Diversity management in three Latin American countries: an institutional theory perspective

Andrés Raineri (Escuela de Administración, Pontificia Universidad Catolica de Chile, Santiago, Chile)

Academia Revista Latinoamericana de Administración

ISSN: 1012-8255

Publication date: 4 June 2018

Abstract

Purpose

The purpose of this paper is to use an institutional theory (IT) approach to analyse the institutional context for diversity management (DM) in Chile, El Salvador and Guatemala, the influence of such a context on the DM issues that organisations’ face, and the DM practices that firms implement.

Design/methodology/approach

Focus groups and a survey are used to assess managers and workers’ perceptions about DM in their countries, while an analysis of the content of firms’ web pages is used to assess the formal public information about DM provided by firms.

Findings

Results suggest prevalence of perceptions, among both managers and employees, of discrimination at the workplace, facilitated by cultural forces that undervalue human diversity. Firms’ DM responses seem to be led by multinational companies, and focussed on fighting discrimination, facilitating inclusion and pursuing external legitimacy.

Originality/value

This study is among the first to analyse DM in Latin American organisations. Furthermore, three studies, with different methodologies, support several IT propositions that emphasise the role of institutional forces in explaining organisations’ implementation of DM practices. The relevance and challenges of developing DM programmes in Latin America are discussed.

Propósito

Este artículo utiliza la Teoría Institucional para analizar el contexto que rodea la Gestión de la Diversidad (GD) en empresas operando en Chile, El Salvador y Guatemala, y la influencia de este contexto en los temas de GD que enfrentan las organizaciones, y en las prácticas de GD que éstas implementan.

Diseño/metodología/enfoque

Se utilizan grupos focales y una encuesta para evaluar la percepción de administradores y trabajadores sobre la GD en sus países, y se utiliza un análisis del contenido de las páginas web de las empresas para evaluar la información pública y formal proporcionada por las empresas sobre la GD.

Resultados

Los resultados sugieren la prevalencia, tanto entre los directivos como empleados, de percepciones de discriminación en el lugar de trabajo, facilitadas por fuerzas culturales que desvalorizan la diversidad humana. Las intervenciones de las empresas en GD aparecen lideradas por compañías multinacionales, y se enfocan en combatir la discriminación, facilitar la inclusión y buscar legitimidad externa.

Originalidad/valor

Este artículo es uno de los primeros en analizar la GD en organizaciones latinoamericanas. Además, se presentan tres estudios que, con diferentes metodologías, apoyan varias proposiciones de la Teoría Institucional, las que enfatizan el rol de las fuerzas institucionales en explicar la implementación de prácticas de GD por parte de las organizaciones. Se discute la relevancia y desafíos del desarrollo de programas de GD en América Latina.

Keywords

Citation

Raineri, A. (2018), "Diversity management in three Latin American countries: an institutional theory perspective", Academia Revista Latinoamericana de Administración, Vol. 31 No. 2, pp. 426-447. https://doi.org/10.1108/ARLA-08-2016-0220

Download as .RIS

Publisher

:

Emerald Publishing Limited

Copyright © 2018, Emerald Publishing Limited


Introduction

Human diversity has been labelled a double-edge sword in management literature. On the one hand, diversity has the potential to increase the quality of decision making, enhance creativity and innovation, and facilitate relationships with customers from different backgrounds. On the other hand, human differences might generate prejudice, discrimination, interpersonal conflict and decrease feelings of inclusion (Stevens et al., 2008). Diversity management (DM) refers to the management of human diversity at work, including cultural differences, physical differences, such as appearance and disabilities, and inherent differences, such as race and sex (Ng and Stephenson, 2015). Furthermore, previous literature shows that properly designed and implemented DM programmes and activities can enhance the positive outcomes associated to diversity and prevent its concomitant problems (Ng and Stephenson, 2015; Alhejji et al., 2016).

The origin of DM has been associated with the affirmative action and equal employment mandatory legislation initiated in the USA in the mid-1960s (Klarsfeld et al., 2016). This early approach emphasised developing legislation to reduce discrimination and levelling the field for disadvantaged groups. However, it was not until the 1990s that a second approach gave rise, in the USA, to the concept of DM both in literature and practice, in reference to voluntary and discretionary interventions of management, intended to reap the benefits of human diversity at the workplace, and adapt to demographic changes generated by globalisation (Ng and Stephenson, 2015; Kelly and Dobbin, 1998). Later, in the 1990s, both voluntary and mandatory DM programmes were introduced in parallel in Europe (Klarsfeld et al., 2016).

There is no standard set of diversity management practices (DMPs) that every organisation should adopt, but rather the set of DMPs appropriate for a particular firm is context dependent, including issues such as its own workforce, labour market composition, local legislation and cultural issues (Ng and Stephenson, 2015; Syed and Özbilgin, 2009). Firm-specific DMPs should be designed to take advantage of the benefits of its workforce diversity, and minimise its potential disadvantages (Ng and Stephenson, 2015). Despite the latter, some of the most frequently described DMPs include the active recruitment of minorities, training supervisors in leading with respect and without biases a diverse workforce, providing fair compensation systems and developing formal organisational policies on DM (Ng and Stephenson, 2015). A comprehensive approach to classify DMPs is provided by Kossek and Pichler (2006) who proposed three main types of DMPs, based on the type of value generated by their implementation. First, some DMPs can decrease the costs of products or services. If DMPs defuse the negative consequences of diversity, such as prejudice, discrimination, conflict or employee turnover, then they should reduce a firm’s human resources costs, for example, by training managers to avoid personal biases when involved in a personnel selection process. The second type of DMPs are intended to configure a more diverse human capital to pursue an objective, as is the case when forming teams in order to generate innovation, creativity or better quality of products and services. Finally, the third type of DMPs are intended to induce feelings of inclusion in the workplace, lead employees to care about their organisations, and make their best contribution to work, for example, through formal recognitions (e.g. celebrations) of the value of human diversity within the organisation. Previous research shows that human diversity issues vary from firm to firm (Ng and Stephenson, 2015). For example, global companies need to prioritise the development of intercultural managers, while some local firms might need to prioritise the management of specific local minority groups. Furthermore, cross-country studies show that human diversity issues, and country-level responses such as legislation, also vary from one country to another (Klarsfeld et al., 2016). For example, India has given a prominent relevance to legislation related to social segmentation through casts (Donnelly, 2015), while, in the USA, race and ethnicity have been dominant issues in diversity-related legislation (Klarsfeld et al., 2016).

Most of the research on DM is conducted in the USA and Europe, leaving unanswered questions about the validity of such results in other national and cultural contexts (Klarsfeld et al., 2016; Syed and Özbilgin, 2009). Because of the limited literature on DM in Latin America (LA), a first objective of this study is to gain insight into the perceptions that managers and employees have, from firms operating in Chile, El Salvador and Guatemala, about DM in their countries. In particular, this study explores for commonalities and differences in the perceptions of these employees in relation to three topics: the diversity issues that organisations face in these countries; the institutional context that surrounds DM in their countries; and their perceptions of the responses that companies give to DM issues within such context. Furthermore, this study tests several propositions from an institutional theory (IT) perspective (Scott, 1995), about how the characteristics of the institutional environment surrounding firms influence employees’ perceptions about DM, and on firms’ implementation of DMPs (i.e. Yang and Konrad, 2011, p. 26). In particular, this study tests for IT propositions about how institutional forces influence the types of DMPs implemented, the purposes intended by firms when implementing DMPs, and the types of firms that engage in DMP implementation.

In the following sections, first, the previous literature on the institutional and historical context for DM in LA is described. Second, hypotheses are derived from IT proposing how the particularities of the LA institutional environment affect employees’ perceptions about DM, and DMPs implemented by firms. Then, a multi-method research design is used to analyse managers and subordinates’ perceptions about DM, and test the proposed hypotheses. Finally, theoretical contributions, limitations and practical implications are discussed.

Literature review

The Latin American workplace as an appropriate setting in which to study diversity management

During the last five centuries, the mingling of a multiplicity of races and cultures has shaped a rich mosaic of people in LA. When Spanish and Portuguese Europeans first disembarked in LA, during the fifteenth and sixteenth centuries, an estimated 43m people inhabited the region (Salzano and Bortolini, 2005). A myriad of different population groups has been recognised to exist before the arrival of the Europeans (Guzmán et al., 2006). These groups included large and complex civilisations, such as the Aztecs, Mayas and Incas, as well as much smaller less well-known groups, such as the Selknam in Tierra del Fuego, or the Nukak in the tropical rainforests of Colombia. In the next four centuries, approximately 10m enslaved African people were disembarked in LA and the Caribbean, accounting for near 90 per cent of victims in this tragedy (Voyages Database, 2016). Smaller migrations from almost every region in the world settled more dispersedly throughout LA (Salzano and Bortolini, 2005). All these ethnographic changes have created an ample range of differences within and across LA countries. For example, in Brazil, more than half of the population is Afro-descendant, and in Argentina, a large majority is of European origin. Within this regional mosaic of races and cultures, Chile and El Salvador are prototypical mestizo countries, which comprise people of Amerindian and Spanish decent, as is the case in most LA countries. On the other hand, Guatemala is better profiled as a LA country where more than half of its population is from one of several Amerindian cultural groups, which speak more than 22 different Mayan languages (Gibbons and Ashdown, 2010).

LA’s economic structure also varies across countries. In most South American countries, as is the case for Chile, natural resources contribute to national GPD in a proportion far above the world average of 5.6 per cent; on the other hand, most countries located in Central America, as is the case for Guatemala and El Salvador, agriculture and low-tech manufacturing (e.g. maquila) are some of their main sources of income (The World Bank, 2017). Statistics also show that 59 per cent of the workforce in Guatemala, 32 per cent in El Salvador and 14 per cent in Chile live below poverty levels (The World Bank, 2017). Additionally, roughly 69 per cent of the workforce in Guatemala, 57 per cent in El Salvador (The World Bank, 2017), and 37 per cent in Chile (OECD, 2016) participate in the informal sector of the economy, posing difficulties to enforce labour laws. All these differences make the case for the study of diversity on three different LA scenarios. On the one hand, the mestizo countries of El Salvador and Chile, one on them with relatively low poverty and informal employment rates, and the other with higher poverty and informal employment rates. On the other hand, Guatemala, with the majority of its population being from an Amerindian ethnicity, has higher levels of poverty and informal employment. Despite such historical, social and demographic background, DM at the workplace remains a neglected area of research in this region, and in these three countries in particular.

The institutional scenario for DM in LA

Several authors have proposed IT as a framework to study DM (i.e. Yang and Konrad, 2011; Kostova and Roth, 2002; Klarsfeld et al., 2016). IT places an emphasis on how contextual factors affect organisations’ management stance and practices. IT highlights three types of institutions that exert pressure on organisations: the legal and regulatory system, the professional and education processes and the cultural, ethical and social habits of a community (Scott, 1995). These contextual factors pressure organisations to adapt and conform to their expectations and principles in order to gain legitimacy among their stakeholders. Such legitimisation facilitates organisations’ interaction with stakeholders, such as employee cooperation, customer preferences and suppliers’ delivery of resources.

The LA context has its own institutional particularities, which are relevant to DM. From a legal stance, almost all LA countries have ratified International Labour Organisation’s conventions including anti-discrimination protection, union participation and collective bargaining labour rights, and have adapted local labour laws to protect such rights (Burgess, 2010). However, there seems to be consensus in the literature that the enforcement of such legal rights is difficult (Burgess, 2010; Ronconi, 2012), and it has been attributed to the lack of knowledge or ambiguity of legal rights (Goldín and Dowdall, 2012), cultural issues and prejudice (Dulitzky, 2005), weak enforcement institutions (Ronconi, 2012; Levitsky and Murillo, 2013), and the difficulties in applying labour laws in the informal sector of the economy (Bensusán, 2009). Empirical evidence indicates that despite the existence of legislation to protect minorities, discrimination is frequently observed. For example, Atal et al. (2009), in a study of gender and ethnic wage gaps in LA countries (including those here studied), found significant wage differentials in detriment of women and ethnic minorities, which are persistent throughout countries, age, educational level and occupations. Also, career advancement and personnel selection processes in the region have been described as not being very equalitarian. For example, Maxfield (2005) interviewed women executives from six LA countries, including El Salvador and Chile, who described facing discriminatory practices when advancing their career goals. Similarly, evidence shows that gender and racial discrimination occur during the hiring process in Peru (Moreno et al., 2012; Galarza and Yamada, 2012), and Chile (Bravo et al., 2009). In sum, legal protection to avoid discrimination and protect diversity is present in LA, but its’ compliance has been reported as weak.

IT also proposes that the professional norms brought into the organisation by those employees hired and the training they receive within the firms influence organisations’ management and practices. Literature on DM programmes and practices in LA is very scarce and not encouraging. In one study, Suárez et al. (2015) interviewed managers from 37 hotel chains in Argentina, Colombia, Mexico and Spain, asking them about diversity and inclusion interventions carried out within their hotels, and concluded that whenever DM interventions were made, they responded to the concerns of particular managers, rather than to organisations’ formal DM programmes. In a similar vein, Chiappetta et al. (2011) conducted interviews in 15 large Brazilian companies in order to assess their level of involvement in DM and concluded, “diversity is infrequently systematised or addressed in Brazilian companies”. To the best search efforts of this author, no research has been conducted on DM training in the three countries under study. However, in the case of Chile, there is an indirect evidence that indicates a very low level of training in DM. Registries from a widely used government tax programme that allows firms to deduct part of the investment used in employee training programmes indicates that from a total of 4,784 courses registered between the years 2013 and 2017 under the topic of “human resources management”, only seven courses dealt with DM training (SENCE, 2017). Such a low number of training programmes in DM makes these programmes a rarity in Chile. Unfortunately, it was not possible to find similar data on DM programmes in Guatemala or El Salvador. Despite the scarcity of public information about DM in LA, existing research seems to suggest that professionally driven DM is an infrequent activity in the region.

IT also proposes that the cultural, ethical and social habits of a community pressure organisations to adapt and conform to these expectations and principles (Scott, 1995). Large cultural studies consistently describe LA as culturally homogeneous, and characterised by high “uncertainty avoidance” and “high power distance” cultural values (i.e. Hofstede, 1980; House et al., 2004). These cultural traits can be indirectly or directly associated to lower levels of acceptance of individual differences. For example, uncertainty avoidance is described as “the extent to which a society feels threatened by uncertain and ambiguous situations and tries to avoid these situations by […] establishing more formal rules, not tolerating deviant ideas and behaviours” (Hofstede, 1980). Also, power distance is described as the extent to which organisations are hierarchically stratified, authority is based on obedience, and subordinates are not expected to express disagreement (Hofstede, 1980). In this sense, both uncertainty avoidance and power distance could be considered as cultural traits that do not highlight the value of individual differences. Taras et al. (2010) conducted a meta-analysis of research with Hofstede’s cultural values consequences on several outcomes and found, at country level, a significant negative relationship of power distance with measures of human rights, gender equality and income equality. Similarly, Taras et al. (2010) found a significant negative relationship of uncertainty avoidance with income equality and innovation. Furthermore, in the “gender egalitarianism” index from the GLOBE study that measures the degree to which a society minimises gender inequality, El Salvador and Guatemala received very low scores when compared worldwide (Javidan et al., 2006). Unfortunately, Chile did not participate in this last study.

Paternalism and authoritarianism have also been frequently considered as LA culture dimensions (Elvira and Davila, 2005; Pellegrini and Scandura, 2008). Paternalism has been described as a mixture of personalised caring for subordinates, sometimes getting involved in their non-work lives, maintaining strong authority and status hierarchy, while expecting loyalty and deference (Pellegrini and Scandura, 2008). A more authoritarian side of paternalism has been related to exploitative relations and to lower levels of commitment and trust (Pellegrini and Scandura, 2008). From this perspective, it could be argued that paternalism and authoritarianism are cultural traits that do not promote the value of individual differences, but rather of conformity.

Additionally, previous literature indicates the presence in LA of cultural traits that directly undervalue human diversity, such as machismo, classism and racism. Some studies indicate that LA women perceive machismo as a problem to their career advancement (Maxfield, 2005; Cárdenas et al., 2014). Classism and racism have their own place in LA history. In LA, race mixture has been a relevant topic of sociological research given that during colonial centuries, Amerindian, Europeans and Africans came into large-scale contact with one another (Telles and Sue, 2009). In most LA nations, the racially mixed person became the national race archetype whether called mulato, mestizo or moreno. In LA, race mixture led to cultural ideologies of racial hierarchy. Researchers have described elaborate caste systems as cultural ideologies used to categorise different types of mixed-race persons (Moreno and Oropesa, 2012; Mazzolini, 2007; Telles and Sue, 2009). These racial ideologies have been related to class segmentation and discrimination (i.e. Dulitzky, 2005; Atal et al., 2009). For example, Nuñez and Gutiérrez (2004) showed empirical evidence that in Chile, amongst business and economic graduates, income is better predicted by socioeconomic class of origin rather than by academic performance. In the same vein, Atal et al. (2009) found significant gender and race wage gaps, disfavouring women and Amerindian ethnic groups throughout LA. In summary, LA culture has been described as possessing several traits that can be associated to lower levels of acceptance of individual differences, as well as other cultural traits that more openly undervalued human differences, and have been associated to discrimination.

From an IT perspective, it has been proposed that employees develop attitudes towards institutions, and the ideas they promote, by processing information from their local environment (i.e. Zajac and Bazerman, 1991). In this study, it is expected that the local environment previously described will influence employees’ attitudes towards local institutions. In particular, due to the previously described historical background and legal reforms protecting human diversity at the workplace, it is proposed that:

H1.

Local employees will show a higher proportion of verbalisations describing a positive influence of legal institutions on DM than a negative influence.

Also, due to the cultural issues mentioned earlier indicating prejudice and discrimination, it is proposed that:

H2.

Local employees will show a higher proportion of verbalisations describing a negative influence of cultural issues on DM than a positive influence.

Finally, due to the apparently low levels of DM professionalisation described above, it is proposed that:

H3.

Local employees will make more verbalisations about legal and cultural institutions having an influence on DM than normative/professional institutions.

The influence of institutional forces on firms’ development of DMPs in LA

IT proposes that the institutional context can shed light on the value that managers attribute to specific resources, the abilities available in that environment to make use of those resources, the regulatory constraints, and the social acceptance of the deployment of such resources (Scott, 1995; Yang and Konrad, 2011). For example, the legal environment can prompt managers to correct unfair practices towards minorities in their organisations (Kelly and Dobbin, 1998). In addition, the normative environment can educate managers to protect and make better use of the human diversity available (Alhejji et al., 2016), and cultural values shared in a society can increase or hinder the value of human diversity and DM investment (Syed and Özbilgin, 2009). Furthermore, IT proposes that when organisations face external institutional pressures, they will tend to adapt to those pressures by attempting to gain external or internal legitimacy (Kostova and Roth, 2002). When pursuing external legitimacy, the intention is to indicate the adoption of certain practices to external stakeholders, such as clients or governmental agencies, in order to gain acceptance and interact with those actors. In the early stages of organisational responses to affirmative action in the USA, when companies had to comply with anti-discrimination laws, and enforcement was weak, such patterns of responses were observed (Kelly and Dobbin, 1998). When pursuing internal legitimacy, the intention is to obtain the acceptance of the practices by organisational members, in order to gain their commitment and facilitate inclusion. The weak enforcement of anti-discrimination legislation, apparently poor DM professionalisation, and the presence of cultural biases that undermine the value of human diversity suggest that LA employees will perceive the adoption of DMPs as intending to gain external legitimacy. The reason for this is that if enforcement of anti-discrimination legislation is weak, pressure for firms to comply weakens. Likewise, if professionalisation of DM is poor, the knowledge and skills to better manage diversity are not available. And, if cultural biases that undermine the value of human diversity are frequently shared, there will be no willingness to invest in DMPs beyond external legitimacy:

H4.

DMPs used by firms in LA will be perceived by local employees as attempting to gain external rather than internal legitimacy.

The types of DMPs implemented by firms should also be shaped by the institutional context. As stated earlier, Kossek and Pichler (2006) classified DMPs into three types: DMPs intended to defuse the negative consequences of diversity (e.g. prejudice or discrimination), DMPs intended to create value through creativity and better quality of products and services, and DMPs inducing feelings of inclusion in the workplace in order to create better work climates. Because of the previously described legal pressures to comply with non-discriminatory practices in LA, and historical and cultural evidence of unfair treatment towards several groups in this region, it can be expected that DMPs in LA should be mostly geared towards decreasing discrimination and generating feelings of inclusion. Furthermore, the presence of cultural traits that undervalue human diversity as a resource suggests that DMPs geared to take advantage of the potential value of diversity, in order to generate better decisions, products or services, and should be less frequently observed in LA. Therefore:

H5.

Local employees will perceive more DMPs in their work environment intended to reduce discrimination or generate feelings of inclusion, rather than DMPs intended to add value to product or services.

The types of firms that implement DMPs should also be influenced by the institutional scenario. LA markets are composed of both local firms and multinational companies (MNCs). All firms operating in LA are influenced by the local institutional scenario previously described. However, MNCs from North America and Europe are also subject to institutional pressures from their home countries, which possess a longer stance at demanding and complying with DM needs and requirements (Klarsfeld et al., 2016). MNCs home country institutional pressures, both in terms of legal standards and higher levels of professionalisation, will push their managers to deploy DMPs in their LA subsidiaries. Such pressures should influence MNCs to implement DMPs in LA in order to gain legitimacy in their home countries, and simultaneously give them a local reputational advantage. In contrast, local firms are only exposed to the local institutional scenario, with low enforcement of legislation, low professionalisation of DM, and subject to local cultural “blind spots” and prejudice, and will therefore less frequently pursue the implementation of DMPs. Similar patterns, where MNCs play a role of early adopters of DMPs, have been observed in other regions of the world (Klarsfeld et al., 2016). Hence, it is proposed that:

H6.

MNCs will be viewed by local employees as more involved in the development of DMPs when compared to local organisations.

In summary, it is here proposed that the institutional environment will shape employee perceptions about local institutions, as well as their perceptions about firms’ involvement with DMPs, in terms of type of legitimacy pursued, types of DMPs implemented and types of firms engaging in the implementation of DMPs.

Methodology

A three-stage, multi-method research design is used to collect, analyse and validate data, in order to explore DM and test the hypotheses. Study 1 uses focus groups to explore managers’ verbalisations about the DM issues their firms face, the institutional forces affecting DM in their countries, and the DMPs used by their firms. Managers’ opinions are relevant because most frequently they are the ones responsible for implementing human resources practices, such as DMPs (Ostroff and Bowen, 2016). However, subordinates’ perceptions are also important because they are the ones expected to experience the intended practices (Ostroff and Bowen, 2016). Study 2 uses a survey methodology to gain access to subordinates’ opinions about these issues and validate results from Study 1. Finally, Study 3 uses content analysis of firms’ web pages, in order to assess formal, public information about DM provided by firms, and to serve as triangulation method to test the validity of some of the results obtained in Studies 1 and 2.

Study 1

A focus group methodology is used in Study 1 to provide an in-depth exploration of a topic about which little is known (Stewart and Shamdasani, 2014). Focus groups are considered as a less directive research method, where the moderator prompts the main issues to be discussed, and the participants share their attitudes, feelings, beliefs, experiences and reactions in spontaneous conversation (Stewart and Shamdasani, 2014). Focus groups facilitate “collective sense making” where meanings are negotiated and perspectives elaborated through a process of social interaction. This method helps in detecting topics, patterns and details that might not be revealed through more directive methodologies (Stewart and Shamdasani, 2014). Study 1 uses focus groups to explore middle managers’ opinions about the DM and IT issues here pursued.

Sample and procedure

Managers were contacted through executive education programmes, and did not receive any form of retribution for their participation. A total of 107 individuals from Chile, 62 from Guatemala and 88 from El Salvador participated were allocated to 18 groups in Chile, 10 in Guatemala and 11 at El Salvador, respectively. In every country, slightly over a quarter of the participants were females, the average age was close to 33 years, and average work experience approached eight years. In all countries, subjects worked in an equally proportioned mix of medium- and large-size companies. A third of participants worked at MNCs and the rest in locally owned firms. In every country, participants worked in a variety of industries, which were roughly equally balanced: Agriculture and forestry, commerce, communications, transport, construction, electricity, gas and water, manufacturing, mining, business services, financial services and personal services. A moderator encouraged discussion in the groups by prompting three topics in an open questions format: What are the diversity issues organisations face in your country? How does the national context, in terms of legislation, management professionalisation, and work culture influence DM? How do companies respond to these issues? The questions were presented sequentially in time. Groups were encouraged to explore, describe and exemplify the issues that appeared in the discussions, and to build on the ideas of other group members or disagree with them.

Data analysis

The natural language opinions given in these groups were recorded and transcribed into phrases and sentences. Verbalisations were isolated and numbered in order to identify verbal material referring to a specific issue (Stewart and Shamdasani, 2014). Preliminary readings and an initial classification using the scissor-and-sort technique (Stewart and Shamdasani, 2014) led the author to identify three main themes emphasised in participants’ discussions. The first theme centred on the prevalence of discrimination towards different groups of people. The second theme centred on discussing the influence of local culture on DM. The third theme concentrated on the DMPs used by firms to deal with the diversity issues faced. In order to perform a more complete descriptive analysis of participants’ verbalisations, the researcher searched for all possible emergent subordinate categories available in the participants’ verbalisations for each of these three main themes (Stewart and Shamdasani, 2014). For example, for the main theme of “discrimination”, the different groups being discriminated, and the different practices where discrimination occurs (e.g. during personnel selection) were assigned as secondary categories for that theme. For the main theme, influence of local culture on DM secondary categories was created for each culture trait affecting DM discussed by the participants. Finally, for the main theme, “DMPs used by firms”, secondary categories were created for different forms of DMPs discussed by participants. Furthermore, several additional theoretically derived categories were created in order to facilitate hypothesis testing. To test H1 and H2, and following Roberson and Stevens (2006), categories of “positive impact” and “negative impact” were created in order to evaluate the impact attributed to the institutional forces as exerting on DM. To test H3, categories of the three main types of institutional forces that impact DM (legal, professional and cultural) were created (Yang and Konrad, 2011). Following Kostova and Roth (2002), categories of internal and external legitimacy were created to test H4. Kossek and Pichler’s (2006) three categories of types of DMP were created (avoid discrimination, promote the value of diversity, create better products and services), in order to test H5. Finally, to test H6, categories to label companies executing DMPs as local or MNCs were created. See Table I for the final set of categories used in the focus group data analysis.

Two independent research assistants, with degrees in management and work experience in the HR area, were hired and trained to code the data. The use of multiple coders allows the assessment of coding reliability (Stewart and Shamdasani, 2014). Coders received descriptive explanations and samples for each code alternative. Coders independently classified the focus group verbalisations in a multi-step procedure. In a first step, coders assigned the verbal units (phrases, sentences or verbal exchanges) to the three general themes previously identified. Then, the coders used the secondary coding categories to pre-classify transcribed verbalisations and test the coding categories and coding process. Corrections and clarifications were made whenever necessary, and secondary coding categories, not previously identified, were added to the final set of categories (Stewart and Shamdasani, 2014). In a final coding stage, coders were requested to independently examine the data and categorise verbal material with the assigned code categories. Inter-rater agreement for all coded data between coders reached 94 per cent. Descriptive results of focus groups verbalisations are presented next, followed by statistical tests for the proposed hypotheses.

Study 1 results

Descriptive results

Table II shows two indexes to analyse the focus groups’ verbalisations. Frequencies of verbalisations (fV) reflect how often a descriptor was mentioned by participants within each country. A second index, intended to analyse the reliability of verbalisations of specific topics across focus groups, is the percentage of focus groups in each country (%G) where such verbalisations were at least mentioned once. The most prevalent DM issue that surfaced in focus groups discussions is discrimination. The upper part of Table II shows frequencies of verbalisations, indicating that in Chile, discrimination based on gender and social class is among the most frequently mentioned. In El Salvador, gender, age and disabilities, and in Guatemala, race and gender discrimination are most frequently mentioned. Other forms of discrimination are less frequently mentioned, or receive a moderate rate of responses. In all three countries, the human resources practices which participants most frequently report as being biased by discrimination are personnel selection, promotions and compensations (see the mid-upper part of Table II). An example of a participant’s verbalisation in Chile describes how discrimination occurs: “[…] There is discrimination in hiring and promotions towards lower social classes […] this is a less conscious, informal process […] selection criteria are not made explicit […] sometimes it is the high school where they studied, their neighbourhood of residence […] or how the person speaks or dresses”.

A second issue, discussed in the focus groups, is the influence of institutional forces on DM. Most mentions of institutional forces referred to culture (65 per cent), and smaller percentages referred to legislation (25 per cent) and management professionalisation or educational processes (10 per cent). The middle part of Table II shows the cultural issues most frequently associated with DM by participants. Classism is discussed in 83 per cent of the Chilean groups, machismo is discussed in 67 per cent of the groups in El Salvador and 50 per cent of those in Guatemala; finally, racism was brought up in 70 per cent of the groups in Guatemala. These same issues were also raised in the other countries, but discussed in less than 50 per cent of the groups. Examples of participants’ verbalisations illustrate these culture traits “[…] in Chile we have more labels to classify each other in social classes than Eskimos have words to classify types of snow”. A female participant from El Salvador describes Machismo as “[…] a prejudice that we should stay at home to care for the family […] or that we can only do humanitarian work […]”. Other cultural issues, such as authoritarianism and paternalism, were discussed in a smaller proportion of groups. Paternalism and Authoritarianism are seen as two sides of the same coin. “If you are lucky […] you get a nice boss who treats you well […] If you have bad lack, you get a boss that can be inconsiderate and abusive […]”. On the whole, cultural traits were mostly described as precursors of prejudice and discrimination.

A third issue, raised in focus groups, refers to the DMPs that participants had observed or experienced at their workplace (see the lower part of Table II). A first set of verbalisations refers to the use of DMPs intended to inhibit discrimination. Interventions to avoid biases in personnel selection, and to facilitate access to technical training to underrepresented groups of employees, are most frequently mentioned across countries. An example of a practice used to inhibit discrimination in personnel selection is described by a participant in Chile “[…] in an attempt to set the company apart from a common practice […] we have established a rule not to request photos in candidates’ CVs, in order to avoid prejudice attached to their physical appearance”.

A second set of DMPs discussed by participants referred to interventions geared towards promoting the value of human diversity, and generating feelings of inclusion. Diversity training is most frequently mentioned, while other practices, such as infrastructure for the disabled, inclusion activities and ethics codes, are less frequently mentioned. An example of a DMP promoting the value of diversity is offered by a participant from El Salvador, expressing that at her company a campaign was held “[…] to promote a culture of diversity and inclusion, and respect for all, regardless of their beliefs or ideology. The campaign ‘Watch Your Fingers’ was launched to promote respect for differences, emphasising that we need all our fingers and that no one finger is the same as another”. In another inclusion DMP example, a participant from Guatemala describes “[…] when we plan for the construction of a new facility, we usually include diverse groups in the decision making. This has led us to include ramps, and elevators with Braille labels on the button panels […] we have also agreed to have no meetings in the late afternoon, so that we don’t affect employees who car pull, or have to pick up their children at day-care”.

Very rarely participants described DMPs intended to take advantage of business opportunities through human diversity. One example came from Guatemala, where a telecommunications company had hired and trained sales employees from several ethnic minorities. A participant described that “[…] people feel more confident when talking to someone of their own ethnic group, so that our strategy has been to hire staff from local ethnic groups to conduct sales activities in different regions of the country […] also advertising is transmitted in the dominant language of the region”. Another participant from Guatemala reported to have hired staff that speaks the Mayan language at the company’s call centre, because “[…] our Mayan clients have difficulty expressing their concerns or complaints in Spanish”.

Hypotheses testing from focus group verbalisations

All hypotheses were tested with χ2 statistics by comparing frequencies of verbal responses in relevant categories. When discussing the institutional forces having an impact on DM, most mentions to culture depicted a negative impact (vs positive) of local culture on DM (χ2 (1, n=125)=81.61, p<0.01), while most mentions of legislation (χ2 (1, n=34)=30.12, p<0.01) were associated with a positive impact of these forces on DM, thus supporting H1 and H2. Verbalisations about DM professionalisation or educational processes were considerably few (n=19); however, all of them were associated with a positive impact of normative forces on DM. Furthermore, more verbalisations are made in focus groups about legal (χ2 (1, n=80)=22.05, p<0.001) and cultural institution (χ2 (1, n=172)=104.39, p<0.001), rather than normative/professional institutions, giving support to H3. Participants’ verbalisations also suggest that firms are perceived as intending to gain external rather than internal legitimacy, when implementing DMPs (χ2=5.23, df 1, p<0.05), supporting H4. Furthermore, focus group verbalisations indicate that firms are perceived as implementing a larger proportion of DMPs geared towards reducing discrimination 2=22.23, df 1, p<0.01), and generating feelings of inclusion (χ2=37.16, df 1, p<0.01), rather than towards generating better products or services. This supports H5. Finally, MNCs are viewed as being more involved in DMP implementation, when compared to local firms (χ2=37.39, df 1, p<0.01), supporting H6.

Study 2

The focus groups methodology can be subject to method-induced biases that posit potential limitations on its results’ validity. For example, if questions are asked in a leading manner, if an extremely dominant participant inhibits other group members from sharing their opinions, or if the interaction of the group creates pressures to conform to a particular view (Vogt et al., 2004). Triangulation methods help to enhance the validity of inferences and deal with such methodological problems (Vogt et al., 2004). Triangulation occurs when comparing one set of results with another set, acquired with a different methodology or sources of information. Study 2 uses a survey methodology to validate descriptive results from Study 1, as well as to validate hypotheses related to the purposes and types of DMPs implemented (H4 and H5), and the types of firms implementing DMPs (H6).

Sample and procedure

A sample of workers from Chile (n=385), El Salvador (n=214), and Guatemala (n=192), half females and half males, working in a wide variety of sectors, answered a survey developed on the basis of Study 1 results, and previous literature. A convenience sampling was carried out using a snowball procedure through managers attending executive education programmes, who provided contacts to invite voluntary participants. Respondents worked at firms with a similar industry composition as in Study 1. No more than 3 subjects per firm were allowed in the sample to avoid in-company nested effects. Participants were asked to judge the frequency at which the discriminatory practices happen in their workplace, the human resources processes where discrimination occurs, the prevalence of cultural traits that facilitate discrimination, and the degree of use of several DMPs, using the verbal categories identified in Study 1 and in previous literature. Six additional items were included to test whether firms use DMPs to gain external and/or internal legitimacy. Based on Kostova and Roth (2002), these items describe the achievement of external or internal legitimacy by DMPs. Four additional items were included, in order to assess the implementation of DMPs using Kossek and Pichler’s (2006) DMP categories (reducing discrimination, generating feelings of inclusion, generating better products or services). Finally, two items asked survey respondents to rate the frequency of DMP implementation by MNCs and local firms. A six-point Likert scale with response categories ranging from “Extremely Infrequent” to “Extremely Frequent” was used.

Results

Table III shows accumulated frequencies of survey responses for the “extremely infrequent”, “very infrequent” and “infrequent” categories (1, 2, and 3), and for the “frequent”, “very frequent” and “extremely frequent” categories (4, 5, and 6), segmented by country. Table III also shows χ2 test results for differences between the “Low frequency” and “High frequency” accumulated response sets. In consistency with the results obtained in Study 1, Table III shows significant χ2 scores, indicating that a majority of respondents believe that discrimination occurs with a high frequency in relation to gender, race, social class, physical appearance, disabilities, sexual orientation and age. The only exception to the above is observed in El Salvador, where the majority of respondents consider Race discrimination is infrequent. However, in all three countries, the majority of respondents believe that discrimination at the workplace based on religion or nationality occurs with low frequency. Table III also shows that, in all three countries, the majority of respondents consider that discrimination is highly frequent during personnel selection, promotion and compensation processes. In Chile and El Salvador, the majority of respondents consider that discrimination during performance evaluations and access to training is infrequent, while in Guatemala, the survey respondents were split in their opinions.

Furthermore, in all three countries, the majority of respondents consider that classism, machismo, paternalism, authoritarianism and racism are very frequently seen cultural biases and prejudices in their countries’ workplaces. The only exception is observed in El Salvador, where the majority of respondents consider racism as infrequent (see Table III).

Responses to survey items asking participants to rate the prevalence of DMPs in their countries show that roughly half of respondents consider that DMPs intended to avoid discrimination (interventions in personnel selection, fair promotions, access to technical training, fair performance evaluations and compensation) are implemented very frequently, and the other half consider that they are implemented infrequently. This was seen in all three countries, with the exception of personnel selection and access to promotions practices, whereby in Chile, the majority of respondents consider them to be implemented infrequently (see Table III). For the diversity training, infrastructure for disabled, and ethics code DMPs, intended to promote a positive climate of diversity, either the majority of respondents consider them to occur with low frequency, or there are similar numbers of respondents choosing the frequent and infrequent response alternatives. However, in all three countries, the majority of respondents consider that Inclusion activities, for example, celebrations of human diversity, occur frequently. Furthermore, in all three countries, the DMPs intended to create value in decisionmaking, products or services (decision-making diverse teams, product/service diverse teams, knowledge gaining diverse teams) are considered to be used infrequently by the majority of respondents, in all three countries (see Table III).

Participants’ responses rating the extent to which DMPs are implemented to gain external/internal legitimacy indicate that external legitimacy intentions (comply with laws and regulations, Strengthen firms’ image in community, and Promote firm’s social responsibility) are perceived as frequent by the majority of respondents, in all three countries. On the other hand, internal legitimacy intentions (strengthen fair treatment to employees, gain internal support from employees, create a better work climate) are perceived as frequent by the majority of respondents in Chile, infrequent by respondents in Guatemala, while in El Salvador 50 per cent of respondents’ opinions correspond to “frequently” and the other 50 per cent to “infrequently” (see Table III).

In order to test H4H6, indexes were constructed, by averaging the “external legitimacy”, “internal legitimacy”, “DMPs intended to reduce discrimination”, “DMPs intended to generate feelings of inclusion” and “DMPs intended to generate better products or services” items. Table IV shows means and standard deviations for these indexes. Then t-tests were used to compare the means of these indexes, in order to test H4H6, for the overall sample across countries. Results indicate that DMPs are perceived as more frequently used by organisations as intended to gain external rather than internal legitimacy (t=15.19, df=1,508, p<0.001), confirming H4. Results also show that DMPs are perceived to be used with the intention to reduce discrimination (t=10.31, df=707, p<0.001) and generate feelings of inclusion (t=12.11, df=700, p<0.001), rather than to generate better products or services, confirming H5. Finally, survey respondents consider that MNCs use DMPs more frequently than local firms (t=25.76; df=708; p<0.001), confirming H6. When these hypotheses are tested at the single-country data level, results remain significant at a p<0.001 level.

Study 3

Purpose

Study 3 adds a relevant third perspective to the implementation of DMPS in the three countries under study, by pursuing the analysis of firms’ web pages, regarding their involvement in implementing DMPs. This approach allows the assessment of firms’ public information about their involvement in DM, providing a formal organisational stance about DM, in contraposition to Studies 1 and 2, which assessed managers and subordinates’ experiences and perceptions. Previous research conducted by Uysal (2013) used a similar methodology, in a sample of US firms. Uysal’s (2013) findings suggest that firms’ DM communication on their websites is intended to position DM as a competitive advantage, and rarely as a corporate social responsibility, or a corporate value.

The type of content provided by firms through websites gives formal and public information about the types of DMPs used, and the types of firms using DMPs. Therefore, webpage content analysis allows to validate previous results from Studies 1 and 2, in relation to H5, which tests for types of DMPs (avoid discrimination, create a positive diversity climate, create value through products and services), and H6, which tests for types of firms using DMPs (local vs MNCs).

Sample and procedure

The content of web pages was analysed for references regarding the DMPs implemented. A total of 337 web pages belonging in similar proportions to local firms and MNCs in each country (see Table V) were analysed by two independent coders. Local stock markets, trade associations and company directories served to assemble the samples. In every country, for both local firms and MNCs, a variety of industry sectors were balanced in similar proportions as those used in Studies 1 and 2. All material posted on the websites describing the use of DMPs was downloaded and the coders categorised this material with the code categories reported in Table I, for H5 and H6 (coders inter-rater agreement t=94 per cent).

Results

Table V shows frequencies and percentages, of firms describing DMPs in use on their web pages, for each country. Unequal variance t-test comparisons, between local and MNC firms, were performed for the total sample, in order to compare local firms and MNCs’ declaration of DMP use on their web pages, the purpose intended for the use of DMPs, and the types of DMPs used. Overall, the countries’ comparison results indicate that 15 per cent of local firms and 40 per cent of MNCs report using DMPs on their web pages, making for a difference, which is statistically significant (t=−5,31; df=302; p<0.000). Also, among those organisations using DMPs, MNCs posted on their web pages a larger number of DMPs per firm (t=−2.32; df=74; p<0.01), with an overall average of 3.35 DMPs in MNCs mentioned web pages, and an average of 2.46 DMPs mentioned by local firms. These results help validate focus group verbalisations, suggesting that MNCs use DMPs more frequently than local firms, supporting H6. Finally, web page content analyses also indicate that firms in the sample report more DMPs used as intending to reduce discrimination (t=18.97, df=95, p<0.001) and generating feelings of inclusion (t=12.83, df=106, p<0.001) rather than to generate better products or services, supporting H5. In data analyses not reported here, H5 and H6 are also confirmed when tested in each of the countries’ data separately.

General discussion

Comparing descriptive results in Studies 1 and 2

Studies 1 and 2 offer views, about the DM environment in the three countries, from different types of employees. Study 1 explores managers’ opinions, which are relevant because of their role in implementing firms’ DMPs, while Study 2, designed to validate Study 1 results, analyses the views of employees, who are the primary receptors of the DM initiatives that firms implement (Ostroff and Bowen, 2016). Results from both studies present similarities and differences. The results from both studies show that discrimination in terms of gender, disabilities and social class is perceived consistently across countries. On the other hand, race discrimination seems to be perceived mostly in Guatemala and Chile, but not in El Salvador. Additionally, both studies in all three countries suggest that perceptions of work discrimination based on religion or national origin are much less frequent. However, Studies 1 and 2 differ in their results regarding the prevalence of discrimination based on physical appearance, sexual orientation and age. For subordinates (Study 2), such forms of discrimination are frequent, while managers in the focus groups for Study 1 mentioned them less frequently. The research design used in this study does not clarify whether this difference is due to the methodology or response groups. Furthermore, Studies 1 and 2 also show agreement in the cultural traits perceived as affecting DM. Managers and subordinates’ perceptions seem to agree that classism, machismo, paternalism and authoritarianism are frequently encountered in all three countries. However, racism is more prevalent only in Chile and Guatemala, and not in El Salvador, as both Studies 1 and 2 show.

Theoretical contributions

Theory-driven categories were built to analyse data from all three studies in order to empirically test several hypotheses about the influence of the local institutional scenario on DM. Results from Study 1 suggest that institutional forces do have an impact on employees’ perceptions about DM, validating the IT proposition that managers’ perceptions about institutions are influenced by their institutional scenario (H1H3). Furthermore, both studies validate through managers and subordinates’ perceptions that firms tend to concentrate on DMPs intended to gain external legitimacy (H4), providing additional support for IT.

H5 and H6 are tested through managers and subordinates’ perceptions, as well as through content analysis of firms’ web pages’ public information about their DM policies and activities. In all three studies, and in the three countries, results indicate that the DMPs most frequently used focus on fighting discrimination and facilitate inclusion. H6 is supported in all three studies: MNCs use DMPs more frequently than local firms. The frequent use of DMPs intended to fight discrimination and facilitate inclusion, considered in conjunction with the tendency to search for internal legitimacy, shows that firms in these countries follow a pattern of implementation of DMPs, which is similar to the one described in the USA during the early stages of organisational responses to affirmative action laws, when companies needed to comply with anti-discrimination laws, and enforcement of legislation was weak (Kelly and Dobbin, 1998). In summary, the confirmation of H1H6 supports IT propositions about the role of the institutional environment shaping organisational forms and responses (Scott, 1995), and in particular shaping firms DMP implementation (Yang and Konrad, 2011).

These studies findings also have implications for resource-based theory (RBT). Mayer and Sparrowe (2013) proposed that theory integration can be pursued by specifying the impact of different explanatory variables, arising from different theories, to the same dependent variable, leading to a more comprehensive understanding of the phenomenon (Mayer and Sparrowe, 2013, p. 917). In particular, Yang and Konrad (2011) proposed that IT can assist RBT in explaining firms’ engagement in DM. RBT focusses on how managers, through rational analyses, attempt to configure and develop their firms’ resources (e.g. DMPs) as different and unique from their competition, in order to achieve competitive advantages. However, RBT has been criticised for its’ silence about the effects of a firm’s external environment, including the institutional environment, on managing resources (Yang and Konrad, 2011). Yang and Konrad (2011, p. 26) also proposed that a contribution to better understand firms’ implementation of DMPs can be made by analysing the institutional environment antecedents of such practices. This paper’s results suggest that DMPs implemented by organisations are in part explained by the institutional pressures they receive, for example, through local legislation, home country legislation for MNCs, cultural traits that undervalue human diversity, and possibly low levels of professionalisation in DM. However, managers’ rationality can lead firms to pursue DMPs and differentiate from their competition, for example, in order to become a preferred employer, as was illustrated through some verbalisations in Study 1’s qualitative results. Therefore, this study suggests that heterogeneity in investment in DMPs might in part depend on the different pressures firms receive from environmental institutions, as proposed by IT, and in part on managers’ rational decision making, as proposed by RBT, giving support to Yang and Konrand’s (2011) proposition that IT complements RBT for a better understanding of DMP implementation.

Limitations

Several limitations of these studies must be acknowledged. First, the convenience samples of the studies limit the generalisability of their results, which cannot be extended to other organisations or employees within these countries. Another potential limitation is the reliability and validity of the coding procedures. Coding categories were developed from Study 1’s verbal responses and theory, coders worked independently, and systematic biases in coding were detected and clarified. However, it is not possible to discard that other measurement errors may affect these results. Finally, alternative explanations are possible for some of the results found. For example, the higher usage of DMPs by MNCs, in comparison to local firms, could be in part due to the size effect. MNCs by definition operate with employees from different countries, which might make it more necessary for MNCs to use DMPs. Also, the size of MNCs might reach a critical point, where investing in DMPs renders higher returns rates.

Practical implications

These studies suggest the need for top management and human resources units to acknowledge the relevance of DM in these countries. Participants had a critical opinion of the current status of DM at the workplace in their countries. Both local firms and MNCs can reap the benefits of developing policies and practices that help avoid discrimination and unfair treatment of employees and candidates, and as such become preferred places of employment. The potential to use DM for value generation through new products and processes seems to be a field for competitive advantage, which is underexploited in LA. Engaging in DM in LA will provide early adopters with potential competitive advantages, for example, by means of becoming preferred employers and attracting better quality human capital.

Lack of compliance with legal systems, scarcity of managerial skills to handle diversity issues, or cultural biases that can make human diversity less valued, or socially acceptable to manage, can make it difficult for managers to benefit from the advantages, or eliminate the disadvantages, of a diverse workforce. The results from the studies here presented suggest that human resources units, from firms intending to implement DMPs in these countries, might need to begin by addressing cognitive–emotional biases of social stereotyping and prejudice, which might be shared by some groups within a firm. As a matter of fact, Stevens et al. (2008) noted that the simple lack of attention paid by management to human differences can facilitate the emergence of unfair practices towards diversity. Furthermore, previous research suggests that the institutional environment can bias managers’ perceptions creating in them “blind spots” towards different aspects of their firms’ competitive environment, and that these biases hinder sound rational decisions (Zajac and Bazerman, 1991). In particular, the cognitive–emotional biases of social stereotyping and prejudice can be culturally transmitted, and act as an automatic response from which the decision maker is unconscious (Fiske, 2000). Therefore, stereotyping and prejudice might obstruct managers’ decisions in terms of DM, for example, by excluding highly competent individuals, or undermining the work climate. Because of the latter, one of the first issues that human resources units might need to address when intending to implement a DM programme are any potential cognitive–emotional biases of social stereotyping and prejudice that might hinder the capacity of different groups to properly assess and engage in DM work.

Coding categories for focus group verbalisations

Code categories Coding alternatives
Coding for general descriptive purposes (derived from focus group verbalisations initial analyses)
Types of discrimination Gender; disability; ethnicity; social class; age; physical appearance; nationality; sexual orientation; religion; political affinity; other
Types of human resources management practice involved in discrimination Selection; compensation; training; promotions; dismissals; performance evaluation; other (job stability; participation; biased job descriptions)
DMPs used to deal with diversity issues Human resources management practices to avoid discrimination (classified as any of the above); inclusion activities (celebrations/after-work activities); infrastructure accommodations; code of ethics; diversity training; diversity teams for decision making or innovation
National culture trait affecting DM Classism; paternalism; racism; authoritarianism; machismo; other
Coding for testing hypotheses (derived from literature)
H1 and H2: types of impact attributed to the institutional forces exert on diversity management (Roberson and Stevens, 2006) Positive impact; negative impact
H3: types of institutional forces that impact diversity management (Yang and Konrad, 2011) Legal and regulatory system
Professional and education processes
Cultural, ethical and social habits
H4: type of legitimacy firms’ attempts to gain through DMPs (Kostova and Roth, 2002) External legitimacy; internal legitimacy
H5: purpose intended in the use of the DMPs (Kossek and Pichler, 2006) Practices intended to avoid discriminatory practices; practices intended to facilitate a sense of inclusion; practices intended to add value to products or services
H6: types of companies executing DMPs Local firms; MNCs

Within country frequencies of verbalisations across focus groups (fV), within country percentage of focus groups where such verbalisations were mentioned (%G), and coders inter-rater agreement for data categories (I-RA)

Chile El Salvador Guatemala
fV %G fV %G fV %G
Groups towards which discrimination occurs (coders I-RA: 94%)
Gender 38 83 35 92 20 80
Social class 37 83 15 50 11 80
Ethnicity 14 50 3 17 29 90
Physical appearance 17 56 11 42 7 50
Disability 15 61 14 75 11 70
Age 12 44 27 83 8 50
Political affinity 14 28 7 50 1 10
Sexual orientation 27 56 8 50 3 20
Religion 14 22 5 33 3 30
Nationality 15 61 1 8 1 10
Human resources practices where discrimination occurs (coders I-RA: 93%)
Selection 60 78 64 100 28 50
Promotions 32 56 31 75 13 40
Compensation 7 28 11 33 3 20
Training 1 6 0 0 0 0
Dismissals 1 6 5 8 0 0
Cultural traits that facilitate discrimination (coders I-RA: 95%)
Classism 33 83 8 33 2 30
Racism 5 22 1 8 13 70
Machismo 1 6 9 67 5 50
Authoritarianism 14 44 3 17 4 30
Paternalism 10 44 1 8 1 10
DMPs implemented to avoid discrimination (coders I-RA: 95%)
Interventions in personnel selection 37 83 30 83 20 80
Access to technical training 7 33 15 58 12 60
Access to promotions 1 6 6 33 2 20
Compensation 3 17 4 33 1 10
Dismissals 0 0 1 8 0 0
DMPs intended to promote the value of diversity and create a positive diversity climate (coders I-RA: 95%)
Diversity training 19 50 25 75 8 50
Infrastructure for disabled 2 11 4 33 3 30
Inclusion activities 13 44 1 8 2 20
Ethics code 1 6 0 0 0 0
DMPs intended to create value through products and services (coders I-RA: 94%)
Decision-making diverse teams 4 17 1 8 0 0
Product/service diverse teams 4 11 3 17 1 10
Knowledge gaining diverse teams 1 6 0 8 1 10

Accumulated frequencies of participants’ responses in the Likert scale for all the “low frequency” response categories f(1, 2, 3), and for all the “high frequency” response categories f(4, 5, 6), for Study 2 survey items, and χ2 test results for differences between the “Low frequency” and “High frequency” response accumulations

Chile (n=385) El Salvador (n=214) Guatemala (n=196)
f(1,2,3) f(4,5,6) χ2 f(1,2,3) f(4,5,6) χ2 f(1,2,3) f(4,5,6) χ2
Groups towards which discrimination occurs
Gender 142 220 16.8** 92 72 4.2* 78 114 6.8*
Race 143 219 16.0** 168 46 69.6** 57 135 31.7**
Social class 91 271 89.5** 85 129 9.1** 77 115 7.5**
Physical appearance 121 241 39.8** 85 129 9.1** 80 112 5.3*
Disabilities 114 246 48.4** 82 132 11.7** 58 134 30.8**
Sexual orientation 131 230 27.2** 88 125 6.4* 73 119 11.0**
Age 151 210 9.6** 82 132 11.7** 75 117 9.2**
Religion 290 72 131.3** 180 34 99.6** 143 49 273.1**
Nationality 191 169 ns 178 36 94.2** 154 38 102.4**
HR processes where discrimination occurs
Selection 93 268 84.8** 69 145 27.0** 55 137 35.0**
Promotions 140 214 15.5** 91 123 4.8* 75 117 9.2**
Access to training 224 136 21.5** 133 81 12.6** 98 94 ns
Performance evaluations 214 147 12.4** 142 72 22.9** 103 89 ns
Compensation 131 230 27.2** 98 116 1.5ns 78 114 6.8*
Cultural traits that facilitate discrimination
Classism 48 308 189.9** 66 148 31.4** 44 148 56.3**
Machismo 77 279 114.6** 53 161 54.5** 45 147 54.2**
Paternalism 81 275 105.7** 57 157 46.7** 37 155 72.5**
Authoritarianism 78 178 112.4** 67 146 29.3** 35 157 77.5**
Racism 131 225 24.8** 155 59 43.1** 66 126 18.8**
DMPs implemented to avoid discrimination
Interventions in personnel selection 202 159 5.2* 103 111 ns 93 99 ns
Access to promotions 205 155 6.9** 94 119 ns 100 92 ns
Access to technical training 190 171 ns 111 103 ns 92 100 ns
Performance evaluations 170 190 ns 98 115 ns 84 108 ns
Compensation 179 181 ns 102 110 ns 85 107 ns
DMPs implemented to promote value of diversity and create a positive diversity climate
Diversity training 239 120 39.5** 123 88 5.8* 107 85 ns
Infrastructure for disabled 205 155 6.9** 115 99 ns 143 49 46.0**
Inclusion activities 126 233 31.8** 66 148 31.4** 81 111 4.7*
Ethics code 209 151 9.3** 84 130 9.9** 97 95 ns
DMPs implemented to create value in decision making, products or services
Decision-making diverse teams 262 98 75.5** 150 61 37.5** 135 57 31.7**
Product/service diverse teams 227 134 24.0** 149 62 35.9** 140 52 40.3**
Knowledge gaining diverse teams 263 98 75.42** 153 61 39.6** 142 50 44.1**
Extent to which DMPs are implemented to gain external legitimacy
Comply with laws/regulations 41 321 216.5** 72 142 22.9** 57 135 31.7**
Strengthen firms’ image in community 41 321 216.5** 59 155 43.1** 39 153 67.7**
Promote firm’s social responsibility 79 283 114.8** 56 158 48.6** 63 129 22.7**
Extent to which DMPs are implemented to gain internal legitimacy
Strengthen fair treatment to employees 111 251 54.1** 97 117 ns 101 91 ns
Gain internal support from employees 160 202 4.9** 108 106 ns 121 71 13.0**
Create a better work climate 104 258 65.5** 112 102 ns 116 75 8.8**

Notes: *p<0.05; **=p<0.01

Means and standard deviations for Study 2 indexes for hypotheses testing

Chile (n=385) El Salvador (n=214) Guatemala (n=196)
Mean SD Mean SD Mean SD
Extent to which DMPs are implemented to gain
External legitimacy 4.44 0.47 4.15 0.86 4.18 0.79
Internal legitimacy 3.86 0.65 3.52 0.91 3.26 1.03
Types of DMPs implemented
DMPs intended to avoid discrimination 3.44 0.98 3.61 1.08 3.54 0.85
DMPs intended to create a positive diversity climate 3.35 1.00 3.60 1.28 3.27 0.84
DMPs intended to create value in decision making, products or services 2.80 1.08 2.88 1.18 2.76 1.20
Extent to which types of firms use DMPs
Local companies 3.29 1.16 2.96 2.19 2.73 1.58
Multinational companies 4.84 1.00 4.20 1.66 4.21 1.46

Frequencies, and percentages (in parentheses), of local firms and MNCs, describing through their web pages DMPs in use, at each country

Chile El Salvador Guatemala
Sample size Local (n=51) MNCs (n=56) Local (n=55) MNCs (n=54) Local (n=63) MNCs (n=68)
DMPs implemented to avoid discrimination
Interventions in personnel selection 7 (14%) 11 (20%) 0 (0%) 8 (15%) 1 (2%) 11 (19%)
Access to technical training 6 (12%) 15 (27%) 2 (0%) 8 (15%) 3 (5%) 7 (12%)
Performance evaluation 3 (6%) 8 (14%) 0 (0%) 2 (4%) 1 (2%) 4 (7%)
Access to promotions 3 (6%) 6 (11%) 1 (0%) 5 (9%) 1 (2%) 6 (10%)
Compensation 0 (0%) 7 (13%) 0 (0%) 5 (9%) 3 (5%) 5 (9%)
Dismissals 1 (2%) 4 (7%) 0 (0%) 3 (6%) 0 (0%) 4 (7%)
DMPs intended to promote the value of diversity and create a positive diversity climate
Diversity training 0 (0%) 1 (2%) 0 (0%) 2 (4%) 2 (3%) 6 (10%)
Infrastructure for disabled 1 (2%) 3 (5%) 0 (0%) 0 (0%) 0 (0%) 3 (5%)
Inclusion activities 9 (18%) 17 (18%) 4 (7%) 11 (20%) 0 (7%) 15 (25%)
Ethics code 7 (14%) 17 (30%) 4 (7%) 7 (13%) 0 (0%) 12 (21%)
DMPs intended to create value through products and services
Decision-making diverse teams 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%)
Product/service diverse teams 0 (0%) 2 (4%) 0 (0%) 2 (4%) 0 (0%) 2 (3%)
Knowledge gaining diverse teams 0 (0%) 0 (0%) 0 (0%) 2 (4%) 0 (0%) 0 (0%)

References

Alhejji, H., Garavan, T., Carbery, R., O’Brien, F. and McGuire, D. (2016), “Diversity training programme outcomes: a systematic review”, Human Resource Development Quarterly, Vol. 27 No. 1, pp. 95-149.

Atal, J., Ñopo, H. and Winder, N. (2009), “New century, old disparities: gender and ethnic wage gaps in Latin America”, Working Paper, No. 25, Inter-American Development Bank, SSRN, available at: https://ssrn.com/abstract=1815933; http://dx.doi.org/10.2139/ssrn.1815933

Bensusán, G. (2009), “Legislation and labor policy in latin America: crisis, renovation, or restoration?”, Comparative Labor Law & Policy Journal, Vol. 34 No. 1, pp. 655-676.

Bravo, D., Sanhueza, C. and Urzúa, S. (2009), “Using an experimental approach to identify labor market discrimination based on gender and social class in a developing economy”, unpublished manuscript, Universidad de Chile, Departamento de Economía, Santiago.

Burgess, K. (2010), “Global pressures, national policies, and labor rights in Latin America”, Studies in Comparative Development, Vol. 45 No. 2, pp. 198-224.

Cárdenas, M.C., Eagly, A., Salgado, E., Goode, W., Heller, L.I., Jaúregui, K., Quirós, N.G., Galarza, N., Gormaz, N., Bunse, S., Godoy, M.J., Rocha, T.E., Navarro, M., Sosa, F., Aguilera, Y., Schulmeyer, M., Tanure, B., Naranjo, M., Soto, B.H., Darre, S. and Tunqui, R.C. (2014), “Latin American female business executives: an interesting surprise”, Gender in Management: An International Journal, Vol. 29 No. 1, pp. 2-24.

Chiappetta, C.J., Serotini, F., Caldeira, J.H., Martínez, J.C. and Gomes, R.A. (2011), “Diversity management: challenges, benefits, and the role of human resource management in Brazilian organizations”, Equality, Diversity and Inclusion: An International Journal, Vol. 30 No. 1, pp. 58-74.

Donnelly, R. (2015), “Tensions and challenges in the management of diversity and inclusion in IT services multinationals in India”, Human Resource Management, Vol. 54 No. 2, pp. 199-215.

Dulitzky, A.E. (2005), “A region in denial: racial discrimination and racism in Latin America”, in Dzidzienyo, A. and Oboler, S. (Eds), Neither Enemies Nor Friends: Latinos, Blacks, Afro-Latinos, Vol. 3, Palgrave Macmillan, New York, NY, pp. 93-113.

Elvira, M.M. and Davila, A. (2005), “Emergent directions for human resource management research in Latin America”, The International Journal of Human Resource Management, Vol. 16 No. 12, pp. 2265-2282.

Fiske, S.T. (2000), “Stereotyping, prejudice, and discrimination at the seam between the centuries: evolution, culture, mind, and brain”, European Journal of Social Psychology, Vol. 30 No. 3, pp. 299-322.

Galarza, F.B. and Yamada, G. (2012), “Labor market discrimination in Lima, Peru: evidence from a field experiment”, Unpublished Manuscript No. 12-03, Universidad del Pacífico, Department of Economics and Centro de Investigación (CIUP), Lima.

Gibbons, J.L. and Ashdown, B.K. (2010), “Ethnic identification, attitudes, and group relations in Guatemala”, Psychology, No. 1, pp. 116-127.

Goldín, L. and Dowdall, C. (2012), “The rule of law and the enforcement of the law workers’ understanding of labor rights in the central highlands of Guatemala”, Latin American Perspectives, Vol. 39 No. 6, pp. 133-154.

Guzmán, J.M., Rodríguez, J., Martínez, J., Contreras, J.M. and González, D. (2006), “The demography of Latin America and the Caribbean since 1950”, Population, Vol. 61 No. 5, pp. 519-620.

Hofstede, G. (1980), Culture’s Consequences: International Differences in Work-Related Values, Vol. 5, Sage Publications, Beverly Hills, CA.

House, R.J., Hanges, P.J., Javidan, M., Dorfman, P.W. and Gupta, V. (2004), Culture, Leadership, and Organizations: The GLOBE Study of 62 Cultures, Sage, Thousand Oaks, CA.

Javidan, M., Dorfman, P.W., De Luque, M.S. and House, R.J. (2006), “In the eye of the beholder: cross cultural lessons in leadership from project GLOBE”, The Academy of Management Perspectives, Vol. 20 No. 1, pp. 67-90.

Kelly, E. and Dobbin, F. (1998), “How affirmative action became diversity management employer response to antidiscrimination law, 1961 to 1996”, American Behavioral Scientist, Vol. 41 No. 7, pp. 960-984.

Klarsfeld, A., Ng, E.S., Booysen, L., Castro Christiansen, L. and Kuvaas, B. (2016), “Comparative equality and diversity: main findings and research gaps”, Cross Cultural & Strategic Management, Vol. 23 No. 3, pp. 394-412.

Kossek, E.E. and Pichler, S. (2006), “EEO and the management of diversity”, in Boxell, P., Purcell, J. and Wright, P.M. (Eds), Handbook of Human Resource Management, Oxford University Press, Oxford, pp. 251-272.

Kostova, T. and Roth, K. (2002), “Adoption of an organizational practice by subsidiaries of multinational corporations: institutional and relational effects”, Academy of Management Journal, Vol. 45 No. 1, pp. 215-233.

Levitsky, S. and Murillo, M.V. (2013), “Building institutions on weak foundations”, Journal of Democracy, Vol. 24 No. 2, pp. 93-107.

Maxfield, S. (2005), “Modifying best practices in women’s advancement for the Latin American context”, Women in Management Review, Vol. 20 No. 4, pp. 249-261.

Mayer, J.K. and Sparrowe, T.R. (2013), “From the editors: integrating theories in AMJ articles”, Academy of Management Journal, Vol. 56 No. 4, pp. 917-922.

Mazzolini, R.C. (2007), “Las Castas: interracial crossing and social structure, 1770-1835”, in Müller-Wille, S. and Rheinberger, H.J. (Eds), Heredity Produced: At the Crossroads of Biology, Politics, and Culture, MIT Press, Cambridge, MA, pp. 1500-1870.

Moreno, M. and Oropesa, R.S. (2012), “Ethno-racial identification in urban Peru”, Ethnic and Racial Studies, Vol. 35 No. 7, pp. 1220-1247.

Moreno, M., Ñopo, H., Saavedra, J. and Torero, M. (2012), “Detecting gender and racial discrimination in hiring through monitoring intermediation services: the case of selected occupations in Metropolitan Lima, Peru”, World Development, Vol. 40 No. 2, pp. 315-328.

Ng, E.S. and Stephenson, J. (2015), “Individuals, teams, and organizational benefits of managing diversity”, in Bendl, R., Bleijenbergh, I., Henttonen, E. and Mills, A.J. (Eds), The Oxford Handbook of Diversity in Organizations, Oxford University Press, Oxford, pp. 235-254.

Nuñez, J. and Gutiérrez, R. (2004), “Class discrimination and meritocracy in the labor market: the case of Chile”, Estudios de Economía, Vol. 31 No. 2, pp. 113-132.

OECD (2016), OECD Employment Outlook 2016, OECD Publishing, Paris.

Ostroff, C. and Bowen, D.E. (2016), “Reflections on the 2014-decade award: is there strength in the construct of HR system strength?”, Academy of Management Review, Vol. 41 No. 2, pp. 196-214.

Pellegrini, E.K. and Scandura, T.A. (2008), “Paternalistic leadership: a review and agenda for future research”, Journal of Management, Vol. 34 No. 3, pp. 566-593.

Roberson, Q.M. and Stevens, C.K. (2006), “Making sense of diversity in the workplace: organizational justice and language abstraction in employees’ accounts of diversity-related incidents”, Journal of Applied Psychology, Vol. 91 No. 2, pp. 379-391.

Ronconi, L. (2012), “Globalization, domestic institutions, and enforcement of labor law: evidence from Latin America”, Industrial Relations: A Journal of Economy and Society, Vol. 51 No. 1, pp. 89-105.

Salzano, F.M. and Bortolini, M.C. (2005), The Evolution and Genetics of Latin American Populations, Vol. 28, Cambridge University Press, Cambridge.

Scott, W.R. (1995), Institutions and Organizations, Sage, Thousand Oaks, CA.

SENCE (2017), “Servicio Nacional de Capacitación y Empleo, Desarrollo y Mantención Unidad de Informática”, available at: www2.sence.cl/cursos_sence.htm (accessed 10 February, 2017).

Stevens, F.G., Plaut, V.C. and Sanchez-Burks, J. (2008), “Unlocking the benefits of diversity: all-inclusive multiculturalism and positive organizational change”, The Journal of Applied Behavioral Science, Vol. 44 No. 1, pp. 116-133.

Stewart, D.W. and Shamdasani, P.N. (2014), Focus Groups: Theory and Practice, 3rd ed., Sage Publications, Thousand Oaks, CA.

Suárez, E., Susaeta, L., Perello, M.R., Colón, C., Gutierrez-Martinez, I.O., Cunha, R.C., Leguizamón, F., Idrovo, S., Weisz, N., Correa, M., Apascaritei, P. and Pin, J.R. (2015), “CSR: does it foster inclusion and diversity? An exploratory study of Ibero-American countries”, Working Paper No. WP-1113-E, IESE Business School, University of Navarra, Barcelona, pp. 1-21.

Syed, J. and Özbilgin, M. (2009), “A relational framework for international transfer of diversity management practices”, The International Journal of Human Resource Management, Vol. 20 No. 12, pp. 2435-2453.

Taras, V., Kirkman, B.L. and Steel, P. (2010), “Examining the impact of culture’s consequences: a three-decade, multilevel, meta-analytic review of Hofstede’s cultural value dimensions”, Journal of Applied Psychology, Vol. 95 No. 3, pp. 405-439.

Telles, E. and Sue, C. (2009), “Race mixture: boundary crossing in comparative perspective”, Annual Review of Sociology, Vol. 35 No. 1, pp. 129-146.

The World Bank (2017), Data Catalog, available at: http://data.worldbank.org/data-catalog/world-development-indicators (accessed 20 February 2017).

Uysal, N. (2013), “Shifting the paradigm: diversity communication on corporate websites”, Public Relations Journal, Vol. 7 No. 2, pp. 8-36.

Vogt, D.S., King, D.W. and King, L.A. (2004), “Focus groups in psychological assessment: enhancing content validity by consulting members of the target population”, Psychological Assessment, Vol. 16 No. 3, pp. 231-243.

Voyages Database (2016), “The Trans-Atlantic Slave Trade Voyages Database, the national endowment for the humanities”, Emory University Digital Library Research Initiative, Atlanta, GA, available at: www.slavevoyages.org (accessed 10 January 2017).

Yang, Y. and Konrad, A.M. (2011), “Understanding diversity management practices: implications of institutional theory and resource-based theory”, Group & Organization Management, Vol. 36 No. 1, pp. 6-38.

Zajac, E. and Bazerman, M. (1991), “Blind spots in industry and competitor analysis: implications of interfirm (mis)perceptions for strategic decisions”, Academy of Management Review, Vol. 16 No. 1, pp. 37-57.

Acknowledgements

The author appreciates to the two blind reviewers and editors whose contributions enhanced the quality of this paper.

Corresponding author

Andrés Raineri can be contacted at: araineri@uc.cl