Diversity and difference research: A reflection on categories and categorization

Introduction*

The categories used in this research tend to be pre-established categories: gender, race/ethnicity, class, religion, sexual orientation, and disability (Litvin, 1997; Lorbiecki and Jack, 2000), age, values, and education, among other variables (Egan and Bendick, 2003; Loden and Rosener, 1991; Milliken and Martins, 1996), nationality and culture (Zander et al., 2010) with the goal of explaining management of diversity and difference and the behaviors of others. Pre-established social categories accordingly form the core of diversity and difference research, and categories are the focus of our article.

Critical diversity scholars address pre-established social categories in their critique of mainstream diversity research claiming that these categories both reflect unequal power and contribute to reproducing it (see Zanoni et al., 2010, for a more extensive review). Pre-established categories can lead to static accounts of diversity in organizations, which ignore the dynamic nature of power and inequality relations (Tatli and Özbilgin, 2012). Consequently, approaching diversity issues through pre-established categories may lead to a risk to ‘essentialize difference in framing of workforce diversity and produce flawed empirical, theoretical and policy insights’ (ibid: 181). Critical scholars claim diversity and difference researchers tend to overlook the issue of power and power structures in their analyses (Primecz et al., 2016; Tatli and Özbilgin, 2012; Zanoni et al., 2010). Power is the focus of intersectional theorists in their analysis of intersecting categories of difference and diversity (e.g. Essers and Benschop, 2009; Adib and Guerrier, 2003). Therefore, we let intersectional theory and research inspire our reflections on categories in diversity and difference research.

The purpose of this paper is to contemplate the power aspects of categories and the possible implications of using pre-established categories in diversity and difference research. We begin with a short overview of intersectionality to set the stage for our further reflections. Thereafter, we continue with a brief reflection on why humans categorize. We continue to ponder over the ways in which power is inherent in the categories used in theories aimed at fighting inequality and discrimination and in managing increasing diversity in organizations. We discuss the use of intersectionality in existing research on diversity in organizations to inspire our discussion. We conclude by reflecting on possible consequences of taking categories for granted in research on discrimination and inequality.

Intersectionality: A means to introduce power in diversity and difference analysis

Intersectional theorists explore power relationships through a lens of mutual constructions of sociocultural categories (Collins and Bilge, 2016; Lykke, 2010). The legal scholar Kimberlé Williams Crenshaw coined the term intersectional as a way of analyzing ‘the various ways in which race and gender interact to shape the multiple dimensions of Black women’s employment experiences’ (Crenshaw, 1991: 1244). Intersectional research ‘centers the experiences of subjects whose voices have been ignored’ (Nash, 2008: 3).

With a focus on structural inequalities and discrimination, intersectional theorists aim to explain and understand how the interdependence and mutual constitution of social categories on an individual level reflect systems of power, oppression, and privilege on a socio-structural level and an organizational level (Acker, 2006; Bowleg, 2012; Rodriguez et al., 2016). Marfelt (2016) identifies two main intersectional research streams. Loyal to the origins of intersectionality (Crenshaw, 1991), scholars adhering to the first stream study various forms of oppression (e.g., Collins and Bilge, 2016; Holvino, 2010; May, 2015). Recently, however, scholars have begun to see intersectionality as extending beyond oppression (e.g., Diedrich et al., 2011; Kelly and Lubitow, 2014), using it more as an analytical tool for understanding and explaining the interacting and simultaneous effects of different factors (Hancock, 2007; Lykke, 2010; Zander et al., 2010). Nash (2008) suggests broadening the explanatory power of intersectional theory to also focus on ways in which privilege and oppression are co-constituted on the subject level.

One of the main attributes of intersectional theory is its focus on the analysis of social categories on multiple axes rather than dealing with single categories. Many diversity and difference scholars determined to depart from single-dimension and essentializing category analysis turn to intersectional theory, as it promises a more complex view of the multi-dimensional constructions of identities (e.g., Essers and Benschop, 2009; Essers et al., 2010; Holvino, 2010). Not only is the multiplicity of categories emphasized in intersectional analysis, but also their simultaneous and intertwined nature (e.g. Adib and Guerrier, 2003). Intersectionality adds complexity and more fine-grained explanations and understandings to diversity and difference analyses. Intersectional diversity and difference researchers focus also on societal discriminative processes in addition to social categories in their analysis of inequality in organizations (e.g., Carrim and Nkomo, 2016; Johansson and Śliwa, 2016). We bring with us these notions from intersectionality into our further reflections. First, though, we will contemplate on why humans categorize.

The human need to categorize

Categorization is at the heart of human activities and sense making. Categories are mechanisms for organizing. ‘A “classification system” is a set of boxes (metaphorical or real) into which things can be put to then do some kind of work – bureaucratic or knowledge production’ (Bowker and Star, 2000: 10). Categorization sorts reality into comprehensible categories built on a worldview whereby the attributes of an object tend to correlate (Rosch, 1999; Rosch et al., 1976). A classification system is chosen for a specific purpose (Bowker and Star, 2000) and should ideally demonstrate: 1) consistent, unique classificatory principles, 2) mutually exclusive categories, and 3) a complete system that provides total coverage of the world it describes.

Categorization helps us by simplifying and guiding our actions and behaviors in our everyday lives, routinizing them, providing structure, bringing order to a complex world (Banton, 2011; Vergne and Wry, 2014). It offers a coping strategy and tool for structuring experiences and remaining in control (Jacob, 2004). Categorization is a way of sorting perceptions and actions (Bowker and Star, 2000) – a human activity subject to habitual and organized action. The first time one encounters an object, it may be difficult to categorize it; in time, the process becomes routine, unconscious, embedded in organizational routines, and the categories become taken for granted (Berger and Luckmann, 1966). By studying the process of categorization, one can explain why certain attributes are considered more significant than other attributes.

The functions of categories in organizational diversity and difference research

In organizations categories can provide individuals with guidelines for behavior in certain situations. If employees are sorted into categories according to perceived social identities, the categories serve to facilitate interaction and common understanding within and across groups. Categories aim to make the vast diversity of individual entities we encounter in daily life manageable (Bodenhausen et al., 2012), satisfying a basic human need for cognitive parsimony (Hogg and Abrams, 1988).

Diversity viewed as business logic highlights how diversity can be an organizational resource (Risberg and Søderberg, 2008; Williams and O’Reilly, 1998), enhancing creativity and innovation, for example, by presenting a variety of viewpoints (Roberge and van Dick, 2010; Stahl et al., 2010). Categories are here used to identify the parts of a person’s identity that provide a resource for the organization.

In the light of internationalization and globalization of business, diversity and difference research – driven by business logic – includes such categories as culture and nationality (Nishii and Özbilgin, 2007). Many cross-cultural researchers consider a multi-cultural environment as a risk for success-driven organizations, and potential problems are associated with miscommunication and conflict arising from cultural differences (e.g., Hofstede, 2001; 1980). This ‘problem-focused view’ of cultural diversity (see Stevens et al., 2008) leads researchers to culture- and nationality-based categorization, in their attempt to solve problems stemming from cultural or national differences. Categorization can therefore be seen as a tool to ‘smooth out’ cross-cultural business encounters.

Diversity seen as a moral and social-justice issue builds on the assumption that individuals of certain categories are disadvantaged in comparison to others. Social-justice logic has its origin in anti-discrimination legislation aimed at protecting a group of people who are discriminated against in workplaces or in society at large (see European Union, 2000; ILO, 1958, and various national legislations). Categories are used to identify discriminated minority groups, to strengthen their position in organizations and society, and to combat the discrimination they face – often the same categories found in national anti-discrimination laws, EU directives, and the UN Declaration for Human Rights.

A growing body of critical diversity management studies sets out to study – hidden or not – dimensions of power embedded in diversity management practices (Marfelt and Muhr, 2016). Many critical diversity scholars point out that if hierarchy, privilege, and discrimination, among other factors, are left unattended, systemic oppression will continue (e.g., Oswick and Noon, 2014; Tomlinson and Schwabenland, 2010). Categories are here used to criticize perpetuation of inequality regimes.

In our focal theories, categorization serves to fulfill theoretical and analytical objectives. In business-logic diversity and difference theories categorization serves a more normative objective: how to help organizations manage situations and events involving people representing differing social categories. Social-justice and critical driven diversity research uses categories as a tool to identify groups targeted for systemic inequality or discrimination, pointing to power aspects within organizations.

In business-logic diversity and difference theories, it may seem that categories are relatively neutral concepts, the purpose of which is to facilitate the daily lives of humans in the workplace and in society. However, following the insights from intersectional research and social-justice driven diversity and difference research, it becomes clear that is not the case. Therefore, we continue with a critical discussion of political and power aspects embedded in categories.

Implicit assumptions embedded in categories

Categorization can be a process of power demonstration. Bowker and Star (2000) point to epistemological, political, and ethical aspects in processes of building classification systems. Categories are subjective, as the location of the individual perceiving them influences the perception of opposition and dissimilarity (Berger and Luckmann, 1966). Ahonen and Tienari (2015) argue that although diversity management discourse emphasizes the goal of equality and social justice, thereby claiming neutrality of judgment, the differences in focus are selective (see also Swan, 2010). Those determining the categories have the power to define the norm (what is counted as a difference, in relation to what?). They hold the power to decide what attributes of difference should be categorized (see also Marfelt and Muhr, 2016). The significance given to attributes is influenced by what the individual creating the category perceives to be an important distinction. The attributed distinction is strongly influenced by the perception of difference and a belief that the difference is noteworthy (Harrison and Klein, 2007).

To categorize diversity and create diversity knowledge is to ‘identify and enable or set in motion the diverse subject, to make the diverse subject a visible, legitimate, subject’ (Ahonen and Tienari, 2015: 281). Diversity entails elements of construction and agency. The correlational structures of objects are subject to the individual perceiving the attributes which in turn depends on the individual’s location, including cultural background and specific capabilities of perception. Ahonen and Tienari (2015) add underlying political agenda to the list, pointing to underlying norm structures by which an organization’s performance and competitive advantage guide the understanding and classification of diversity (see also Ahmed, 2012). This sustains the business-based view that diversity contributes to organizational performance. Diversity management categories, whether business- or social-justice based, are infused with constructions dependent on the location of the individual categorizer.

If certain characteristics are assumed to correlate with an attribute, giving it significance may lead to dubious or stereotypical conclusions about individuals who are sorted into a certain category. Work attitude may be associated with ethnicity, for example, if a specific ethnic group is perceived as hardworking. If these correlations are considered fixed, naturalized essences of a group, even though subjective and dependent on individual perceptions, they serve as grounds for prejudices and stereotypes.

Power aspects of chosen categories in diversity and difference studies rarely receive attention (for some exceptions, see e.g. Ahonen et al., 2014; Ahonen and Tienari, 2015, Primecz et al., 2016) and there is an inadequate theorizing of power in diversity management research (Zanoni et al., 2010). A reason could be the most commonly used social categories in diversity research – gender, age, ethnicity, sexuality, disability, and religion – are taken for granted, used without deliberations.

Power relations ingrained in category constructions in diversity and difference research

Notions of inferior or superior, privileged or non-privileged, powerful or powerless seem to be ingrained in most social-identity categories used in organizations – if not in society at large. Social categories occur when we differentiate with regard to gender, skin color, race, and class, in reference to social, economic, or power status. Each category covers a certain group of individuals; the category ‘male’ reflects male attributes that differ from female attributes. Yet, the dimension is often collapsed within a certain category. A reading of gender in organization studies reveals that whereas one talks about the dimension of gender, in practice only the category of women is referred to. For example, an extensive body of research investigates the disadvantages female employees experience in male-dominated organizations (for a review, see Broadbridge and Simpson, 2011). In line with the purpose of categorizing to combat discrimination, one category moves to center stage as the group of individuals associated with it is seen to be disadvantaged – a situation to be overcome. The category is therefore legitimized by the conditions its members supposedly experience. Female employees are seen to suffer from obstacles in career advancement and lower pay than men; great importance is therefore accorded the category of women, in order to combat organizational inequalities. But, such single focus on women assumes that women need to change their unprivileged, inferior position, not that the unequal structures need to change. A power relationship is ingrained in the categories of woman and man.

Acker (2006: 444) observed how race ‘refers to socially defined differences based on physical characteristics, culture, and historical domination and oppression, justified by entrenched beliefs’. Such understanding can be traced back to Carl von Linné’s (1735) taxonomies of biologically defined human races. His taxonomy came to serve as a foundation for eugenics (Lindquist, 1996; Lindquist and Olsson, 1991) and as basis for colonial power oppression. Eugenics builds on assumptions that humanity can be divided in more or less distinct races that differ in physical and mental aspects. The differences are assumed to originate from inherited characteristics transferred from generation to generation and on the basis of these differences a value hierarchy is constructed (Lundmark, 2007). Linné’s taxonomy divides Homo sapiens into four categories based on geographical location and skin color: Africanus Niger, Americanus Rubescens, Asiaticus Fuscus, and Europeaus Albus (Broberg, 1975). He described, for example, Africans as phlegmatic, black, slow, relaxed, and negligent; and Europeans as pale, muscular, fast, smart, and inventive. Inequalities and power structures were (re)produced in the construction of race categories as they point to inferiority and superiority based on human features (e.g. skull form) and correlated with intellect and behavior. Historical uses of human categories have clear power structures built into the categories and the categorization processes as it attributes superiority and greater intelligence to some races over others. The South African apartheid, by which people were classified in order to be separated, serves as a prime example of race oppression at work. In a more modern context, Vesterberg’s (2013) study of immigrants points to how members of the category ‘Somali’ are constructed as being as inferior by being difficult to integrate into the Swedish society and labor market. Using race categories uncritically increases the risk of perpetuating power, inequality, and discrimination.

Structured representations of national cultures (e.g. House et al., 2004; Hofstede, 2001) are infused with binary oppositions of Western and non-Western cultures. Said (1978) describes how Western descriptions of the East are tied to Western imperialist interests, whereby descriptions of the East gain meaning through their reference to the West. Typical binary oppositions such as good-evil, honest-dishonest, civilized-primitive, controlled-aggressive accentuate the difference and hierarchical opposition (ibid.). Positive terms are used to describe the norm (Western man) and negative terms to describe the Other. European cultures are seen as more positive than African cultures and North American as more positive than Asian cultures. Relationships and comparisons are made with Western terms of reference (Zanoni et al., 2010), and dualisms or binary oppositions are introduced to make cultural relations manageable (Lorbiecki and Jack, 2000).

Discourses on cultural diversity are imbued with power relationships, as categories are constructed in relation to Western norms, yet assumed to be neutral. Ahonen et al. (2014) argue that diversity and its categories must be understood in relation to those who conceptualize it, which reflects the contextual conditions in which diversity and its categories are constituted. The classification, calculation, inclusion, and exclusion of categories of cultural diversity are rooted more in context than in the taxonomical tool. What Ahonen and Tienari (2015) call ethico-politics is constantly involved in categorization, because context is influenced through value statements and political-ethical statements made in that context (ibid; Willmott, 2014).

These examples show how categories are constructed with reference to a value system wherein the existence of a category is given sense through norms (Jenkins, 2000) and notions of normalcy. The category ‘Somali immigrants’ makes sense in relation to their labor market status; the category ‘female employees’ is given significance with respect to their relative positioning in the organization. Dominating norms of equality and a work ethic consequently influence the framing of categories, and minority groups are distinguished in relation to that norm. Categorization serves to frame these groups. As Zanoni and Janssens (2004) demonstrate, human resource managers understand diversity under the umbrella of grand discourses such as compliance and work ethic. These discourses influence the process of categorization and reaffirm existing management practices and inequalities (see also Zanoni and Janssens, 2015). Power relations and notions of deviation from established norms accompany the categorization of people in organizations.

Intersectional diversity and difference research

If intersectional theory explores power in the intersection of social categories, is intersectionality the answer to the above critique? Could a more conscious categorization process be a way to focus on unequal power relations and norms reflecting the interests of the privileged? Below we bring forward what intersectionality has brought to diversity and difference research in terms of categories and power issues.

Much intersectional diversity and difference research is conceptual and theoretical, suggesting how intersectionality could contribute to the understanding of diversity and difference in organizations. Many, for example, argue for multiple rather than single category research (Özbilgin et al., 2011) or to extend the classic intersecting categories (Holvino, 2010; Metcalf and Woodhams, 2012; Prasad, 2012; Strolovitch, 2012). The number of empirical diversity and difference studies using intersectional analysis is growing. Contributions to earlier diversity and difference research demonstrate that these studies bring in macro contextual factors in the analysis as well as study processes of inequality. These studies also point to the fluidity of intersecting axes and point to power processes.

Carrim and Nkomo (2016) in their study of South African Indian female managers emphasize how the women’s managerial identity is formed by personal and social identities (categories) and institutionalized systems and processes (apartheid and cultural norms). Also, Johansson and Śliwa (2016) point out that the intersection of language and social identity categories are important to understand processes of social and organizational differentiation. Atewologun et al.’s (2016) study shows how intersecting categories are fluid as individuals may have advantaged or disadvantaged social identities at the same time. The advantaged positions shift according to intersecting categories. For example, a male non-white senior manager’s position may be advantaged in relation to a subordinate but disadvantages in relation to a white senior manager. Power positions are thus not stable but fluid according to axing social categories. Adding a category beyond the classic axes of social categories – hierarchical lines – Boogaard and Roggeband (2010) use intersectionality to show how organizational structures and individual agency can construct inequality in organizations. In a study of Muslim immigrant entrepreneurial business women, using intersectionality as an analytical tool, Essers et al. (2010) could point to tensions encountered by respondents because of the intertwining of gender and ethnicity.

Intersectionality makes it possible to untangle and change the differential impact of everyday practices in organizations, and identify and link internal organizational processes and structures of oppression and power with external societal processes and structures (Holvino, 2010). It offers ‘simultaneous processes of identity, institutional and social practice, brings more complete and accurate analyses and better organization and policy change applications’ (ibid: 266). Thus, intersectionality used in diversity and difference studies offers more thorough analyses of inequality and oppression, potentially leading to better-grounded organizational and policy changes.

What intersectional analyses bring to diversity and difference studies are thus more dynamic views of categories, adding new intersections and linking the organizational context to the societal context – though not all studies do all these things. Yet, most of the studies build on pre-defined categories.

Discussion

Our reflections point to how social identity categories are accompanied by power and relationships of inequality and have universalist and essentialist tendencies (Tatli and Özbilgin, 2012). Despite the aim of intersectional theorists to move beyond attention to single categories and to proclaim the simultaneity and interdependence of categories, the core argument is built on stable and durable categories (McCall, 2005). Thus, using multiple categories does not necessarily eliminate the risk of perpetuating power relations ingrained in the categories. Simply categorizing someone as a woman or non-white person perpetuates inequality regimes. Categories of gender and race affect assumptions about skills, responsibility, behaviors, and even appropriate salaries (Acker, 2006). The woman category is accompanied by expectations of how the person will behave and perform on the job, and how she should be remunerated. Labeling someone a refugee migrant from a certain region includes assumptions about literacy, education, and competences. Prasad (2012) claims that simply by using fixed categories, scholars risk reifying and perhaps unintentionally legitimating public assumptions about pre-defined categories, especially binary categories. Categories also shape the behaviors of those associating certain assumptions with certain categories. As Acker (2006: 451) notes ‘supervisors probably shape their behavior with subordinates in terms of race and gender in many other work situations, influencing in subtle ways the existing patters of inequality’.

So, many critical diversity management scholars argued there is little value in using pre-established categories in diversity studies (Marfelt and Muhr, 2016). Instead Tatli and Özbilgin (2012) propose the use of emergent categories that are locally and empirically identified and defined according to their role in generating power, privilege, advantage, disadvantage, discrimination, and inequality. This would generate emic knowledge about diversity, acknowledging the dynamics of power, advantage, and privilege alongside disadvantage, inequality, and discrimination, they claim. The starting point for emic categories is the analysis of power relations, including the historical, institutional, and socio-economic settings.

Primecz et al. (2016), on the other hand, call for caution with emergent categories as the categorization process is not unbiased. They point to how emerging categories are unpolitical, not questioning tensions between advantaged and disadvantaged positions. In addition, new categories of diversity and difference are most likely linked to those holding power positions, just as Marfelt and Muhr (2016) conclude.

Do we have a double-bind dilemma here? If one does not use pre-established categories in diversity, difference and intersectionality research, one risks overlooking structural discrimination; yet, by uncritically adopting pre-produced categories, one risks the perpetuation of racist or sexist notions and inequality regimes (Zanoni and Janssens, 2004). There is probably no clear answer but we could conclude that diversity and difference research, using single or multiple categories, could benefit from an approach that attends to the temporal and geographical contextuality of relationships of power, privilege, inequality, and disadvantage (Metcalfe and Woodhams, 2012) inspired by intersectionality. To this approach, one should also add a caution regarding what power and inequality aspects come with the pre-established or emergent categories.

Conclusion

Our aim with this paper has been to reflect critically on the use of categories in traditional and intersectional diversity and difference research. Although categorizing is inevitable in handling complexity, and although it helps to fight inequality and discrimination, it can also contribute to continuing patterns of inequality and discrimination.

Categories should not be taken for granted and should be used with caution. One way forward could be to study the construction of categories – what Eriksson-Zetterquist and Styhre (2007) call the organizing of diversity practices – as the construction process entails power and power positions (Tatli and Özbilgin, 2012). Understanding how categories are used, defined, or constructed can help in understanding the power structures and power relations in an organization in relation to other processes and practices, including recruitment, advancement, salary, remuneration, promotion, and training. Such analysis could help us understand how categorization and categories – accompanied by aspects of power and inequality through assumptions – affect the lives of people working in organizations. A question remains though – to paraphrase Linstead and Brewis (2004) – how can we write about diversity and difference and acknowledge the importance of diversity and difference without reproducing power relations?