Meanings and Motives before Measures : The ‘ What ’ and ‘ Why ’ of Diversity within the Mossos d ’ Esquadra and the Politie Utrecht

This article proposes the definition of and motivation for diversity (policy) as an important research topic that should be studied before focusing on diversity policy measures. As such, it strives to demonstrate the academic potential of an analytical framework that outlines fundamental choices made in these respects. What types of diversity do organizations focus on? And what do they want to achieve with (increased) diversity? In this article the discourses underlying the diversity policies in two regional European police forces—the Mossos d’Esquadra and the Politie Utrecht—are analyzed. The main observation is that the results are surprisingly similar in spite of contextual factors that may lead observers to expect otherwise: they both focus on gender and migrant background, identifying these types of diversity as collective in nature, while striving for equal opportunities for individuals despite these collective differences. This article also explores possibilities for further theory building by formulating possible explanations for the similarities and differences which have been identified, suggesting a possible hierarchy in diversity within European organizations, and describing how the motivation for diversity might influence the effectiveness of diversity policies.


Introduction
Awareness has grown in both the professional and the academic worlds of the importance for organizations of adapting to their diverse surroundings.Public and private organizations alike have adopted specific policies in order to facilitate diversity within.The main goal is often the diversification of the employee base.In Europe, this tendency is reinforced by the European Union's adoption of diversity as a central political priority (Shaw, 2005).
However, there seems to be considerable theoretical ambiguity and controversy with respect to the concept of diversity and diversity policy.Ambiguity results from a lack of articulation of or reflection on the actual meaning of diversity.Controversy arises from the con-siderable disagreement about how to interpret the assumptions, intentions and methods of diversity policies.This is not surprising as the concept 'diversity' is a social construction.To avoid misunderstandings, scholars and applied researchers in the field of diversity studies would do well to make their definitions of diversity explicit to the reader and to reflect beforehand on the implications of such definitions on the scope of their conclusions, observations or results.
One of the observations of Yang and Konrad (2011) is that many, if not most, analyses of diversity policy in the field of organizational studies focus on causal relationships.For example, Roberson and Park (2007) explore how diversity reputation and minority leadership impact firms' performance; Stewart and Johnson (2009) discuss how the quality of leader-member interactions moderates the relationship between gender diversity and team performance; Chi, Huang and Lin (2009) show how HRM is an intermediate factor in the relationship between tenure diversity and innovation; and so forth.However, the definition of diversity and diversity policy is an interesting research topic by and of itself (Van Ewijk, 2011).The way in which diversity issues are framed is critical to policies and/or organizational approaches to the topic, determining, for example, how public budgets are distributed to schools, or which development practices receive company funding (Smith, 1995).Similarly, it might affect a person's social status, financial situation, juridical rights, and so on.
Still, academic research into the definition of diversity is scarce and some notable gaps remain.First, the few studies there are employ almost exclusively quantitative methods (Carrell & Mann, 1995;Point & Singh, 2003), relying on the face validity of key terms such as 'ethnic' or 'black' or 'age', while differences in the practical interpretation of those key terms by respondents are not at all improbable and may considerably affect the reliability and validity of the research's results (Van Ewijk, 2011).This calls for more qualitative studies.Secondly, most studies on diversity and diversity policy were undertaken in the United States, Canada, or Australia, countries with a very specific experience of diversity because of their cultural, social, and historical context (Mor Barak, 2005).When contexts are different, it is important to refrain from universalizing ideas and knowledge, and instead to focus on the context at hand in order to gain more insight into the matter (Carens, 2000;Modood, Triandafyllidou, & Zapata, 2006).This calls for more European studies.Third, all studies carried out on the definition of diversity and diversity policy in Europe have been based on analyses of private organizations (Point & Singh, 2003;Liff, 1999;Wrench, 2007;Van Ewijk, 2012).Public organizations are also important to study, maybe even more than private ones, because of their enormous scope; all citizens are their 'potential customers'.This calls for more empirical studies of public organizations.
In this regard, the police are particularly intriguing because of their almost paradoxical relationship with diversity.On the one hand, the potential impact on society of diversity within the police is large (the police have a highly symbolic role and interact with a broad range of the population in many spheres of life) and the potential for vertical mobility is high (in theory, every police officer can become chief).On the other hand, empirical data on police forces in Europe shows that the level of diversity is low among new recruits and further diminishes as the ranks get higher (Van Ewijk, 2012).
Although there are a considerable number of academic studies on diversity (policies) within police forces, there are also some considerable gaps in the research.There are few comprehensive analyses which take into account recruitment, promotion and retention, and few comparative studies.One barrier to the latter is that key definitions often vary between countries and are not always easily identified or controlled (Mawby, 1999), making comparisons problematic.This seems to be especially the case concerning diversity within police forces.Diversity is always implicitly defined, simply by focusing exclusively on a certain group or groups.These groups are distinguished by gender (Dick & Cassell, 2002;Dudek, 2007;Metz & Kulik, 2008;Zhao, He & Lovrich, 2006), sexual orientation (Blackbourne, 2006;Miller, Forest, & Jurik, 2003), ethnicity, race or skin colour (Cashmore, 2002;Jaeger & Vitalis, 2005;Jain & Agocs, 2008;Johnston, 2006;Holdaway, 1998;Holdaway & O'Neill, 2004;Phillips, 2005;Walker & O´Connor, 1999), and religion (Armitage, 2006).Making this variance explicit would significantly lower this barrier.This calls for studies on the definition of diversity in the context of diversity policies in police forces.
Therefore, this article offers a comparative case study of two European police forces: the Mossos d'Esquadra (the police force of the autonomous region of Catalonia, Spain) and the Politie Utrecht (the police force in the province of Utrecht, Netherlands).These police forces were analyzed using an analytical framework that identifies the fundamental choices that can be made as regards the definition of diversity (the 'what', i.e. the ontological dimension) and the arguments for diversity within organizations (the 'why', i.e. the deontological dimension).Both refer to what Tatli (2011) calls 'diversity discourse'.This analytical framework, which elaborates on what is meant by diversity in this article (see Section 2), has been formulated elsewhere (Van Ewijk, 2011), but has not been applied in empirical research so far.
The goal of this article is twofold.First, it aims to show the academic potential of the analytical framework by applying it to two empirical contexts and outlining the possibilities for academic theory building that are generated by this comparative analysis.In the process, it also aims to offer a deeper understanding of the way in which diversity is conceived and operationalized in the policies of these two police forces.
The article consists of five sections.Following the introduction, the criteria used for case-selection and for the cases themselves is introduced, and the methods used for data collection and analysis are outlined in the second section.The third section analyzes what definition of diversity is used in each context, while the fourth section focuses on all the reasons why diversity within the organization is deemed important.The fifth and final section demonstrates the possibilities for further theory building generated by the empirical application of the analytical framework in these particular contexts.

Methodology
It was decided that the research should take the form of a case study as data on diversity within European police forces is limited (Van Ewijk, 2012).The general advantage of case studies is that they allow for conceptual refinements with a higher level of validity over a smaller number of cases (Eisenhardt, 1989;Yin, 2003) and the possibility of identifying new variables and hypotheses during the course of field-work.A second case study was chosen to reinforce the academic potential of the analytical framework, as the comparison of two cases demonstrates how the framework can be applied equally well in considerably different contexts, and enhances the possibility of gaining new theoretical insights.This section describes the criteria used to select the two cases, and the main strategies employed for data collection and data analysis.

Case Selection
The premise of the contextual approach is that the characteristics of the context are significantly intertwined with what happens within an organization.This coincides with Yanow's (1996) approach of examining the relationship between the meanings given within the organization and those given within the context of this organization (George & Bennet, 2005).The epistemological base of this type of study is the constructionist perspective: meanings are negotiated as a product of history and of social structure (Sapsford & Jupp, 1996).In line with this approach, the contexts of both cases had to be similar in areas that are likely to provide a strong incentive for adapting the organization to diversity, yet different in areas likely to coincide with variation in fundamental choices concerning the definition of and the motivation for diversity within the organization.These similarities and differences increase the likelihood of reaching a deeper understanding of how diversity can be conceived within police forces, while at the same time demonstrating how the empirical application of the analytical framework leads to new possibilities for theory building.
The Mossos d'Esquadra are the police of the Catalan autonomous community in Spain.They first appeared in 1719, and have since then existed in various forms with different mandates.In 1983, they were established as a modern democratic police force.Initially, their tasks were limited and it was not until 1994 that the force became an integral police force for the whole of Catalonia.This expansion was completed by November 2008, by which time the number of police officers had grown to 14,143, up from 5,000 in 1999.
The Politie Utrecht are the police in the province of Utrecht in the Netherlands.From 1945 to 1993, the Dutch police service consisted of numerous municipal police forces, in addition to a national police force.In 1993, the Dutch police service was divided into 25 regional forces and the National Police Services Agency (KLPD), which has various specialist and support departments.The Politie Utrecht is a medium sized police force, which counted 3,852 employees in 2009.This was all still the case when this study was conducted.Afterwards, in 2013, the Dutch police entered a major reorganization process, transforming it from 23 regional forces into a single force with 10 regional units.This reorganization process remains unfinished as of June 2014.Figure 1 displays the areas of jurisdiction of both police forces.
The first similarity that might provide a strong impetus for the development of a diverse police force is the demographical diversity of their area of jurisdiction, which is high in both cases.Previous studies confirm that immigration is one of the main carriers of diversity, and this diversity tends to multiply and consolidate itself mostly in urban areas (Castles, 2002;Penninx et al., 2004;Ireland, 2004).In this sense, the areas of jurisdiction of both police forces can be categorised as highly diverse.To illustrate this, the area of jurisdiction of the Mossos d'Esquadra includes the city of Barcelona (Spain's second largest city) and 13.7% of its residents were born outside of Spain (INE, n.d.), while Politie Utrecht's area of jurisdiction includes the city of Utrecht (The Netherlands' fourth largest city) and 20% of its residents have a non-western background (Atzema et al., 2008).Second, the police force is not only an instrument of the state, used to impose order where necessary; it also presents itself as a public service for citizens.This combination of responsibilities maximizes the number of potential arguments for a more diverse police force (Van Ewijk, 2012).Both the Mossos d'Esquadra and the Politie Utrecht manifest this twofold orientation.According to the Mossos d'Esquadra, their mission is to protect the freedom and safety of the citizenry, in accordance with the law, and this is why their tasks range from the protection of persons and goods to maintaining public order (Mossos d'Esquadra, n.d.).According to the Politie Utrecht, meanwhile, their mission is to be 'vigilant and servient', and maintain public order, fight crime, and offer help in order to ensure the safety of citizens and create/maintain their confidence in the police (Politie Utrecht, n.d.).
Third, both police forces engage in largely generic policing tasks.Generic police forces have the most daily contact with the broadest selection of the population.This might increase the imperative to adapt, when the population becomes more diverse.Therefore, police forces that only have specialized functions, such as traffic, narcotics, or internet crime, were excluded from selection.
The first difference between the two forces which might coincide with variation in the findings, is the history of immigration in their areas of jurisdiction.Immigration in Catalonia has been sudden and explosive since the 1990's, while the Netherlands, including the urban province of Utrecht, has experienced a steady flow of immigration since the 1950s.
Second, earlier studies have identified diverging governmental policies on diversity and integration.Dutch public policies have long been described as multicultural (Penninx, 2005), although some commentators have detected a shift towards more individual integration (Vertovec & Wessendorf, 2009) or even assimilation under the name of 'new realism' (Prins & Saharso, 2008).Spanish and Catalan public policies, meanwhile, are said to display a practical philosophy, a way of managing diversity which is not based on established and preconceived ideas, but rather on questions and answers generated by the practice of governance of diversity (Zapata-Barrero, 2010).
Third, the governance structure for the two police forces is different, and the decision-making power on organizational policies, such as HRM, lies with different actors.The Mossos d'Esquadra are directed by the Secretary of Security, part of the Department of Interior Affairs.This public organ is responsible for the selection and recruitment of new police officers, together with the public administrators from the Catalan police academy, the Institut de Seguretat Pública de Catalunya (ISPC), and involving police officers of the Mossos d'Esquadra.The same public organ also formulates, supervises and controls financial planning and staff management, leaving the Mossos d'Esquadra with exclusively operative management responsibilities.As such, the Mossos d'Esquadra are wholly composed of operative personnel, not administrative staff.The Politie Utrecht, on the other hand, is overseen by the Minister of the Interior and Kingdom Relations.Every four years the Minister of Justice draws up a policy plan detailing national policing priorities, national policing objectives, and the contribution expected from each police force.All forces must provide information on how they spend their grants.Additionally, the regulation of communication systems, uniforms and police pay is managed at central government level.The Politie Utrecht decides on its own staff requirements and applications).It also has its own career development policies, although the most senior officers are appointed by the Crown.Sixty-five percent of the Politie Utrecht's staff are sworn police officers, while 35 percent are administrative employees.It should be noted that all the above was valid at the time of the study, but this may change due to the reorganization which the Dutch police is currently in the process of realizing.

Data Collection and Content Analysis
It is important to make a clear distinction between units of data collection and units of analysis.In this article, the units of data collection are people and documents, while the units of analysis are the two dimensions of the analytical framework (the 'what' and the 'why').

Data Collection
This research obtains data from two types of sources: documents and people.Two data collection methods were employed.First, a search through the websites of the police forces and related organizations identified several relevant documents and individuals to include.Second, interviews provided information and access to other relevant documents, including those meant for internal use only.
When applying Prior's (2003) distinction between the content, the production, and the consumption of documents to this research, several observations can be made.First, with respect to the content, this article focuses mainly on the text.Second, documents from four different sources were included to multiply perspectives and diminish bias: the police force itself, the police academy, NGO's and associations of/for police officers, and the Ministry or Department of Interior Affairs.Third, the audience the author had in mind when producing the document may affect its content.Therefore, both public and internal documents were included to ensure that conclusions would be representative of the organization as a whole.When references are made to specific documents in Sections 3 and 4, they are included as footnotes.
Sixteen individuals were interviewed regarding the Mossos d'Esquadra (October 2008 to January 2010), and another 16 were interviewed regarding the Politie Utrecht (May 2009 to May 2010).The interviews were based on a semi-structured questionnaire, lasting between 60 and 90 minutes, generally conducted individually and face-to-face.Respondents were selected from four different backgrounds: the police force itself, the police academy, NGOs and associations of/for police officers, and the Ministry or Department of Interior Affairs.Respondents were selected for interview only if they occupied a position that allowed them to play an influential role in the process of policy-making and/or the implementation of policies related to diversity in the areas of recruitment, retention and promotion.

Content Analysis
A qualitative content analysis was used to interpret documents and interview reports with respect to the two units of analysis of Van Ewijk's analytical framework (Van Ewijk, 2011).These two dimensions are operationalized in Table 1.
Earlier interpretations were revised in the light of later readings and the results were interpreted for the whole body of literature and interview reports (Krippendorf, 2004), instead of evaluating each information source individually.
To make valid inferences, it is important that the classification procedure is reliable in the sense that it is consistent (Weber, 1990).By making the so-called qualitative areas in the research process recognizable, possibilities of objectifying individual case studies are opened up (Bos & Tarnai, 1999).Here, a discursive dialogue with four experts (two for each case) took place in order to critically evaluate the questionnaire and the operationalization of the analytical framework, which led to some modifications.As the content of the sources cannot be deduced independently of the reader's interpretation (Krippendorf, 2004), triangulation also took place during the data collection process, whereby preliminary conclusions were shared with fellow researchers, and after the empirical material had been analyzed and the results presented in this article, two experts were asked whether they agreed with the author's interpretation of the texts.

What Does Diversity Mean?
Table 2 offers a visual summary of the definitions of diversity used in the contexts of the Mossos d'Esquadra and the Politie Utrecht.As can be observed, these definitions are strikingly similar, and the difference lies in the nuances.Sections 3.1, 3.2 and 3.3 elaborate on the analyses that led to this conclusion.

A. Selection of modes of differentiation
Existence and frequency of references to specific 'forms of diversity', such as gender, religion, views, migrant background, age, etc.

B. Interpretation of modes of differentiation
Practical meaning of key terms related to these forms of diversity in specific contexts, such as 'immigrants', 'old', 'extrovert', etc. C. Categorization of modes of differentiation: hard or soft, old or new, collective or individual Hard-soft: practically unchangeable/permanent versus changeable/less permanent.Old-new: historical references to this form of diversity in the context versus no historical references.Collective-individual: where the mode of differentiation is linked to specific characteristics which are seen as constituting a group identity versus where it is linked to characteristics that are randomly distributed among individuals and no group identity is indicated.

Motivation for diversity within the organization (deontological dimension: the 'why')
A. Type of arguments: moral or practical Key concepts related to moral arguments: justice, social (in)equality, (in)equity, rights, (anti)discrimination, human agency, representativeness, symbolic (dis)advantages, responsibility, etc. Key concepts related to practical arguments: effectiveness, efficiency, results, practical (dis)advantages related to goal achievement, such as productivity, creativity / innovation, absenteeism, staff turnover, etc.

B. Desired base of difference: individual or collective
Individual-collective: end goal is to support individuals (ensuring equal opportunities) versus end goal is to support groups (ensuring collective characteristics are recognised and positively valued).

Selection of Modes of Differentiation
Although other modes of differentiation (such as age, religion, colour, physical validity, sexual orientation, educational background) also come to the fore 1 , the modes of differentiation that are most dominant in both contexts are gender and migrant background.
In terms of recruitment, for instance, specific attention has been paid to gender diversity in the the Politie Utrecht's recruitment communications since the 1980's and 40% of vacancies are reserved for suitable female candidates until the last possible moment, after which these openings are filled with other suitable candidates 2 .Also, there has been specific attention paid to diversity in terms of migrant background.Since the 1990's 30% of vacancies have been reserved for suitable candidates with a migrant background until the last possible moment 3 , and several programmes from the Politie Utrecht (IOOV, 2009) and a programme of the Police Academy (Politie Utrecht, 2008a) have been designed to reach possible candidates with a migrant background.Finally, the national recruitment website displays photos of female police officers and police officers with a migrant background more often than expected based on the proportion of police officers they represent within the police.The only other mode of differentiation that occasionally comes to the fore in recruitment communication is sexual orientation: the Politie Utrecht participates in Gay Pride Parades, and the 2009 national recruitment campaign included a famous openly gay Dutch magician, Hans Klok.These conclusions are illustrated in Figure 2.   ).The Regional Expertise Centre of Diversity (RECD) of the Politie Utrecht (with 3.4 FTE and a budget of 100,000 Euros in 2009) focuses heavily on diversity in terms of background, immediately followed by gender, and some attention is also paid to age and sexual orientation (RECD, 2007).Finally, the Politie Utrecht has various police officers' associations, such as the Kleurrijk Politie Netwerk Utrecht (KPNU), which supports police officers from migrant backgrounds (KPNU, 2007) and the Homo Informatie Punt (HIP)6 , which supports gay and lesbian police officers and seeks contact with the gay and lesbian communities (Politie Utrecht, n.d.).Furthermore, female police officers of the Politie Utrecht participate in national female networks7 .Other associations mentioned are national police unions, which also provide specific counselling for certain categories of police officers, for example young, female, gay, ethnic minority, or senior officers (ACP, n.d.; ANPV, n.d.; NPB, n.d.).
The Mossos d'Esquadra are provided with basic and advanced education on diversity at the Catalan Police Academy (ISPC).This focuses mostly on the diversity of migrant backgrounds in Catalan society8 , although there are also some short courses on gender diversity9 and diversity in sexual orientation10 .UNESCOCAT offers sessions on diversity of background in society outside of the organization on request11 .AIL-MED (AIL-MED, n.d.), an association that strives to improve the position of police officers that have become handicapped in the line of duty, has also received some juridical attention (Generalitat de Catalunya, 2008).Gaylespol is an association for gay, lesbian and transsexual police officers in Catalonia, which also has members from other (local and national) police forces in Catalonia.Its activities have no formal place within the organization of the Mossos d'Esquadra, in contrast with the Comisió de les Dones Policies (Female Police Officers Commission), which was founded in 1999 on the request of the police commissioner of Girona (Gubianes, 2000) and, after a few years of minimal attention, was formally reinstated in 201012 .Finally, some female police officers from both the Mossos d'Esquadra and the Politie Utrecht have or have had contact with the European Network for Policewomen, the ENP (ENP, n.d.).
Finally, as regards promotion, statistics concerning recruitment and membership of both police forces could only be found on the inflow, outflow, and through-flow of female police officers and officers with a migrant background.
In sum, the first choice made by both police forces with respect to the definition of diversity is that the same two modes of differentiation are selected: gender and migrant background.

Interpretation of Modes of Differentiation
'Gender' has the same practical meaning in both contexts: it refers to the biological distinction between male and female.'Migrant background' is also interpreted in a similar way: it refers to people who have been 'born abroad' or who have one or two parents that were born abroad, where in both contexts 'born abroad' usually refers to 'born outside of the European Union'13 .However, there are some small differences between the two contexts.
The Politie Utrecht use the term 'allochtoon' in relation to diversity in terms of migrant background: a person with at least one parent who was born abroad (CBS, n.d.).Afterwards, a distinction is made between those who were born abroad themselves ('first generation'), those who were born in the Netherlands but whose parents were born abroad ('second generation'), and those whose grandparents were born abroad ('third generation').Registration of the category 'allochtoon' versus 'autochtoon' is voluntary for police officers, which has lead to a significant third category of 'registration denied' in HRM statistics.
The most dominant term in the context of the Mossos d'Esquadra is 'immigrants' (ADN, 2007;Publico, 2009).The ISPC registers only 'students who were born abroad, or have parents who were born abroad'14 .In contrast, no reference is made to diversity in terms of migrant background in internal documents, such as HRM statistics.
In sum, the second choice made with respect to the definition of diversity is that in both contexts gender is interpreted as the physical differences between male and female, and migrant background is interpreted as being born abroad (i.e.outside of the European Union) or being a descendent of those born abroad.The only difference is that more differentiation in terms of generations are included in the interpretation of the Politie Utrecht.

Hard versus Soft
In the contexts of both the Mossos d'Esquadra and the Politie Utrecht, gender is considered a relatively hard mode of differentiation: a human difference that is inborn and/or exerts an important impact on early socialisation and has an ongoing impact throughout life (Griggs, 1995).For example, the message of the Program for Gender Equity of the Department of Interior Affairs in Catalonia is that 'women are equal, but not the same' (Gender Equity Programme, unpublished) 15 .In both contexts, this is seen as a consequence of an influential mix of physical difference (women as child bearers), combined with socio-cultural convictions about the difference in gender roles, especially for families with young children (women as caretakers), resulting in gendered preferences and ambitions, and gendered evaluations of capacities (Kop & Van der Wal, 2008)  16 .
In the context of the Politie Utrecht, migrant background is also considered as a hard mode of differentiation.It is not considered temporary, as often the second and even the third generation of immigrants are included in the concept of 'allochtoon' in policy documents (Politie Utrecht, 2011;RECD, 2007).
In the context of the Mossos d'Esquadra migrant background is seen as a soft mode of differentiation: a human difference that helps to distinguish the self from the other but is seen as less permanent and hence adaptable (Litvin, 1997).It is considered temporal, until some administrative (ie.obtaining Spanish nationality) and practical (ie.learning Catalan) differences associated with this mode of differentiation disappear.For example, a UNESCOCAT 17 spokesperson stated that the entrance of those with a migrant background into the Mossos d'Esquadra was 'just a matter of time', while the director of the ISPC 18 stated that it was a 'matter of time and critical mass'.

Old versus New
It is complicated to characterize gender as either an old or a new mode of differentiation.On the one hand, respondents tend to consider it as old, for gender has almost always constituted an important difference in human societies.On the other hand, only since the Police Emancipation Plan was published in 1990 have policies been formulated to enhance the position of female police officers within Dutch police forces, resulting in the foundation of the Association for Police and Emancipation (SPE) in 1991 and the official establishment under Dutch law of the European Network for Policewomen (ENP) in 1994.Gender only became an issue within the police after national, external developments on the topic 19 .Also, attention to issues of gender in the context of the Mossos d'Esquadra did not result in institutional action until the Female Police Officer Commission was founded in 1999.The issue received attention once again in 2007 with the establishment of the Program for Gender Equity, and only in 16 Interview with the Secretary of Security (Department of Interior Affairs): 13 January 2010, Barcelona.  1  It is also less clear whether migrant background is perceived as old or new by the Politie Utrecht.On the one hand, migrant background has been a prominent form of diversity since the 1960s and was already on the agenda of the Politie Utrecht in the 1980s.On the other hand, this attention only resulted in institutional action in 1998, with the foundation of Expertisecentrum Politie & Allochtonen (EXPA), focusing on the interaction between the Politie Utrecht and persons with a migrant background (Initiatives for Change Nederland, n.d.).EXPA's focus changed to include the integration of officers with a migrant background when it was transformed into the Regional Centre of Diversity Expertise in 2005 (RECD, 2007).
Migrant background in the Catalan context is considered as new in the sense that the presence of minorities with a migrant background from outside of Europe is relatively new in Catalonia (Ajenjo, 2008), but it is considered as old in the sense that Catalonia has been presented as a nation of immigration throughout its history, and new immigrants are expected to integrate with the society that already exists, just as, for example, immigrants coming from the south of Spain have done.

Individual versus Collective
In both contexts, gender is perceived as a mode of differentiation that is essentially collective.It is linked to a collective whose members are generally considered to have some specific characteristics (that imply meaningful difference) in common because of their gender, and these characteristics are believed to create common experiences that contribute to a group identity.In this study, sources refer to the physical capacity to bear children, preferences or ambitions with respect to work, and 'feminine' capabilities.
Migrant background is also perceived as a collective mode of differentiation in both contexts.Being born abroad or having (grand)parents who were born abroad is linked to the observation that, in general, almost all candidates in this group lack specific requisites (for example, mastery of the Dutch or Catalan language21 , or Spanish nationality) or experience in specific capacities (for example, assertiveness (Politie Utrecht, 2011)).Also, it is linked to a specific value that members of this col-lective can offer the police force: inside knowledge of other habits, religions, languages, etc.
In sum, the third choice with respect to the definition of diversity is that, on the one hand, gender is seen as a hard, old, and collective mode of differentiation in both contexts.Migrant background, on the other hand, is considered as a collective mode of differentiation in both contexts, but soft and new in the case of the Mossos d'Esquadra and hard and old in the case of the Politie Utrecht.

What Is the Motivation for Increased Diversity?
Figure 4 offers a visual summary of the Mossos d'Esquadra and Politie Utrecht's motivations for diversity.As described in Table 1 (in Section 2), a distinction is made between moral and practical arguments on the one hand and a desired base of difference that is either individual or collective on the other hand.As with their definitions of diversity, also the two police forces' motivations for diversity are strikingly similar.This section elaborates on the analyses that led to this conclusion.

Mossos d'Esquadra
In general, the end goal of diversity policies for the Mossos d'Esquadra is more often related to supporting individuals and ensuring equal opportunities in spite of collective differences than to supporting collectives and recognizing their value.In terms of the analytical framework, the desired base of difference is individual.For example, the end goal of the Social Diversification Programme is to level the playing field by educating candidates in areas they have more problems with because of their migrant background, such as the Catalan language (Miró, 2007).Also, the Gender Equity Programme confirms that the Department of Interior Affairs wants to use its human capital to the maximum, and it can only do that when all employees work under equal conditions, and differences such as gender do

Mossos d'Esquadra
Politie Utrecht not create inequalities (Gender Equity Programme, 2010).Finally, Gaylespol strives for a working environment without discrimination, in which everyone can be open about their sexual orientation and is treated with respect regardless (Sneijder, 2006).Both moral and practical arguments are employed in favour of diversity in the police force.On the one hand, diversity within the police force is linked to: representing the society it serves, combating stereotypes among police officers and ensuring they value colleagues with a diverse profile, and ensuring respectful behaviour towards the population 2223 .On the other hand, the same sources refer to the added practical value of increasing the quality of the police service by responding more adequately: offering a more diverse repertoire of reactions, more knowledge of and sensitivity towards different cultures (related to gender, class, migrant background, sexual orientation, etc.), and increasing the willingness of the population to cooperate by increasing their trust and sense of proximity to the police (UXXS, 2008).
Finally, it is interesting that moral arguments appear to be more dominant with respect to gender diversity (related to equality, (social) justice, and nondiscrimination) (Gender Equity Programme, 2010; Gubianes, 2000) 24 , while diversity in terms of migrant background is almost exclusively called for on the basis of practical arguments (to be more effective against gangs with foreign origins (ADN, 2007), to teach colleagues how new communities think and act (García, 2007), because of foreign language skills, and to solve conflicts, prevent riots and solve crimes 25 ).

Politie Utrecht
According to various sources (LECD, 2008b), the case for diversity used to be almost exclusively made with moral arguments (as stated literally by one of the respondents: "something fun and politically correct for poor target groups"), but has been presented as a business issue (´a strategic theme related to multicultural skills´) since 2005.However, when analyzing the sources with the analytical framework a more nuanced image appears.First of all, the end goal of the Politie Utrecht's diversity policies is more often related to supporting individuals and ensuring equal opportunities regardless of collective differences, than to supporting collectives and recognizing their value.In terms of the analytical framework, the desired base of difference is individual.For example, the Politie Utrecht claims that their mission is to be a police force for everyone, which implies proactive efforts to combat discrimination and exclusion mechanisms outside as well as inside the organization.As such, the goal of diversity management is meant to contribute to equality so that all members of the Politie Utrecht can develop their own talents and qualities (RECD, 2007).In this context, equal opportunities for all are emphasized (Politie Utrecht, n.d.), and contribute to a safe working environment that allows everyone to develop themselves and apply for promotions (LHP, n.d.).
Second, arguments for diversity within the police force are both moral and practical, although practical arguments are more dominant.On the one hand, it is said that the police force should be representative of society in order to symbolize that the police is there for everyone, and to show that the organization is open and accessible and offers an equal working environment for all (KPNU, 2007;Snijder, 2008) 26 .On the other hand, diversity within the organization is said to enable the police to do their work better because the cooperation of the population will increase, and because the quality of the police service will rise, which produces better results (Kop & Van der Wal, 2008;LECD, 2005;LHP, n.d.;Ministry of Interior Affairs, 2006;National HRM Programme, 2008;Politie Utrecht, n.d.;Politie Utrecht, 2008b) 27 ; a more diverse police force is more innovative and will be able to use more variation in techniques and procedures in different situations; diversity within the police force will make it an attractive employer for talented candidates-even when there is competition on the labour market; and police officers with a diverse profile bring specific capacities with them which increases professionalism.For example, female police officers bring feminine leadership qualities, which might stimulate creativity in their teams.They are also said to have potential added value in cases of domestic violence or a de-escalating effect in cases of potential riots.Police officers with a migrant background might go undetected when performing surveillance duties in diverse neighborhoods.They can also teach their colleagues about their communities.Gay police officers have a similar effect.

Opportunities for Further Theory Building
Assertions on causality are outside the scope of this article.However, the empirical application of the analytical framework does provide input for various hypotheses that might be tested and confirmed through further research, which demonstrates its potential to contribute to academic theory building.

Exploring Explanations
The two cases were surprisingly similar in their definition of diversity and their motivation for encouraging diversity within the organization, regardless of differences in contextual factors (outlined in Section 2.1) that would lead one to expect otherwise.This implies a causal relation between the contextual factors that were similar on the one hand and the uniformity of the conception of diversity on the other.Sound confirmation and explanations of this causality, however, require additional research.In this regard, institutional isomorphism, as introduced by DiMaggio and Powell (1983) could be a plausible direction to explore: police organizations might either emulate the governments they work for (coercive isomorphism) or copy each other's practices (mimetic processes).The fact that gender only became an issue within the Politie Utrecht due to external developments on a national level, and it only gained attention within the Mossos d'Esquadra in 2007 with the initiation of the Catalan Program for Gender Equity corroborates this line of thought.Another interesting factor to explore would be the influence of the European Union in both cases.
Again, it is surprising that similarities are more dominant than differences, when one considers the amount of publications in political theory dedicated to what Castles (2002) referred to as 'national integration models' and Penninx et al. (2004) as 'policy discourse on diversity', which was also mentioned in section 2.1 as a contextual factor that differs for both cases.Even so, this factor could be related to specific findings.For example, national discourse on diversity and integration in the Netherlands could have influenced the definition of diversity in terms of migrant background as a hard mode of differentiation.According to Ghorashi (2006) the experience of pillarization and the specific Dutch implementation of the welfare state led to Dutch policies regarding (ethnic) minorities that have a strong categorical base.Also, regional discourse on diversity and integration in Catalonia-which is a practical philosophy, according to Zapata-Barrero (2007)could have influenced the definition of and motivation for diversity within the Mossos d'Esquadra, even more so because the authority concerning HRM policies re-sides mostly within the Department of Interior Affairs.
The second discrepancy between the two police forces (migrant background as an old and hard mode of differentiation for the Politie Utrecht versus a new and soft mode of differentiation for the Mossos d'Esquadra) might be explained by differences in the two regions' histories of immigration.Over the past few decades, Catalonia has mostly received small numbers of immigrants from countries in Latin-America with similar cultures and a shared history, and has experienced a sudden increase in immigration from other cultures first from the early1990's.The Netherlands, meanwhile, experienced the arrival of relatively more immigrants with relatively more diverse cultural origins several decades earlier, and the presence of their second and third generation descendents since then.
Further research is necessary to test the abovementioned causal relations.Also, it would be enlightening to examine the relative impact of all factors on the variation in definitions of and motivations for diversity within police organizations.Another option is to conduct a longitudinal study and compare one of the regional units in the future Dutch police force (as soon as the reorganization is finalized) with the former Politie Utrecht.Finally, the same causal relations could be explored in other public service sectors, such as health or education, or in the private sector.

The Influence of Definition and Motive on Effectiveness
One of the preconditions for effective policies is that all involved actors have the same goal in mind, they understand it in the same way, and they share the motives for that goal (Dunn, 1994;Kotter, 1996).As such, the definition of and motivation for diversity are expected to have a significant influence on the effectiveness of diversity policies.For example, differences in the definition of and motivation for diversity between those in command and those on the front line could create resistance against these policies, which would obstruct their implementation.Following the thinking of Paoline ( 2003), who argues that organizational style, rank and individual officer style create variation in the cultural homogeneity of officers, this would not be unlikely.Liff (1999) also mentions this potential pitfall.
Both the Politie Utrecht and the Mossos d'Esquadra seem still to be looking for a narrative for diversity within the police force that includes both moral and practical arguments.The Politie Utrecht have gone from one side of the spectrum to the other (from mostly moral to mostly practical), while the Mossos d'Esquadra display a simultaneous dichotomous perspective (gender is linked to moral and migrant background to practical arguments).
However, a mix of moral and practical arguments for each selected mode of differentiation is more con-vincing than an exclusive focus on one type of argument.By using only moral arguments, an organization runs the risk of creating a feeling of charity, rather than a necessity which is high on the list of priorities, especially in times of economic decline.By using only practical arguments, an organization becomes very dependent on studies that demonstrate the direct relationship between diversity and better results, while in many cases this relationship has not yet been conclusively demonstrated by empirical research (Wise & Tschirhart, 2000).The hypothesis, then, would be that future effectiveness of the police force will be positively influenced by a balanced, multi-layered motivation (i.e.including both practical and moral arguments).This will need to be confirmed by comparing outcomes of diversity policies between organizations with singular and multiple-type motivations.

Conclusions
The first goal of this article was to demonstrate the research potential of the analytical framework on diversity (policies) as defined elsewhere by Van Ewijk and briefly outlined in Section 2.2.2.Applying this framework to two empirical cases, the Politie Utrecht and the Mossos d'Esquadra, has confirmed that diversity policy is, in effect, a multi-layered concept both in theory and in practice: the distinction between the definition of diversity and the motivation for diversity within the organization is not merely theoretical, but one that can be observed empirically.Further, the application of the analytical framework has generated possibilities for academic theory building.This article formulates (new) hypotheses on the determinants of variation in the conception of diversity, on a possible hierarchy in modes of differentiation in the European Union, and on the impact of the type of motivation for diversity on the effectiveness of diversity policies.
Second, in the process of applying the analytical framework, this article has provided a deeper understanding of the conception of diversity in the context of the Mossos d'Esquadra and the Politie Utrecht.This has in turn enabled the formulation of clear observations from a multitude of raw empirical data.The most important observation is that the findings were in general surprisingly similar.For example, both police forces focus on the same modes of differentiation (gender and migrant background) and both attribute similar practical meanings to these modes of differentiation.The only minor difference in their definitions of diversity is that migrant background is considered to be a soft mode of differentiation in the context of the Mossos d'Esquadra and a hard mode of differentiation in the context of the Politie Utrecht.While in both cases gender and migrant background are mainly identified as collective modes of differentiation in nature (which is an ontological observation), both the police forces strive for an organization in which individuals have equal opportunities regardless of collective differences (which implies a deontological choice for diversity as an individual concept).Both moral and practical arguments are employed to call for diversity within the police organization, although practical arguments seem far more dominant in the case of the Politie Utrecht.

Figure 1 .
Figure 1.The areas of jurisdiction of the Politie Utrecht and the Mossos d'Esquadra in 2011.

1
Interview with the commanding officer responsible for the RECD (Politie Utrecht): 11 August 2009, Utrecht. 2 Double interview with the administrator and police officer responsible for facilitating and monitoring diversity within the organization (HRM department of the Politie Utrecht): 19 June 2009, Utrecht. 3 Double interview with the administrator and police officer responsible for facilitating and monitoring diversity within the organization (HRM department of the Politie Utrecht): 19 June 2009, Utrecht.

Figure 2 .
Figure 2. Diversity within the police force on the Dutch national recruitment website.No target groups have been officially defined to be recruited for the Mossos d'Esquadra.Even so, photos of female police officers are usually included in the general leaflets that the Catalan government produces and distributes 4 , and in 2008 and 2009 the Gender Equity Program of the Department of Interior Affairs launched campaigns specifically aimed at women.Efforts to connect with the gay community, such as participation in the 2009 Gay Pride Parade, have been initiated exclusively by the gay police officer association, Gaylespol (Gaylespol, n.d.).Finally, there was a separate publicity campaign to recruit candidates from migrant backgrounds for the Social Diversification Programme, designed to help these candidates prepare for the official selection procedure in 2007 and 2008.Figure 3 displays a general call for candidates (basic education in 1997 and 2009), a call for female candidates (basic education in 2008) and a call for candidates with a migrant background (Social Diversification Programme in 2007).

Figure 3 .
Figure 3. Diversity in publicity for the Mossos d'Esquadra.In 2011, the Dutch Ministry of Interior Affairs formulated gender and migrant background targets, with financial consequences for the Politie Utrecht.In 2008, the Ministry of Interior Affairs and Police Commanders founded the 'PolitieTop Divers' working group in order to stimulate and monitor these objectives (Politietop Divers, n.d.).Also, the Dutch Police Academy's basic and advanced education on diversity focuses almost exclusively on migrant backgrounds, both within and outside the organization (SPL, n.d.) 5 , just as the 'knowledge lectorate' founded in 2008 focused on multicultural professionalism (LECD, 2008a).The Regional Expertise Centre of Diversity (RECD) of the Politie Utrecht (with 3.4 FTE and a budget of 100,000 Euros in 2009) focuses heavily on diversity in terms of background, immediately followed by gender, and some attention is also paid to age and sexual orientation(RECD, 2007).Finally, the Politie Utrecht has various police officers' associations, such as the Kleurrijk Politie Netwerk Utrecht (KPNU), which supports police officers from migrant backgrounds (KPNU, 2007) and the Homo Informatie Punt (HIP) 6 , which supports gay and lesbian police officers and seeks contact with the gay and lesbian communities (PolitieUtrecht, n.d.).Furthermore, female police officers of the Politie Utrecht participate in national female networks 7 .Other associations mentioned are national police unions, which also provide specific counselling for certain categories of police officers, for example young, female, gay, ethnic minority, or senior officers(ACP, n.d.; ANPV, n.d.; NPB, n.d.).
Interview with the coordinator of the diversity courses for the police (UNESCOCAT): 5 November 2008, Barcelona. 18Interview with the director of the Catalan Police Academy (ISPC): 19 January 2010, Mollet del Vallès. 19Double interview with the administrator and police officer responsible for facilitating and monitoring diversity within the organization (HRM department of the Politie Utrecht): 19 June 2009, Utrecht.
2009 did this lead to a request by a few female Mossos d'Esquadra for an independent internal team, which was approved in 2010 20 .

22
Presentation at the conference of UNESCOCAT on police and diversity on the 17th of December 2009 by the research and international cooperation coordinator of the ISPC; Interview with the director of the Catalan Police Academy (ISPC): 19 January 2010, Mollet del Vallès; Interview with the Secretary of Security (Department of Interior Affairs): 13 January 2010, Barcelona. 23Double interview with the president and vice-president of Gaylespol: 12 May 2009, Mollet del Vallès. 24Interview with the work group representative of the Gender Equity Programme (Department of Interior Affairs): 21 April 2009, Barcelona. 25Interview with the Secretary of Security (Department of Interior Affairs): 13 January 2010, Barcelona.

Table 1 .
Operationalizing the 'What' and 'Why' of Diversity.