 0.70 in another adds imprecision, but such differences are often treated as sufficiently manageable to allow comparison (Leung, 2008).
Conceptual equivalence Issues of translation equivalence merge with issues of conceptual equivalence since both are rooted in language. Concepts correspond between societies to different degrees. For example, Farh et al., (2004) report an inductive study of organization citizenship behaviour (OCB) in China. Their analysis indicates that some constructs, such as 'taking initiative', appear to be almost identical to US concepts, such as 'conscientiousness'. Other Chinese concepts, such as 'helping
coworkers', shows nuances of a I s m this case 'altruism'. However,. as 'interpersonal harmony',are mn.L , „ linked to China's heritage. St .p|  , ' ' reviews indigenous managemein ' from understudied parts of the woi I   |,,, ^ roughly correspond to ideas fan i ir „ {,' English literature.
As with measure equivalence, di , i|„ of conceptual equivalence are ii   i i , | framed in statistical terms. That is,'Lin,.-llr j '■ equivalence is evident when am i in h into a set of relationships with othci mcaiimn in theoretically meaningful ways, • i, h ... nomologicaf net, in the same way in different parts of the world. When viewed i I.., , discussions of conceptual equiv^^e ,, typically reflected in hypothesis dc I   i, L„i and testing rather than in validity ; i     ^llt. lt Data analysis techniques used tc t , n,| such relationships range from shu > im_ i, ,| regression slopes are equivalent in , i i t ii nations to applying even more , m. j multigroup structural models (Van i \n-,... and Leung, 1997). Controversies about the technical merits of different methods for assessing cultural differences in relm u-i -I among measures concern the strcnglhs and limitations of regression, structural irijimn modelling, and hierarchical linear mi ski i j (Raudenbush and Bryk, 2002).
Response bias
Language differences suggest that ilem phrasing is likely to affect inter-item covariancc, reliability coefficients, or measuren u 11 - n .l I equivalence. However, language equivalence issues and culturally linked communiciilion style preferences can also affect the teni I. k y to use high, low, or intermediate values in responses (Smith, 2004). Significant anecdotal evidence has been collected concerning cultural differences in response patterns (Van de Vijver and Leung, 1997). Nal. 'i. differences have recently been explairii i 11 part by a link to Hofstede's (2001) concept of power distance (Smith, 2004), A'.c responses to surveys in large power distance nations are generally found to be m-"
.  . man those in small power distance i^i"'
iii.i
11 jion to cultural patterns in response lencics, nuances reflecting degree in , A words can create language differ-s |   particular items. The sensitivity of . j to the phrasing of alternatives is '" II , i grazed for English-language surveys *   s   | al, 1974). However, adverbs of i ■i^l i1 ray not match precisely between ■ 11! i n s. For example, in a set of response r, .ii 'es ranging from 'strongly agree' .„ i.i lily disagree', the English 'slightly' „ee uii convey slightly more agreement / p |l   Chinese you diem agree. Similarly, I,  I i i ;ihg of the question stem may have II unt degree of positive, neutral, or negative connotation in two languages. In a simple question like 'How hungry are . ,n       Dutch translation that represents |„ n_*. ■ «s hanger relates to the upper ranges of this phenomenon (overlapping with starvingl, whereas the alternative Dutch trek relates more closely to the more ordinary dc »ves o! hunger experienced in developed nations,
One solution has been to design measures using forced-choice alternatives or rankings that make response patterns difficult or impossible. For example, the Meaning of Working project (1987) asked respondents to rank a set of fourteen goals to avoid possible .i iturlil differences in the propensity to say that: all work goals are very desirable. Even i respondents would like to say that each ii-ial is the most important, the answering ■.' item only allowed a respondent to rank one as the most important. Another approach has been-to adjust responses to Likert-scale mea-'ii es! using within-subject standardization or i can-centring. Within-subject standardization requires subtracting the average for each uspondcnl across all items from the score of each item, and dividing the result by the st ndard deviation of the respondent's answer ■u. 'Oss all items (Hofstede, 2001). Some researchers have, then, helpfully rescored the result so that it corresponds to the original response alternatives (Hanges and Dickson,
04). Mean-centring takes the first step of
subtracting each respondent's overall mean, but leaves out the second step of dividing by each respondent's standard deviation.
Unfortunately, these solutions to the response pattern problem have significant limitations. They all produce dependencies among items that limit the reliability of indices that combine items (except when adjustments are made to Likert-scaled items after indices are formed), and create covariances among the adjusted items that may not be meaningful. Consider, for example, two questions from the Organizational Commitment Questionnaire (Mowday et al., 1979) about which most people are likely to have similar attitudes; 'My values and the values of this organization are quite similar', and T am proud to work for this organization'. If respondents were asked to rank these obviously similar items as '1' and '2', their answers would necessarily have a correlation of -1.0. A similar negative correlation appeal's if respondents answer these questions using Likert scales, and the researcher then standardizes the responses within subjects as described in the preceding section.
The extent to which such within-subject adjustments are troublesome is affected by two considerations. One is the number of items, such that the dependencies produced are smaller as the number of items increases. The other is the original covariances among the items, such that the tendency to reduce meaningful variance is increased as the original covariances increase (Saville and Willson, 1991; Smith, 2004).
Sample equivalence
Cross-cultural researchers also face the problem of sample equivalence. Referring to the example of commitment, one would not want comparisons to be confounded by other demographic characteristics besides nation. Demographic confounds are readily resolved in domestic research by statistically controlling for factors such as age or gender before comparing commitment levels across organizations. Such controls, though, typically assume that any implications of age and gender are reasonably consistent
332
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
CROSS-CULTURAL COMPARATIVE STUDIES
333
across organizations. Such an assumption is generally reasonable within a single nation, but is less reasonable across nations. Cultural variability in the meaning of demographics is reflected in showing respect for age and gender (Peng and Peterson, 2008). For example, in Japan, only those women who are extraordinarily committed to remaining employed tend to continue working for a major employer beyond their mid-20s. It is also unusual for a man to remain with a major employer beyond age 55 unless he holds a senior management position. Glazer and Beehr's (2005) study of nurses in four nations indicates that simple statistical controls for other sample characteristics such as occupation or employer type are similarly problematic where the social status or employment conditions affecting a group vary by nation. The consequence is that implications of demographic characteristics may well need to be considered, but the limitations of the simple statistical controls used in domestic research must be recognized.
Data collection process equivalence National or cultural differences can also affect the meaning of data collection procedures. In some nations, postal systems work efficiently, but in others mail surveys can be lost or delayed. In some places, personal interviewers who come to a private home are viewed as normal visitors, but elsewhere they are viewed as potential threats and may be at risk themselves. In some nations, research that includes separate-source performance evaluations or access to personnel records requires respondents to sign a consent form, but in other nations, it does not. In a study of local governments, I found that if a Japanese organization agrees to participate, then nearly all employees will voluntarily participate as well; in the United States, however, organization consent and employee consent tend to be separate issues. Superficially equivalent data collection procedures are not the same as substantive equivalence, and a degree of nonequivalence in the data collection process
is likely in comparative projects (Lincoh Kalleberg, 1990).
Adapting to equivalence constraint*
How   do   we   solve   issues   of In equivalence? The magic of randoi i, (Campbell and Stanley, 1963) is ,pil,|||!!i out of reach in cross-cultural ,iVcr|t We are left with the partially satisfdctorvf recourse of making statistical ; In to our measures, matching resj..| . n. best we can, and interpreting I hi ].,.„ implications   of  uncontrolled contextual^ differences.   Precise   equivalem ■ .-hiv way to sufficient comparability i    ,\, w reasonably meaningful comparison m," caveats and needs for replicatioi ■; jim 2008). Subjectively evaluating i m.i,-give meaning to comparisons c; n i..|w, even a quantitatively oriented rev ikI.. (j be something of a linguist, antlnopol j,-economist, sociologist and psycholugi .> n-to invite collaborators from other discin m.i (House et al., 2004).
Level of analysis
Is culture a characteristic of a society or at its members? As a theoretical question,! >m1i. deal with this again in the following sei mih but it also has a technical side. '"
Many cross-cultural projects collectsu. ,tj data from individuals then aggregates I. -j data to create measures that are typii :i l> used to represent a national collective, The; mechanics of creating nation scores is sin p c take an average across a set of ileitis .i u across a set of people. The basis for in i<i i|s    * I: decisions about what items to combine tot create nation scores, however, is Irequ.- itl> misunderstood even by experienced scholars. Researchers typically create measures based on individual-level measurement stracfi i^s> then perform a second set of analyses ti whether there is sufficient variance ai i" it units at a higher level (groups, organizations,, nations, etc.) to aggregate the individi .il-level measures to that level (Peterson n.J Castro, 2006). Hofstede (2001), how"'-notes that the correlations and covaru'ii.."-between items on which measure decisioi ■
he i J <
Jiffer depending on the level of analysis R n fhey are calculated.
ter, for example, a hypothetical study -cks to create measures to represent I, ,v.p. > 'in °f individualism and collectivism ■j   fjujerman et al. (2002) identify in la-analytical review of individualism-"llectivisnT. In principle, one might be "lik     construct ten items that provide "lt, ,,i  I indicators of individualism and .,      len relevant to collectivism. It is f,.   hi r reported that, when such items .|C , i l.scd at the individual level, two or m ,    ustructs emerge that are internally ,tm gJ> Qv* and on4' modestly correlated iln .ii: another (Kirkman et al., 2006; Oyserman, et al., 2002). That is, it is not , ii .ii, I for one person to express both i kI '.ii iijlistic and collectivistic values, or • ulici I rexpress neither individualistic nor 'tic. values. However, when such i o aggregated to the nation level, and £ortelations among these items are calculated, i 'j i c i s tend to combine into one rather than two dimensions (Hofstede, 2001; Kirkman ft .1, 2i 06). That is, societies in which many Ippk _■ < press individualistic values also tend hi be societies in which few people express u lecuvistio values.
More generally, measurement structures arc often affected by level of analysis (IV.imhi and Castro, 2006) and need to In1 js.issed at each level at which fhey in. In be used (Inglehart and Baker, 2000). Whsther oi not individual-and nation-level n i tsurement structures differ depends on ■In. topic of concern. Personality measures ■o. m to have similar structures at both levels 'f'-lade. et al,, 2006), while measures of iallies often seem to differ by level (Hofstede, 'I'll), The measures of managers' use of social; structures as sources of meaning show some tendency to collapse into a smaller "umber of dimensions at the national level '■"ipared with the individual level (Smith et al., 2002). Decisions about what items s' 'aid be combined to create measures of I -i ional perceptions, attitudes, or values related to culture will differ from decisions ■ii : lit what items should be combined based
on nation-level measurement structures to create nation-level measures of culture.
Perspectives and controversies
The technical issues of equivalence and level of analysis tend to be the focus of most commentary on cross-cultural research methods, but these are expressions of more basic issues concerning the nature of culture and its representation. Cross-cultural studies have progressed from the 1980s when culture was treated as an amorphous concept evoked to describe differences between nations without providing an explanation (Child, 1981; Roberts, 1970). Subsequently, controversies have centred on which ways of understanding and representing culture are most useful. These debates often concern the adequacy of Hofstede's (2001) conceptualization of culture as 'the collective programming of the mind' (based on the dimensions of individualism-collectivism, power distance, uncertainty avoidance, masculinity-femininity, and short-long term orientation), and the utility of cultural distance and country cluster research that is closely tied to these constructs.
Although these controversies continue, others have recently arisen. One set of controversies concerns the relative utility of alternative systems for categorizing cultural values compared to the one that Hofstede proposed. A second concerns whether organization studies will benefit more by comparing societies based on value dimensions, or whether it is better to compare belief systems, social structures, or concepts suggested by neoinstitutional theory or theories of institutional systems (see David and Bitektine, Chapter 10, this volume). Another is whether national culture analysis should be abandoned in favour of research about the values and the cognition of individuals, or whefiier collective culture characteristics can instead provide a better basis than surveys of individuals for drawing inferences about cognition. A further controversy concerns whether there is so much variability within nations, or such strong patterns of multiple-nation similarity,
334
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
CROSS-CULTURAL COMPARATIVE STUDIES
335
that other boundaries around culture should be considered. I will deal briefly with each of these after discussing Hofstede's view of culture and cultural dimensions.
Culture and the 'collective programming' metaphor
Many of the controversies surrounding Hofstede's project relate to theoretical perspectives on the nature of culture. Researchers often cite Kroeber and Kluckhohn's (1952) review of competing definitions. The metaphor of 'collective programming' does not derive from any particular hue of theory, although it suggests several. Currently, the many definitions of culture in use reflect major theoretical perspectives. For psychologists who study attitudes, values, or personality, Hofstede's metaphor draws attention to constructs that will have adequate psychometric properties in a given society, indicating that individuals use them frequently and understand them intuitively (Kirkman et al., 2006). The metaphor also evokes ways of tmnking that are shaped by the social norms reflected in what children are rewarded and punished for by their parents, teachers, and childhood friends (Rokeach, 1968,1973). For psychologists who are closer to theories of scripts, schemas, and cognitive structures, it implies that the way in which such ideas are structured are shaped by socialization (Markus and Kitayama, 1991; Nisbett et al., 2001). For linguists and anthropologists, the idea of mental programming invokes the Sapir-Whorf view that language is an aspect of culture that shapes what people intuitively understand (Sapir, 1921; Whorf, 1956). For sociologists and political scientists using survey methods, the metaphor suggests the sort of nation-level measures of societal norms noted in the preceding discussion of level of analysis. For sociologists and political scientists interested in social institutions, collective programming might call to mind the sort of influence that institutions like governments and churches generate (Whitley, 1999), as well as the influence of less tangible institutions such as normative systems (Kostova and Roth, 2002).
"Hofstede proposed the idea c programming at a time when ■ based theories were prominent in and sociology, and this idea ha.s |i n,,,, ..." frequently been interpreled i i ( '* value dimensions. Many of the , I , theoretical perspectives developed Luc,, not have been anticipated when Hon,- ,]K , ' sentedhis perspective in 1980. Th ■> ■ u ,.. tives, however, permit useful reft iim | tl of the 'collective programming' each of which is likely to produi    | ||f , responses to the issues discuss ,-i| m ti.| chapter.
Dimensions and nation scores ChallengestousesofHofstede'sn: li, n are many and well known. The use 0f these scores is criticized for relyi ir i _ ,, data about phenomena that arc  u mfc to change, for being collected ui i suit; company (IBM), for being based oi .'t ft that have little face validity, and I .[ I _|r^ too quantitative and missing loo n ut i the most important aspects of c i I i i 'ih responses are also well known. F who seek to explain culture tend lo consider aspects of a nation's heritage read     i w.-longer into the past than the date <■ ih IH" survey, studying a single company i.: n i U for many extraneous variables, the i   - ir.t have construct validity in relation to .i !:■ i.".-. external criteria, and the measures ate useful but are incomplete compared to i I'  11>. ■+ gained from learning a language and living and working in a society. Hofstede iii. and responds to these controversies i 11 li | th i the second edition of Culture's Coi • • ,'i i ' (Hofstede, 2001).
Alternative national culture dimeni ic i: schemes
The first edition of Culture's Consequences in 1980 established a paradigm for national culture research. One key character n I paradigm is that questionnaire survey s can bo used to collect individual data, whiel i ai e (hen combined based on nation-level m(. .'ii. i-i' structures. The dimensions measured dK indicators of values that charai i  i ■
i n Jims and institutions that societies ''' [c niiiigcnericsocialfunctions.Anumber 13 ,n fal I within this paradigm, but do so '' i'i I > by proposing alternative systems of I, ii'ii'-
■ ni.rBiost influential are projects based ■ 'it ,-iliwartz Value Survey (Schwartz,
I ,i | hi 'iwarl/ and Ros, 1995), the Riding a] Culture model (Trompenaars '.ijl nipden-Turner, 1998), the World Value ,i (InglchiiTl and Baker, 2000), and f: Leadership and Organizational i effectiveness project (GLOBE; \'\\    I al, 2004). The Schwartz Value r, w     -roject introduces seven nation-
,uu in tare dimensions, which contrast mih one another—conservatism-intellectual ,.i iniiii ir and affective autonomy, hierarchy-ri |||L i ii sm, and mastery-harmony. Nation-j",f| in is hased on data from secondary <](. i nl I ichers and university students are available for a number of these (Schwartz,
I n Siijwarl/. and Ros, 1995). Trompenaars j-j ll.mpden-itirner (1998) collected an (■( (. is.1 s : consulting-based data set that |. -ii it ei seven variables, the first five of •mi I ire based on the 'theory of action' i-«(.|m| cd by Parsons and Shils (1951); uriivci salism-particularism, individualism-„iu ii i litarianism, specific-diffuse, affective-i-iii il, achievement-ascription, sequential-svn. ■■n-iiic, and internal-external. Smith etal. ( 1 in- applied multidimensional scaling to this data^ set to identify two dimensions;
; I tarian commitment and utilitarian involv-ic it; contrasted with loyal involvement— an interesting nuance on the individualism-
i kclivissn theme. The GLOBE project
Mi use el. al., 2004) augments Hofstede's " i .'it by creating nine culture dimensions;
'«. ir-dtstance, uncertainty avoidance, future orientation, institutional collectivism, in-group collectivism, performance orientation, gender egalitarianism, assertiveness, and li 'i iane orientation. Inglehart and Baker (2000) identify two national cultural dimen-■K n i.;: traditional/secular versus rational, and
li ival versus self expression. I he illustrations of culture dimensions drawn from these projects should not discount
their significance as alternatives to Hofstede's original dimensions. Each project is at least as extensive in the number of individuals surveyed and nations represented, as was the Hofstede project. Each is also linked to a significant theoretical tradition, and they each provide alternatives within the paradigm of nation-level research about culture dimensions.
Alternatives to cultural value dimensions The paradigm of national dimensions based on values competes with a reasonably compatible set of alternatives. The social axioms model, for example, represents cultural differences in attribution propensities or systems of beliefs (Leung et al., 2002). Another approach renews a neglected theme in comparative sociology, organization studies, and social psychology concerning the analysis of societal differences in emphasis on a set of social structures including roles, rules, and norms that influence the behaviour and thinking of the society's members (Peterson and Smith, 2000 and 2008; Smith et al. 2002). This research identifies five organizationalrole categories— superiors, subordinates, colleagues, staff specialists, and one's self, as well as organizational rules and organizational and societal norms.
Individual level research on cultural themes
Several cross-cultural researchers have suggested that the field should move away from studying national culture dimensions. Instead of treating national culture as an indicator of a range of experiences that shape cognitive structures, characteristics of individuals should be measured using personal values surveys, and these measures should be used to test individual-level hypotheses. A recent review (Kirkman et al., 2006) suggests that many if not most studies that evoke Hofstede's (2001) conceptual basis and culture dimensions proceed in this way by linking the values of individuals, rather than national culture values, to various criteria.
The challenge of research into individual expressed values is that such constructs
CROSS-CULTURAL COMPARATIVE STUDIES
337
represent the more rationalized but less behaviourally irrfluentiaL aspects of cognition (Peterson and Wood, 2008). It is tempting to extend the value label too far by interpreting expressed values as implicit, taken-for-granted aspects of cognition. People do think deliberately and act based on conscious values or beliefs under circumstances that are unfamiliar, threatening, or emotional. The argument for representing national culture by aggregating responses to attitude and value surveys rests on the effects of socialization on such conscious thought. Unfortunately, using expressed personal values as indicators of culture has made it tempting for researchers to mistakenly equate culture with patterns of conscious thought. Cognition research since the 1970s suggests that most aspects of behaviour are driven by less accessible cognitive mechanisms (Kahneman, 2003; Nisbett and Willson, 1977). Indicators of the culture in which a person has been socialized may be better indicators of aspects of these cognitive processes than are the values that individuals overtly express.
Extending the national culture paradigm to individuals
The distinctions in meaning reflected in differences between the structures for measures of values at individual and national levels make it challenging to connect levels of analysis. Since the structure of values at the nation level differs from that at the individual level, it is inappropriate to conclude without testing that research findings at one level apply to the other. For example, if societal values of individualism and gross national product are associated at the nation level, then preferences for individualism may or may not be associated with wealth at the individual level. Nevertheless, individuals are certainly exposed to, experienced in, and socialized into the cultures of their nations, so it is reasonable to expect that societal culture does affect individuals. How, then, does one do research that connects levels?
Studies that connect societal culture characteristics to nonconscious cognitive characteristics other that! attitudes or values
typically   follow   cognitive , pi research paradigms (Peterson and Woqh 2008). Individuals are randomly , „ at birth to a cultural group. In e;i|    Pli |( . studies, a researcher then assigns |. ,n . more immediate experimental co study the joint effects of culture a l( ,ln experimental manipulations. Almoin-like any other experimental ma'!|u| |, )r individuals   respond   differenlly   to tk culture to which birth assigns them, culture can be treated as a quasi-ex . n,,. condition to which they have been exposed (Campbell and Stanley, 1963). \, r „, quasi-experimental context, the anal »»\, experimental manipulation is apiin.ii |t cultural groups are more comple1 hi „ ordinary experimental raanipulndon, and, those who participate in researcl atypical of their cultural group. Nev. i th k-experimental studies of culture and 1111.1 I n often treat the implications of natioiu, u.,t,iii, for structuring the cognitive pro< society's members as more releva values that individuals express in i.ur. v
Studies that include a method to i applicability of national values dimension* provide one option that can be i ■clul I ir individual level research that cvo! l>lI lur explanations. Brett and colleagues (IIn example, report several negotiati'i'i .u. 1.-, using Japanese and US managers attending executive training as participant. Iln-. researchers recognize that the j u i iriii% in their studies are not a randc of the participants' national culluics, but a sample of people who have found I ■ rr way into managerial roles. Nevertheless, the researchers recognize that their participants live and work in a national cullural environment. The researchers, therefore, >-i group differences with respect to pi 11 k ip. h" cultural values, but do not expect individual participants' expressed cultural values to act as mediators. Brett and her colleagues use their participants' cultural values as a torn* of quasi-experimental sampling r!n evaluate whether the relative positioning ft the national samples of managers is consistent with cultural theory (Jeanne M/Bretl.
| ,| communication, October 2007). This
!       I*     .ft is most evident among psychol-* .1  who emphasize cognitive structures, i Markus and Kitayama (1991) and
0 Pl -et al. (2001). However, a similar
^looking for ways to evaluate the v i all ' of society-level culture implications i idiViduals can also be found in research sizing values in the Rokeach (1968)
U i I 1 Di"
hst-(subculture, or region i, in ■ i ig individuals rather than cultural ,,, i| -, is not the only alternative to studying iii |i,i tlcultures (Peterson and Smith, 2008). .\|-li ni'ih some subcultures are typically a. Mil-zed as exceptions in the major compazine studies of nations, the significance ,- i-,i liiii-nation subcultures is sometimes J , ited even more strongly (Lenartowicz j nil Uoth, 2001). Projects advocating the | . n ■> s of within-nation subcultures have • _ii■] li .sized the use of regions for explaining mu.e in the values of individuals and i,i ..■ tudied relatively small segments of the ivi i Furtherworkofthiskindmustprogress v. i "i 1 these limitations to develop and
i. i n .iiesmeasures based on subculture-level
1 l is i eniont structures that are analogous i lime that have been used for develop-
ii. ii ition-level measures. This work also mi I-10 compare within-nation variability to lu v. i n-nation variability on a global scale.
\Studies of multiple-nation regions, ■nnii i mes referred to as 'civilizations' fin lglon, 1993), are another alternative i'1 .1 lalyzing cultural groups (e.g., Ronen,
I lii Ordinarily, however, multiple-nation t-_.i iii rare proposed as a way of identifying -lul-i cultural forces that supplement and r' il-.ui rather than replace analyses of
•rim it cultures.
International research collaboration
Handling the technical and metatheoretical v' 'in international research is a social
II i as well as an intellectual one. Two - i ■; that remain in the background of 11,11 domestic research surface in debates in
international collaborative teams. One is the way researchers need to interact in order to move between theory and experience. Polanyi (1958/1962, pp. 69-131) discusses this in his chapter about 'articulation' as a social process of explicating tacit knowledge. A second issue concerns bringing ideas, theories, and methods developed in one part of the world into contact with the system of meanings used in another (Peterson and Pike, 2002). These issues appear in international collaboration in part due to issues of conceptual equivalence; ideas that researchers from one locale find important are seen as less important or omit the concerns of researchers in other locales. Hofstede (1996) and Erez (1990) identify topical preferences in different parts of the world, and these also occur due to the cultural basis of intellectual traditions that influence preferred theories and methods (Weber, 1951).
Two other contingencies that affect international research team dynamics are the researchers' institutional context and their project aims (Peterson, 2001). Unpleasant politics can develop when viewpoints become tied to the images and identities of particular parties (Dutton and Dukerich, 1991). From the GLOBE project, Hanges et al. (2005) provide examples of how viewpoints and identities can become Linked in international collaborative projects.
Normative and descriptive international collaboration
The literatures about international research collaboration include both normative material about how such research should be conducted, and descriptive reports about the experiences of international teams. Normative commentary includes recommendations about how to avoid over-reliance on concepts and theory from one nation, particularly United States (Boyacigiller and Adler, 1991; Peng et al., 1991; Peterson, 2001), and how input from the locales in which data are to be collected should be obtained. Some recommended procedures focus on how to draw the most from researchers from different nations (Brett et al., 1997), while others focus on how best
338
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
CROSS-CULTURAL COMPARATIVE STUDIES
339
to learn from respondents in different locales (Morris et al., 1999). These approaches to de-centring international research are complementary.
The effectiveness of procedures for de-centring research is also now debated as part of a descriptive literature on international team dynamics (Hofstede, 2006; Javidan et al., 2006). A few international teams have been reflective about their dynamics. Easterby-Smith and Malina (1999) describe the conduct of an ethnographic project. Graen et al. (1997), Hanges et al. (2005), IDE (1981), MOW (1987), and Teagarden et al. (1995) describe the dynamics of very large-scale projects. Mezias et al. (1999) illustrate what can happen when US researchers study the global programmes of a US multinational organization. Tannenbaum et al. (1974) provide an example of how to incorporate different researchers' interpretations of the same data. Peterson (2001) provides a useful review of these projects. An analogous descriptive literature about handling process issues in international management teams is also relevant (Baba et al. 2004; Maznevski and Chudoba, 2000), Current commentary on international research collaboration suggests that normative models should be seen as alternatives. No model is universally optimal. The recommendations that follow draw both from the normative and descriptive literatures of international collaboration.
Research purposes affect collaboration At least four purposes affect why researchers become involved in leading or participating in an international collaborative research project (Peterson, 2001). A collaborative group is composed of people who are likely to have a variety of purposes, and groups as a whole are likely to vary in the level of agreement about purposes. Choices about the methods, concepts, and viewpoints discussed in the foregoing section are influenced by the purposes of individual collaborators. The distinction among the first three of the following purposes has been frequently discussed (Boyacigiiler and Adler, 1991).
First, research that focuses on > is constrained to repeat procedure ■ i, |    r 1 in the original research. To the exti u | ( ^ purpose is accepted, debates aboni i v _, and concepts derail the project, h ., fixed procedures are so central thai   ,, , ■ personal involvement of the re,<.,  \ ] .t group that did the original work can ■ | , (l to providing advice rather than din v ,| new team. An example would be cum .,, use of the Schwartz Value Survey i Vh.i 11, and Sagiv, 1995) once the me; ,i i become established.
Second, some research uses £ lis-vij,»i set of collaborators to develop in< i'm i|, research problem that is viewed a; This purpose characterizes partici' n n ^ bers of many research teams, altnn tl i v less often characteristic of a team : , i » In this case, it is the theoretical and technical views of the collaborators that are central, rather than the unique qualities of their cultural backgrounds, or 11   u il m which they will collect data. For :m<i| i- ji the time of writing, I am collabora in ■ mi'i j group of negotiation researchers to(.n n ih li-ideas about culture research methoi • \li c this knowledge is viewed as not bun.; L-,-rt on my cultural background (Metcalf ct al, 2007).
Third, international collaborativi -ii li frequently draws upon the cultural, and setting of collaborators. Tin i j collaborator is part of a group, eitf..'i h:i 1.1 of a personal intuition for how a i .■.-.■jivIi problem is likely to be best Formulate) in their setting, or because of Iho ability to collect data in a particularly -il u f This is consistent with integratin1: ti il" i~»ltr methods and perspectives. For example, the negotiation study that I am assisting in I' '' » researchers from several nations ■ 'i- I" "L an intuition about how negotiati >i 1 l-in their country, and whose coulribiiunii reflects the purpose of promoting thi pn |i ■ applicability in many nations.
Fourth, intended outlets and i influence   project   design   and   i ' iJ' Universities in United States and -ii""11' are increasingly driven by criteria : 1
"■irdiW'S illK' funding agencies that base , i .-sources and rewards on research is published in a list of frequently cited -1 rnals The functioning of international K ms is then shaped by personal values and „ , -i stems that influence the importance r   i, i members place on conforming to . ,    mis associated with producing such ,[ v , il uns. Those who manage or collaborate in i hi iiiniial teams must recognize that these ( j,i     create very different ways of ,. r 111. Purposes for being involved in a ,r hough, are not personality traits. One o-m be involved simultaneously in multiple pr_ j-, that have quite different mixes of Ili ii poses.
Ri iouiic'implications
- ,1- hi nalive literature on international „ Mim. Live research tends to be abstract md i i . nslic. while the descriptive literature Sends to be selective in focus. Figure 19.1 hi.ii pi- to remedy both problems by drawing attention ; to the elements in setting up a research : project. Most of these elements
	Social influence processes
	Collaborators
•	Commentators
•	'Close' audience
	'Genera!' audiences
*	Research participants
	('subjects')
Research design ideals: Iheory
• Literature knowledge
• Theory orientation
• Development/critiquing skills
• Writing ability
Data analysis
• Personal understanding
• Assistants
• Support resources
Figure 19.1   Elements in a research project.
are present in any project, but are easily overlooked when researchers are so intent on a normative model of how to conduct good cross cultural research that they lose track of the resources that are needed to do any kind of research. Resources designated in Figure 19.1 as 'social' are those typically not contracted in exchange for money, but are based on networks, understandings, and relationships. Contracted resources are those that typically require funds. Theory resources and data analysis resources represent the body of knowledge and practice that the group will need to access. Organizing skills are those needed to coordinate a project group and link it to outside parties in its academic disci-pline(s). Collaborators from different nations typically contribute to different subsets of these elements.
The process of setting up a group oscillates between finding collaborators, negotiating purposes, and assembling resources. If these elements seem to be in sequence, such that one is needed before the other, this is not intended. They are set out randomly to indicate that in crystallizing a research project, these tend to
	Research design ideals:
	pragmatics
*	Participant access
•	Funding
*	Computing resources
•	Assistants
*	Travel support
*	Email knowledge
*	Data collection
Organizing skills
• Promoting collaboration
• Gatekeeper roles
• Conference organizing
Plus
• Administrative support
• Personal and distant communication skills
• Personal and collective persistence
340
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
CROSS-CULTURAL COMPARATIVE STUDIES
341
occur chaotically. At some point, through the deliberate, proactive efforts of an individual, a few collaborators, or occasionally a larger group, the elements of a research project come to be formulated with sufficient clarity that the project can be announced. Other collaborators are then recruited and focused work can begin. Whichever elements remain indeterminate are either worked out or remain to haunt the project and threaten its completion.
Regardless of research process ideals or the explicit plan for a project, power considerations suggest that collaborators are likely to have credibility and influence to the extent that they can contribute the resources for handling one or more of the elements in Figure 19.1 that others in the group find important. The ability to provide those resources makes it predictable that large multiple nation projects will become centralized. If most project organizing experience, resources, and research skill comes from one or two people, any project is likely to reflect the preferences of a few individuals. Anyone joining an international research team would do well to consider what resources they bring, and any group taking an ideological stance about how the research process should unfold might consider whether its ideals are consistent with the distribution of resource contributions from its various members.
Project coordination
What are the options for coordinating an international collaborative project if one is in a high-power position, or for influencing the group where one's power is more limited or spread more evenly? Theories of social structures help identify these options (Peterson, 2001; Peterson and Smith, 2000; Smith and Peterson, 1988). Rules or norms about science, rules and norms negotiated by the group, guidance from an experienced leader, or ongoing collegial adjustments among experienced colleagues, all provide alternatives.
Two normative theories about how to manage international collaboration from Brett et al. (1997) and Morris et al. (1999) rely on universal arguments about the nature of doing good social science to develop a set
of procedures. As normative thf >i i ■ t rely on norms of science to guidi ^, t-behaviour, specifically variants <-i| |jL , (1969) adaptation of Pike's (1960) id.,, 7 [" emics and etics, concerning the use < cultural surveys. As attempts to particular scientific norms, these not directly discuss the issue of wl. i[ j ,j members of a collaborative rest, i i |. n, will accept them. The variant pi  ■ i._ . n Morrisetal.(1999)appliesbestwh< i.i|» .. includes a substantial replication i km ,, , well as including the potential lor ad.j t.i i ,u The Brett et al. (1997) variant a| | li , ]ts, where new theory and new nit p. i i T being designed. If a team genuinUi P1, one of these or some other normal , i i , \ the model can well provide a pi ijnn? to follow that will combine prior   m ,r | the unique cultural intuition of col1: i> . li ls and the input of research panicin i . m j fruitful way.
Considering other aspects of son il si. Lt-ture besides norms provides other i I'ti i .i m , for guiding a collaborating groi p > a » , DrenthandWilpert(1980)explainik | .in. i for crystallizing and transforming local group norms into a system of rules by i i i 1u explicit social contract. Such woil-uii ■ 'lei have the potential to make expl in 11,r" of the issues in Figure 19.1. Easterby-Smilh and Malina (1999) describe the relying heavily on norms based o i h ■ "i r that individual collaborators develi ■ I iim isji experience.
Relying on a strong group leadu \ n .'I' i prove useful. Many cross-cultural projects are known by the name of t u pi •■■n who coordinated them. The GLO]>l' ' - if (House et al. 2004) provides an   1 mr«. of how to coordinate a project h.i--d  » t clear lender and carefully thought i i'l i■■''« hierarchy of roles and responsibilities. H i "p et al. (2005) describe a number of challenges faced by the GLOBE project, and how the senior leader and leadership group idu. them, in part by relying on norms about how to do good social science, rules in lhe rorn> of social contracts, and other social ■■ nii-i1 such as collaborators chosen to fit th- ■ | ■■' i 1 of the project.
^national research without 'rjrttionnaires
' |1S discussion  has  centred  on survey ;arch, principally because that is the UDatit approach to cross-cultural orga-itional research. There are other exten-s international literatures that deal with licit systems of rules, laws, and economic ndicators that do not evoke the concept _,i I ire. These draw from literatures of institutional theory (Kostova and Roth, 2), national business systems (Whitley, 11 ■ 19), and sovereignty (Krasner, 1988). One llcnge in such work is to contrast laws in mi ions where they tend to be followed and nrced, with laws in nations where they are lie expressions of ideals. Another chal-n >e is lo know how to interpret economic i ■, lure those providing information about, wages paid or received, operate in a ,.■11 hat rewards distorting the information repotted. Collaboration with local researchers with a through knowledge of the local business context is likely to be particularly is- ful in l hese cases.
A c.-.tain amount of qualitative efhno-;ji iphic analysis is built into survey-based ii (eets: when the modes of decentring methods described in the preceding section are i1 owed (Morris et al., 1999). Occasionally, distinct; ethnographic projects also appear in the: cross-cultural management literature. To date, these have tended to be conducted iv i.'. ively small groups and represent ii f a few nations (Baba et al., 2004; Fnrh et al., 2004; Maznevski and Chudoba, 21)00). Easlerby-Smith and Malina (1999) I tribe the social aspects of their research process in some depth. One larger and more ' i ielycoordinated international comparative analysis, of psychological contracts relied " a combination of historical analysis, ethnography, and survey research (Rousseau Md Schalk, 2000). The collaborators in that . ".e worked from a core of basic constructs i' a loose analytical framework, but the rc'archers from each nation conducted their analyses in individualized ways. This project ■"■ rges with a larger literature in which a Marcher edits a book that includes chapters
on a particular theme, including contributions from multiple nations.
A challenge for research relying heavily on local concepts is to find a basis for comparison. To the extent that much survey research is introduced or informed by historical and ethnographic analysis, survey studies implicitly provide this sort of comparative framing. If a large number of ethnographic projects conducted separately in multiple nations were to produce the sort of fragmentation that characterized anthropology in the mid-twentieth century, the post hoc coordination provided by the Human Relations Area File (Murdock, 1940) would be useful for establishing an accessible index and facilitating comparisons.
CONCLUSIONS
International collaborative research has a technical side, a methodological and theoretical orientation side, and a social process side. The purpose of this chapter has been to identify issues and literatures that can help to think through each of these. International collaborative studies bring together researchers whose cultural backgrounds and intercultural experiences lead them to make different assumptions about how the group will operate. It also brings together people who have different degrees of skill and different views about technical issues, who are likely to have different perspectives about methodological and theoretical issues, and who bring unique personal research interests, career plans, and resources. A review of international collaborative projects suggests that those which produce influential research seldom met all of the ideals of any normative model about how they should, operate (Peterson, 2001). A clear sequence of moving back and forth between prior theory and new experience is rarely followed, but both inputs are typically included in some way. All collaborators rarely contribute equally to the research design or execution, most typically because many have other interests and other demands on their time apart from the interests of a core group.
Human nature can be helpful in international research collaboration. People
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
are able to use different theoretical and methodological tools at different times for different purposes. We are able to use a map containing representations of roads even when such maps do not recognize bees and houses. We can also refer to a photograph without worrying whether or not map making is incommensurate with photography, and vice versa. Similarly, social scientists are frequently able to see analogies between quite different ways of analysing cultural characteristics and understanding societies. Whereas theories can be incommensurate such that one does not recognize the existence of central elements of another, people collaborating in international teams are likely to be more successful if they seek elements of complementarity. Ideologies and attitudes about how research should be conducted tend to give way to pragmatic considerations of personal necessity and resource limitations about how research can be conducted.
In order to become involved in an international project in a way that is likely to be satisfactory to all concerned,x the considerations identified here suggest that one should look for a project having several characteristics. It should have:
• Members with adequate technical skills
• A constructive way to develop technical agreements
• Basic methodological and theoretical orientations where differences are viewed as complementary rather than incommensurable
• Compatible purposes
• A working consensus about the norms, rules, and roles that will guide the research process
To the extent that these qualities are lacking, as they always will be to some degree, efforts to improve the project group need to attend to all of these, and not just to a preferred subset.
REFERENCES
Adler, N.J. and Boyacigiller, N, (1995) 'Giobal organizational behaviour: going beyond tradition', Journal of International Management, 1: 73-86.
Baba, M., Gluesing, f, Ratner, H. and W (2004) 'The contexts of knowing: natural globally distributed team', Journal ofi Behaviour, 25:547-87. '
Bass, B.M., Cascio, W.F. and O'Connor, 'Magnitude estimations of expression and amount', Journal of Applied'P: 313-20.
Berry J.W. (1969) 'On cross-cultural coirpordbility-
International Journal of Psychology, 4:' Boyacigiller N. and Adler N.J. (1991) The parochial
dinosaur: organizational science in a g
Academy of Management Review, 16: Brett, J.M., Tinsley, C.H., Janssens, M., I
and Lytle, A.L. (1997) 'New approachi
of culture in industrial/organizations psychology7"
in P.C. Earley and M. Erez (eds), A/m-v Pmpect/wj "
on International/Organizational Psyihology, Saq
Francisco: lossey-Bass, pp. 75-129. Brislin, R.W., Lonner, W. and Thoi
(1973) Cross-Cultural Research Me^ods, NewVorlr •
John Wiley and Sons. Campbell, D.T. and Stanley, J.C. (1963) i
and Quasi-Experimental Designs for Re:
IL: Rand-McNaily. Child, J. (1981) 'Culture, contingency and i
the cross-national study of organizatii
in Organizational Behaviour, Vol. 3. G
JAI Press, pp. 303-56. Drenth, P.J.D. and Wilpert, B. (1980) 'Therol
contracts" in cross-cultural research', J
Review of Applied Psychology, 29: 29: Dutton, J.E. and Dukerich, J.M. (1991) 'Ki
on the mirror: image and identity in i
adaptation', Academy of Management
517-54.
Earley, P.C. (2006) 'Leading cultural re future: a matter of paradigms and taste', International Business Studies, 37: 922-^-31
Easterby-Smith, M. and Malina, D. (1999)'( collaborative research: toward reflexivit/, Acdtkifti ' of Management Journal, 42: 76-86.
Erez, M. (1990) 'Toward a model of i industrial and organizational psychology', -n H.C , Triandis, M.D. Dunnette and L.M. Hough book of Industrial and Organizational i (second edn.), Vol. 4. Palo Alto, CA: Coi Psychologists Press, pp. 559-608.
Farh, J.L., Zhong, C.B. and Organ, D.'.V (2004) 'Organizational citizenship behaviour in the Peoole s Republic of China', Organization Sc 241-53.
Gelade, G.A., Dobson, P. and Gilbert, P. (2006) 'Natrona! differences in organizational commitn
CROSS-CULTURAL COMPARATIVE STUDIES
343
- econofiy, product of personality, or consequence of oJiurp'''. Journal of Cross Cultural Psychology, 37: 542 -5&-
Glazer 5. Geehr, f A (2005)'Consistency of implications (/three role stressors across four countries', Journal of Organizational Behaviour, 26: 467-87. flraei G.B;, Hui, C, Wakabayashi, M. and Wang, v:ZM. (1997) 'Cross-cultural research - affiances in organizational research: cross-cultural partnership making in action', in P.C. Earley and M krez'(eds), New Perspectives on International Industiial/Organizationa! Psychology, San Francisco: lossey-Bass, pp. 160-89. Hanges, PJ. and Dickson, M.W. (2004) 'The development and validation of the GLOBE culture and leadership scales', in R.J. House, P.J. Hanges, M Javidan, P.W. Dorfman and V. Gupta (eds) Culture,-Leadership, and Organizations: The GLOBE Study of 62 Societies, Thousand Oaks, CA: Sage, pp. 122-51.
Hanges, P.J Lyon, J. and Dorfman, P. (2005) 'Managing a multinational team: Lessons from project GLOBE', tn 0. Shapiro, M.V. Glinow and J. Cheng (eds), H/Sjnagmg Multinational Teams: Global Perspectives, Vol 18.:Qxford: Elsevier, pp. 337-60,
Hofstede, G. (1996) 'An American in Paris: the influence o[ nationality on organization theories', Organization Studies, 17: 525-37.
Hofstede, G (2001) Culture's consequences: inter-ndlional differences in work related values, (second edn.) Thousand Oaks, CA: Sage.
Hofstede G (2006) 'What did GLOBE really measure? researchers' minds versus respondents' minds', j trnal of International Business Studies, 37: 882-96.
House, R I., Hanges, P.J., Javidan, M„ Dorfman, P.W. andiGupta, V., (eds) (2004) Culture, Leadership, and Orqantzations: The GLOBE Study of 62 Societies, Ihousand Oaks, CA: Sage.
Huntingtont S.P. (1993) 'The clash of civilizations', i-oreign Affairs, 72(3): 22-49.
IDE International Research Group (1981) Industrial democracy in Europe, Oxford: Clarendon Press.
lnglehart,;R. and Baker, W.E. (2000) 'Modernization, tural change, and the persistence of traditional ties', American Sociological Review, 65: 19-51.
InkelesA and Levinson, D.J. (1969) 'National character: the:study of modal personality and sodocultural ■ terns', in G. Lindzey and E. Aronson (eds), the.Handbook of Social Psychology Reading, MA: Ac'dison-Wesley, pp. 418-506.
JavidrftiM„ House, RJ„ Dorfman, P.W., Hanges, P.J. and de tuque, M.S. (2006) 'Conceptualizing and measur-■ng cultures and their consequences: a comparative
review of GIOBE's and Hofstede's approaches', Journal of International Business Studies, 37: 897-914.
Kahneman, 0. (2003) 'A perspective on judgment and choice', American Psychologist, 58: 697-720.
Kirkman, B.L, Lowe, K.B. and 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: 285-320.
Kostova, T. and Roth, K. (2002) 'Adoption of an organizational practice by subsidiaries of multinational corporations: institutional and relational effects', Academy of Management Journal, 45: 215-33.
Krasner, S.D. (1988) 'Sovereignty: an institutional perspective', Comparative Political Studies, 21: 66-94.
Kroeber, A.L. and Kluckhohn, C. (1952) 'Culture: a critical review of concepts and definitions', Papers of the Peabody Museum of American Archaeology and Ethnology, 47 (1), Cambridge, MA: Harvard University.
Lenartowicz, T. and Roth, K. (2001) 'Does subculture within a country matter? A cross-cultural study of motivational domains and business performance in Brazil', Journal of International Business Studies, 32: 305-25.
Leung, K. {2008) 'Methods and measurements in cross-cultural management', in P.B. Smith, M.F. Peterson and D.C. Thomas (eds), Handbook of Cross-Cultural Management Research, Thousand Oaks, CA: Sage, pp. 59-73.
Leung, K„ Bond, M.H., Reimel de Carrasqtiel, S., Munoz, C„ Hernadez, M., Murakami, F., Yamaguchi, S., Bierbrauer, G. and Singelis, T.M. (2002) 'Social axioms: the search for universal dimensions of general beliefs about how the world functions', Journal of Cross-Cultural Psychology, 33: 286-302.
Lincoln, J.R. and Kalleberg, A.L (1990) Culture, control and commitment: a study of work organization and work attitudes in the United States and Japan, Cambridge: Cambridge University.
Markus. H., and Kitayama, S. (1991) 'Culture and self: implications for cognition, emotion and motivation', Psychological Review, 98: 224-53.
Maznevski, M. and Chudoba, K. (2000) 'Bridging space over time: Giobal virtual team dynamics and effectiveness'. Organization Science, 11: 473-92.
Metcalf, L.E., Bird, A., Peterson, M.F., Shankarmahesh, M. and Lituchy, f. (2007) 'Cultural influences in negotiation: a four-country comparative analysis', International Journal of Cross Cultural Management, in press.
344
THE SAGE HANDBOOK OF ORGANIZATIONAL RESEARCH METHODS
Mezias, S.J., Chen, Y.R. and Murphy, P. (1999) 'Toto, I don't think we're in Kansas anymore: somefootnotes to cross-cultural research', Journal of Management Inquiry, 8: 323-33.
Morris, M.W., Leung, K., Ames, D. and Lickel, B. (1999) 'Views from inside and outside: integrating emic and etic insights about culture and justice judgment, Academy of Management Review, 24: 781-96.
MOW (Meaning of Working International Research Team) (1987) The Meaning of Working, London: Academic Press.
Mowday, R.T., Steers, R.M, and Porter, L.W. (1979) 'The measurement of organizational commitment', Journal of Vocational Behaviour, 14: 224-47.
Murdock, George P. (1940) 'The cross-cultural survey', American Sociological Review, 5: 361-70.
Nisbett, R.E., Peng, K., Choi, i. and Norenzayan, A.
(2001) 'Culture and systems of thought: holistic vs. analytic cognition', Psychological Review, 108: 291-310.
Nisbett, R.E. and Willson, T.D. (1977) 'Telling more than we can know: verbal reports on mental processes', Psychological Review, 84: 231-59.
Oyserman, D., Coon, H.M and Kemmelmeier, M.
(2002) 'Rethinking individualism and collectivism: evaluation of theoretical assumptions and metaanalyses'. Psychological Bulletin, 128: 3-72.
Parsons, T. and Shils, E.A. (1951) Toward a General Theory of Action, Cambridge, MA: Harvard University Press.
Peng, T.K., Peterson, M.F. and Shyi, Y.P. (1991) 'Quantitative issues in cross-cultural management research: trends and equivalence issues', Journal of Organizational Behaviour, 12: 87-108.
Peng, T.K, and Peterson, M.F. (2008) 'Nation, demographic, and attitudinal boundary conditions on leader social rewards and punishments in local governments', Journal of Organizational Behaviour, 29:95-1 17.
Peterson, M.F. (2001) 'International collaboration in organizational behaviour research'. Journal of Organizational Behaviour, 22: 59-81.
Peterson, M.F. and Castro, S.L. (2006) 'Measurement metrics at aggregate levels of analysis: implications for organizational culture research and the GLOBE project', Leadership Quarterly, 17: 506-21.
Peterson, M.F., Maiya, K. and Herreid, C. (1993) 'Adapting Japanese PM leadership field research for use in Western organizations', Applied Psychology: An International Review, 43: 49-74.
Peterson, M.F. and Pike, K.L. (2002) 'Ernies and etics for organizational studies: a lesson in contrast from linguistics', International Journal of Cross Cultural Management, 2: 5-19.
Peterson, M.F. and Smith, P.B. (2000)- ' of meaning, organizations, and culture' rn ' sense of organizational events', in Ashkjnas N.M., Wilderom, C.P.M. and Peterson, M F (/ooni Handbook of Organizational Culture and Clirns^ Thousand Oaks, CA: Sage, pp.101—16.
Peterson, M.F., Smith, P.B. and 44 coauthors poos) 'Role conflict, ambiguity, and overload a 21 nation study', Academy of Management Journal 38' 429-52.
Peterson, M.F. and Smith, P.B. (2008) 'Social si.n.«tnK and processes in cross-cultural managenn P.B. Smith, M.F. Peterson and D.C. Thomas (fris) Handbook of Cross Cultural Management Research Thousand Oaks, CA: Sage, in press.
Peterson, M.F. and Wood, R. (2008) 'Cognitve structures and processes in cross-cultural managenent' in P.B. Smith, M.F. Peterson and D.C. Thomas (eds) Handbook of Cross Cultural Management Research Thousand Oaks, CA: Sage, in press.
Pi ke, K. L {1960) Language in Relation to a Unifiei of the Structure of Human Behaviour (third edn j, Glendale, CA: Summer Institute of Linguistics
Pike, K.L. (1982) Linguistic Concepts: An Introduction to Taginemics, Lincoln, Nebraska: University Nebraska.
Polanyi, M. (1958/1962) Personal Knowledge: Towards a Post-Critical Philosophy, Chicago: Unrv< si, £ Chicago.
Raudenbush, S.W. and Bryk, A.S. (2002) Hierarchical Linear Models: Applications and Data ^ -Methods (second edn.), Thousand Oaks, CA: Sage.
Roberts, K.H. (1970) 'On looking at an ele phant: an evaluation of cross-cultural research related to organizations', Psychological Bulletin, 74: 327-50.
Rokeach M. (1968) 'A theory of organization and change within value-attitude systems', Journal of Social Issues, 24: 13-33.
Rokeach, M. (1973) The Nature of Human Values, New York: Free Press.
Ronen, S. (2006) The New Cultural Geography: A Uets Analysis of Country Clusters, Invited address, 2bth congress of the International Association for Applied Psychology, Athens, July.
Rousseau, D.M. and Schalk, R. (2000) Psychological Contracts in Employment: Cross-National Perspectives, Thousand Oaks, CA: Sage.
Sapir, E. (1921) Language, New York: Harcourt, Brace and Co.
Saville, P. and Willson, e. (1991) 'The reliabili"        " ': validity of normative and ipsative approaches in the measurement of personality', Journal of Occupational Psychology, 64: 219-38.
CROSS-CULTURAL COMPARATIVE STUDIES
345
'ita/artz.   S'H'   (1994)   'Beyond individualism-* ,clj|[ectivism: new cultural dimensions of values', 3. Kim, H.C. Triandis, C. Kagitcibasi, S.C. Choi G: Yoon (eds), Individualism and collectivism: "rfieory. Method and Applications, Newbury Park, \h Sage. pp. 85-119.
S.H. and Ros, M. (1995) 'Values in the West: a theoretical and empirical challenge to the ■'individualism-collectivism cultural dimension', World psychology, 1: 91-122. jchwartz. S.H. and Sagiv, L. (1995) 'Identifying culture-icifics in the content and structure of values', journal of Cross-Cultural Psychology, 26: 92-115. Jcott, WR. (1995) Institutions and Organizations,
jusand Oaks, CA: Sage. >mith, pj. (2004) 'Acquiescent response bias as an aspect of cultural communication style', Journal of issCulturalPsychology, 35: 50-61. >mtth, P.B. (2008) 'Indigenous aspects of management jes', in P.B. Smith, M.F. Peterson and D.C. Thomas (eds),' Handbook of Cross-Cultural Management ■Research, Thousand Oaks, CA: Sage, pp. 319-32. iitlith, PB., Dugan, S. and Trompenaars, F, (1996) .' 'National culture and the values of organizational .employees: a dimensional analysis across 43 . 'nations', Journal of Cross-Cultural Psychology, 27: 231-64.
Jjraith, P.B. and Peterson, M.F. (1988) Leadership, Organizations and Culture, London: Sage. Smith P.B., Peterson M.F. and Schwartz S.H. with 44 coauthors (2002) 'Cultural values, sources
of guidance and their relevance to managerial behaviour: a 47 nation study', Journal of Cross Cultural Psychology, 33: 188-208.
Tannenbaum, A.S., Kavcic, B., Rosner, M., Vianello, M. and Wieser, G. (1974) Hierarchyin Organizations, San Francisco: Jossey-Bass.
Teagarden, M.B., Von Glinow, M.A., Bowen, D.E., Frayne, CA., Nason, S., Huo, Y.P., Milliman, J., Arias, M.E., Butler. M.C., Geringer, J.M., Kim, N.M., Scullion, H„ Lowe, K.B. and Drost, E.A. (1995) 'Toward a theory of comparative management research: an idiographic case study of the best international human resources management project', Academy of Management Journal, 38: 1261-87.
Trompenaars, F. and Hampden-Turner, C. (1998) Riding the Waves of Culture: Understanding Diversity in Global Business (second edn.), New York: McGraw Hill.
Usunier, J.C. (1998) International and Cross-Cultural
Management Research, London: Sage. Van de Vijver, F. and Leung, K. (1997) Methods and
Data Analysis for Cross-Cultural Research, Thousand
Oaks, CA: Sage. Weber, M. (1951} The Religion of China, NewYork: Free
Press.
Whitley, R. (1999) Divergent Capitalisms: The Social Structuring and Change of Business Systems, Oxford: Oxford University Press,
Whorf, B, (1956) Language, Thought and Reality, Cambridge, MA: MIT Press,