Making Sense of the Challenges of Diversity Management
- An Explorative Study from the Perspective of Change Agents
and a sub-title, if any
Master’s Thesis 30 credits
Program: Master’s Program in Business and Management
Specialization: Organization Department of Business Studies Uppsala University
Spring Semester of 2021
Date of Submission: 2021-06-02
Hanna Glas
Victoria Ulander
Supervisor: Leon Caesarius
Abstract
The importance of incorporating Diversity Management (DM) in organizations has become a viral debate in recent years in response to social pressures worldwide. While there is extensive research on the effects of DM, how the implementation process is managed is much less obvious. Furthermore, DM is a multidimensional concept that lacks a universally accepted definition in terms of organizational practices, which presents operations managers with various challenges. To identify these challenges and how they are managed, a qualitative case study was conducted through semi-structured interviews with Top and HR managers of a listed, Large Cap company in Sweden. The theories of Change Agents and Sensemaking were applied to analyze the implementation process of DM within The Company. The findings were categorized into; (1) the challenge of mobilization, (2) the challenge of creating a shared meaning, and (3) the challenge of the diversity paradox. Based on these findings, the conclusions indicated that unity, communication, awareness, and persistence are important factors to consider when managing these challenges.
Key words: Diversity Management, Inclusion, Organizational Change, Process, Culture, Communication, Change agents, Change recipients, Sensemaking, Sensegiving, Mental models.
Acknowledgements
We would like to thank a number of people who assisted us in the making of this research.
First of all, we would like to thank our supervisor Leon Caesarius for his valuable encouragement, guidance and feedback. We would additionally like to extend our utmost gratitude to the respondents from the case company. Without their support, commitment and valuable participation, this research would not have been possible. Furthermore, we would like to thank the informants from PwC for their guidance in the preparation of this study. We are also grateful for the comments, support and inspiring discussions of our seminar group.
Finally, we would like to give love and thanks to our families and friends for their emotional and motivational support during the entire thesis period.
Uppsala, June 2021
__________________ __________________
Hanna Glas Victoria Ulander
Table of Content
1. Introduction 4
1. 1 Background 5
1.2 Difficulties in Introducing Diversity Management 6
1.3 Problem Statement 7
1.4 Purpose 7
1.5 Research Question 7
1.6 Limitations 7
2. Literature Review 8
2.1 Implementation Complexity 8
2.2 The Role of Top and HR Management 10
3. Theoretical Framework 11
3.1 Change Agents 11
3.2 The Theory of Sensemaking 12
3.2.1 Sensegiving 13
3.2.1.1 Change Conversations 14
3.4 Theoretical Model 16
3.5 Theoretical Limitations 17
4. Method 18
4.1. Research Design 18
4.2 Data Collection 19
4.2.1 Preparation Work and Informants 19
4.3 Case Study 20
4.3.1 The Case Company 21
4.4 In Depth Semi-Structured Interviews 22
4.4.1 Selection of Respondents 22
4.4.2 The Interview Guide 23
4.5 Theoretical Relevance 24
4.5.1 Operationalization 25
4.6 Data Analysis and Coding 26
4.7. Ethical Considerations 28
4.8 Method Criticism 29
4.9 Scientific Quality 30
5. Empirical Findings 31
5.1 Background 31
5.2 Managerial Training 33
5.3 Decision-Making and Communication 34
5.4 Problems and Measures 37
5.5 Evaluation 39
5.6 Endurance and Integration 40
6. Analysis 41
6.1 The Challenge of Mobilization 41
6.2 The Challenge of Creating a Shared Meaning 43
6.3 The Challenge of The Diversity Paradox 47
7. Discussion & Conclusions 50
7.1 Conclusions 51
7.2 Contribution and Limitation 52
7.3 Suggestions for Future Research 53
List of References 55
Appendix A 63
1. Introduction
This chapter introduces the phenomenon of diversity management. It also provides a background to the identified problem, the purpose of the study, and the research question.
As a result of globalization and technical innovation, an increasing mobility in the workforce and migration have made workforce diversity a contemporary topic for organizations (Altman, 2020; Fuchs, 2020; García-Hodges, 2020; Tosczak, 2020). Diversity is a vague yet complex term that can be applied in many different contexts, for instance, by companies to promote their social responsibility (Byrd, 2018), or political parties to demonstrate their openness towards different cultures (Ward & Choi, 2020). The reason being that diversity is a noun that simply describes a condition of variety (Merriam-Webster Dictionary, n.d., diversity). However, research suggests that perceptions of workforce diversity are formed by individuals’ interpretations and biases of others, their attitudes towards organizational diversity policies, and the construction of self-identities (Ely & Thomas, 2001). Therefore, diversity is argued to be subject to sensemaking, meaning that people “make sense” of complex phenomena through social interaction that affects the cognitive aspects of beliefs, values, and knowledge (Roberson & Stevens, 2006).
In a workplace context, diversity can be defined as “the collective number of differences among members of a social unit” (Harrison & Sin, 2006). In turn, differences can be categorized in different dimensions (Alcazar, Fernández & Gardey, 2013; Podsiadlowski et al. 2013). According to scholars, there are primary dimensions, which are often more visible, such as gender, age, and ethnicity (Qin, Muenjohn & Chhetri, 2014), and there are secondary dimensions; for instance, job-relatedness, education, geographical location and work experience (Qin, Muenjohn & Chhetri, 2014). The two categories are not separated but rather interrelated, since “surface-level diversity such as race is indicative of deeper-level differences, such as cognitive processes/schemas, differential knowledge base, different sets of experiences, and different views of the world” (Shore et al., 2009:118).
It is presently recognized that diversity alone can not account for higher performance in comparison to other organizations, as several other factors such as resources, capacities, and competence also can explain variations (Jayne & Dipboye, 2004; Roberson & Park, 2007). In other words, higher performance does not emerge from diversity per se, but from the way it is
managed (Bruna, Chanlat & Chauvet, 2017). The organizational activities that aim to promote inclusion of employees of the different dimensions explained above are called diversity management (DM) (Thomas & Ely, 1996). In practical use, for instance, in policies and annual reports, DM is often referred to as Diversity & Inclusion (D&I) (Byrd, 2018).
There is a wide range of literature on the effects of DM. For instance, increased organizational competitiveness and performance, in terms of return on equity (e.g. Cox and Blake 1991; Richard, 2000), group member morale (e.g. Jehn, Northcraft & Neale, 1999), and team performance (e.g. Tegarden, Tegarden, & Sheetz, 2009). Furthermore, increased group conflicts due to diversity in for instance, race and tenure (e.g. Pelled, Eisenhardt &
Xin, 1999; Grimes and Richard 2003; Van Knippenberg, Van Ginkel & Homan, 2013). While studying the effects of change, most researchers agree that DM is an ongoing change process in organizations. However, fewer studies have investigated the actual implementation process, especially, how operations managers interpret the meaning of change, the challenges that arise, and how to manage them (Janssens & Zanoni, 2014; Holck, 2016).
1. 1 Background
The importance of incorporating DM in organizations has become a viral debate in recent years in response to social pressures worldwide (García-Hodges, 2020; Tosczak, 2020; Fuchs, 2020). In Sweden, the Discrimination Act was implemented in 2008, which covers grounds for discrimination in the workplace, education, labor market, employment services, and conducting of business (Prop. 2007/08:95). The regulative scenery shifted from anti-discrimination to more active DM when the Swedish government settled on new requirements for sustainable reporting (Riksdagsbeslut 2016/17:CU2). The requirements stated that large companies, with over 500 employees, must disclose how they work towards diversity and inclusion in their annual report.
DM can also be an attempt to respond to pressures from stakeholders, societal norms, trends, and expectations (Kalev, Dobbin, & Kelly, 2006). Large investors, such as BlackRock, State Street, and Vanguard, have recently expressed a higher demand for qualitative and quantitative information on how companies work with D&I (Pinilla & Hampole, 2020;
BlackRock, 2020; SSGA, 2020; Vanguard, 2020). Keeping up with modernization and branding are additional stakeholder pressures (Byrd, 2018). DM can also be influenced by cognitive and cultural factors, such as internalized values and beliefs (Romanenko, 2012).
For example, civil right movements like Black Lives Matter and MeToo have had a powerful impact on society at large, but also on company cultures, and values, since they have increased awareness and attention to unfair structures and code of conducts (Ledbetter &
Kinsman, 2019; Stengel, 2020). Following this path, previous research has highlighted that change initiatives that provide cognitive meaning are more likely to be successfully enforced in organizations (Molnar & Mulvihill, 2003).
1.2 Difficulties in Introducing Diversity Management
DM lacks a universally accepted definition in terms of organizational practices, which can serve both in favor and disfavor of its implementation efforts (Byrd, 2018). The implementation of DM can be framed to sound attractive to potential employees or stakeholders, however, it also means practical inconsistencies and that the actual scope of action is determined by the top and human resource (HR) management (Foster & Harris, 2005; Thomas & Lamm, 2012). The board of directors have the mandate to include DM in their business model and HR translates it to action (Byrd, 2018). The lack of definition makes it difficult to hold HR accountable for making the “correct” efforts (Hansen & Sierstad, 2017). Thus, D&I benchmarking has become a popular mechanism in determining the appropriate and normative way of conducting DM (PwC, 20201). Nonetheless, surveys still indicate that companies struggle with translating diversity strategies into execution (ibid).
In Sweden, making objective assessments of diversity efforts are related to both practical and ethical difficulties. The collecting and processing of personal information are regulated in the EU by the GDPR act of 2018 (GDPR.eu, 2021). Instead, companies can receive general statistics of employees of ‘foreign descent’ by anonymously sending the social security numbers to the Swedish Central Bureau of Statistics (SCB). Foreign descent means born abroad or that one or both parents are born abroad (SCB.se, 2021). Furthermore, collecting personal data can bring up cultural and ethical issues. Privacy and equality are two acknowledged pillars of the Swedish culture, thus collecting statistics on ethnicity or other attributes is culturally objectionable (Nordberg, 2015; Helmfrid, 2021; Hofstede Insights, 2021). Furthermore, categorization of personal attributes can raise ethical questions of why such information is relevant in the first place since it should not affect the perception of human value (Chen, 2011).
1PwC: Global D&I Data, 05/21/2020 based on ~ 3000 responses across 25+ industries
1.3 Problem Statement
According to academics and practitioners, implementing DM is a desirable but difficult organizational change (Foster & Harris, 2005; Byrd, 2018; Ward & Choi, 2020). Firstly, diversity is a multidimensional concept that can be interpreted differently depending on context and individual attitudes and values (Ely & Thomas, 2001). Secondly, even though social and contextual demands put pressure on organizations to implement DM, the process lacks definition and means for measurements. Thus, allows the top and HR management to interpret and implement it according to their judgment (Foster & Harris, 2005). Thirdly, given this background, the actors involved in DM often have a discrepancy of what “makes sense”, which can cause resistance in the change process and implementation phase (Weick, Sutcliffe
& Obstfeld, 2005).
1.4 Purpose
The aim of this study is to explore the implementation process of DM by investigating how DM is translated into cognitive and practical activities performed by top and HR managers.
The purpose is to identify the challenges they have faced, furthermore, how they have managed these. The study hopes to contribute with a deeper understanding of the application and progression of DM, furthermore, to provide practical implications for more efficient implementation for managers. In addition, this study aims to provide other stakeholders, that are influenced by DM, with an understanding of what challenges managers face and why they might act in certain ways in response to these.
1.5 Research Question
What are the challenges of implementing diversity management and how are they managed during the ongoing change process?
1.6 Limitations
This study is limited to the perspective of top and HR managers to review and understand the challenges of those who have the mandate and executive power to implement DM in organizations (Foster & Harris, 2005; Thomas & Lamm, 2012). However, reactions and interactions with employees, during the change process, are considered but from the perspective of managers.
2. Literature Review
This chapter will present previous literature within the studys’ problem area, which is relevant for the understanding of complexities related to DM implementation.
2.1 Implementation Complexity
Diversity management has been widely implemented in different sectors and cross-borders.
However, procedures and rationales vary depending on context (Hansen & Sierstad, 2017). In the DM literature, many scholars have investigated the strengths and weaknesses of different activities and initiatives, however, few have highlighted the challenges and reactions of their implementation and how managers have handled them (Holck, 2016). Below follows a review of previous studies within the problematized area.
There is an active debate, in both practice and research, on whether companies should regulate using hard or soft rules in the matter of DM (Spender, 2015). Previous studies have examined the effect of implementing pro-diversity hiring policies, meaning that companies actively seek candidates with diverse backgrounds (Mayer, Warr & Zhao, 2018). The results suggest that such policies increase the pool size of employees from which a company can find and hire talents, that the varied range of perspectives, backgrounds, and experience can contribute to more creative problem-solving (Horwitz & Horwitz, 2007), and finally that it can be effective in attracting and retaining surplus talents who were not necessarily covered by the policy itself (Mayer, Warr & Zhao, 2018). Criticism of such policies is that they can potentially reinforce “positive discrimination” and stereotyping, meaning that recruiters seek individuals purely based on their minority identity rather than their experience or qualifications (Foster & Harris, 2005; Lee, 2011; Byrd, 2018).
Multiple countries in the European Union have implemented mandatory legislative quotas for women in executive positions and corporate boards, whereas others, like Australia and the UK, have incorporated gender diversity into soft regulation as a form of moral responsibility (Villiers, 2010). Freidenvall and Hallonsten (2013) conducted a study on gender quotas in Swedish politics and concluded that gender quotas had the potential to increase the representation of women faster. Madison (2019) argued that quotas rather enforce biased selection and measures based on group membership rather than individual identity. Based on a thematic analysis, where he reviews inconsistent political statements that argue for
legislated sex quotas in Sweden, he concludes that quotas would lead to lower quality and productivity.
Pless and Maak (2004) assume that diversity itself is based on cultural norms and values, thus reshaping habits and norms in an organization is fundamental for the creation of a ‘culture of inclusion’. An inclusive culture means that the organizational environment allows people with different backgrounds and mindsets to work together, thus the authors emphasize reciprocal understanding, standpoint plurality, and mutual enabling as important factors.
Reciprocal understanding means inviting, supporting, empowering, and openly listening to different viewpoints, to recognize and understand each other’s basic assumptions. Standpoint plurality aims to prevent some voices from being marginalized when others are generalized due to unequal distribution of power. While the authors indicate that all situations imply an imbalance in power distribution, the principle of standpoint plurality is to raise awareness of the power aspect in the creation of an inclusive culture. Mutual enabling is linked to standpoint plurality in that different standpoints should be integrated into the decision-making and implementation of organizational processes. In addition, Pless and Maak (2004) suggest that having a normative culture of inclusion reinforces successful implementation of other diversity efforts. “Diversity is, first and foremost, a cultural question and thus a question of norms, values, beliefs, and expectations, /.../ Not before this is taken into consideration, acknowledged and institutionalized, can ‘‘diversity management’’ be successful.” (Pless & Maak, 2004).
Previous research has studied ways to develop more inclusive organizational cultures through information and awareness (Romanenko, 2012). Diversity training for the HR department and managers usually focuses on increasing awareness about social perception biases to avoid discrimination in hiring and promotion (Dobbin, Kim & Kalev, 2011). At the same time, diversity training and education have confronted managers with a paradoxical dilemma.
While they push the agenda for value in diversity and express the company's appreciation of human differences in all levels of the organization, management training also aims at eliminating the use of stereotypes and the acknowledgment of human group differences, presenting a paradox in the message (Lee, 2011).
Pless and Maak (2004) propose a selective view to DM called assimilation. The assimilation approach rather opposes inclusion, which may reduce the positive effects of DM even though
companies have policies or initiatives in place. Furthermore, assimilation ignores human differences instead of integrating them, more or less forcing women and minorities to assimilate into the dominant company culture. The result is that the initiative does not show the positive results expected, such as high turnover rates among minorities, changes in employee’s quality of work-life, or the creation of an inclusive atmosphere. Thus, the approach can enable tension, intrapersonal conflict, knowledge, and experience not being leveraged, loss of potential, employees experiencing barriers in advancing in the organization (Pless & Maak, 2004).
2.2 The Role of Top and HR Management
Barling and Robertson (2013) examine the role of leaders and their behavior in the creation of social norms and organizational behavior. The authors state that implementation processes become faster if the leader communicates information regarding the initiative’s effectiveness and adaptability, which also increases employees' positive emotions towards it. In addition, employees are more likely to be passionate about something of social and organizational importance, if a leader can inspire motivation that creates optimism about the individual’s contributions to the greater good (Barling & Robertson, 2013). An additional study by Byrd (2018) concludes that socially responsible leadership and action by those responsible for strategic planning are key for spreading the intrinsic value of diversity in organizations.
The top management and HR department are often emphasized in the literature on DM as the primary actors of change because they define and enforce strategies and efforts (Byrd, 2018).
The top management decides on how much effort (if any) should be directed to DM, with regard to the board of directors, shareholders, and financial planning (Mullins, 2018).
Furthermore, the board of directors has the mandate to allocate resources to different departments (ibid). Most studies underline the significant role of HR and its influence on DM (Dobbin, 2009; Dobbin, Kim & Kalev, 2011; Byrd, 2018). The reason is that DM is generally categorized as HR-related activities as it involves hiring, administering and developing the organization’s human assets. Thus it is up to them to translate the company values and goals, which are stated in the annual report, into actual efforts of change (Byrd, 2018). Furthermore, previous studies that have investigated DM efforts and results emanate from initiatives implemented by HR of organizations (e.g. Roberson & Stevens, 2006; Dobbin, 2009; Chen, 2011; Dobbin, Kim & Kalev, 2011; Holck, 2016; Mullins, 2018)
3. Theoretical Framework
This chapter provides a review of the theoretical framework, which focuses on the role of change agents, sensemaking, sensegiving, and the use of change conversations.
3.1 Change Agents
In this study, the theory of change agents aims to frame the perspective of the study as well as to highlight their role as decision-makers and operations managers in the implementation process. Change agents is the human factor in organizational change that has the capacity to act upon a situation or change structures and are commonly categorized through the activities they perform or their role in the organization (Iverroth, 2010). In relation to implementing initiatives and detecting challenges, the role of change agents is central to the process of implementing DM.
In Caldwell’s (2003) model of change agency, managers are often classified in the management model. In which, their roles as change agents include adopting (Kirton, 1980), implementing (Storey, 1992), and empowering change initiatives (Lawler, 1986). According to Cummings and Worley (2008), in a planned change approach, the actions of change agents can be categorized into three main activities. First, they collect the knowledge necessary for identifying the problem, which they will define and plan various change interventions for.
Next, the agents guide and lead the change recipients toward the desired end-state. Finally, they use tools and measurements to collect further knowledge and evaluations from the intervention to decide whether additional actions should be taken (Cummings and Worley, 2008). This approach is criticized for its assumption of rational action and democratic values, moreover, the leader’s unrealistic control of the entire process (Caldwell, 2006).
Beckhard and Harris (1987) suggest an extension to the planned change approach that includes the relationship between change actors and change recipients, which are those who adopt and adapt to the change, arguably the employees or other stakeholders in this case. The authors argue that successful change is rather the product of cognitive and analytical processes than rational control, thus the recipients need to intellectually understand the problems related to sticking to current status, or status quo, to want to change it. Furthermore, change agents need to practice cognitive and analytical oriented activities, where the value of
change is internalized by the recipients, thus creating a shared and common ground to successfully enable change (Caldwell, 2006).
3.2 The Theory of Sensemaking
Previous research of organizational change highlights the process theory of sensemaking as a tool for investigating the construction of change agents’ mental models, and to understand how they interpret and manage challenges during a change process (Gioia et al. 1994; Weick et al., 2005; Iveroth, 2010). Sensemaking is a social process by which we retrospectively give meaning to our experiences and socially construct the world around us and our actions in a mental model (Weick, 1995). In an organizational context, the common focus is to create shared meanings of, for instance, corporate objectives between managers, employees, and stakeholders (Starbuck & Milliken, 1988), which in this case refers to the common mental model of DM. The vision is disseminated through communication and interaction, guidance and leadership, professional training, or other types of social interaction (ibid). Furthermore, such processes influence the way people interpret, understand, and act upon change initiatives, hence sensemaking is argued to be essential for the emergent process of organizational change (Weick, 1995).
DM can be argued to be a matter of organizational sensemaking because it can be connected to identity, social contact, ongoing events, cues, enactment, plausibility, and retrospect, which are the seven properties of sensemaking (Weick, 1995:3). Based on identity construction and maintenance, sensemaking processes happen when individuals perceive situations in a way that helps them maintain esteem and consistency in their self-identities (Ring & Van de Ven, 1989). Thus as a part of the identity construction process, individuals categorize themselves and other people in groups in order to make sense of equivocal social situations (ibid).
Furthermore, as categorization based on visual features is common (Fiske & Taylor, 1991), ethnicity and gender are identity relevant (McCann et al., 1985), hence DM in organizations are likely to affect sensemaking in the workplace. Furthermore, sensemaking is also a social activity, in which individuals create and share different explanations to resolve ambiguous situations (Weick, 1995:38). Research on social networks suggests that demographically similar people have stronger interpersonal connections than demographically dissimilar people (Williams & O’Reilly, 1998). Thus it can be interpreted that equivocal events related to DM may be interpreted or socially shared differently depending on such demographic groupings, which can prompt divergent sensemaking of ongoing diversity initiatives and
efforts (Roberson & Stevens, 2006). Given that sensemaking is related to ongoing events and extracted cues (Weick, 1995), it indicates that organizational sensemaking of DM can occur in both formal initiatives and informal cues in everyday interaction. The sensemaking property of enactment entails that sensemaking is triggered by unexpected situations or discrepancies that interrupt the normal environment, hence causing individuals to start registering social cues directed to a certain process (Weick, 1995). Relating this to DM in organizations, enactment can be triggered by, for instance, global news, demonstrations, alone accidents, or another type of situation that can trigger sensemaking. Furthermore, the ongoing process of DM, similar to general organizational activities, can be viewed as
“negotiated orders for ambiguous work that are enacted among organizational partners involved” (Weick et al. 2005), thus analyzing the process of interactions and interpretations can help to make a complex phenomenon plausible (Gephart, 1993). Previous studies have provided evidence that managers temporally make sense of their own experience of change and discovered that there’s a configuring relationship between the past, present, and future (Wiebe, 2011), thus sensemaking is considered to be retrospective (Weick, 1995). By retrospection, previous research refers to a process in which action is not driven primarily by sense, instead sense is guided by action and a retrospective understanding of that action (Weick, 1995; Gioia, 2006). Therefore, the following section focuses on such actions, recognized under the concept of sensegiving.
3.2.1 Sensegiving
Previous studies argue that the interaction between change agents and change recipients is important to ensure successful organizational change (Beckhard and Harris, 1987; Caldwell, 2006). Thus, within the theory of sensemaking is the concept of sensegiving, which describes how individuals try to influence other people’s sensemaking processes and creation of mental models (Gioia & Chittipeddi, 1991). For instance, when a manager tries to influence the way employees interpret change initiatives by communicating their perspective of the change, to create a shared understanding to possibly affect the outcome (Iveroth & Hallencreutz, 2016).
However, managers are unable to fully control the process of sensemaking among employees, since mental models are also influenced by external cues from news, home, and informal discussions with friends, family, and colleagues (ibid).
Sensemaking and sensegiving are equally important for the construction of shared understanding and meaning (Gioia & Chittipeddi, 1991). The mutually reinforcing processes
can explain the process of organizational change, by moving from current to future frames of reality, which is explained in the figure below; A) A change leader, for instance, a manager, engages in sensegiving through the communication of visions and goals of a change initiative. B) The cue is picked up by an employee who tries to interpret and make meaning of the change, C) to then perform sensegiving back to the change leader, to try to influence the vision by providing their perspective. D) Based on this, the leader makes alterations to their sensemaking process, and so on (Iveroth & Hallencreutz, 2016). This model can also be used to understand change processes in retrospect, since a basic assumption of sensemaking theory is that people first act, then make sense of their action (Weick, 1995).
Figure 1. The reciprocal process of sensemaking and sensegiving (Iveroth & Hallencreutz, 2016)
3.2.1.1 Change Conversations
In the theory of sensemaking, people extract cues from narratives, conversations, and other interactions in their everyday life to make sense of the surrounding world (Weick, 1995). In an organizational context, such cues come from dialogues with co-workers, symbols from different communication sources, body movements, material artifacts, or events (ibid).
Furthermore, Weick (1995) argues that such cues are an important foundation of power for leaders because if they can control people’s point of reference, their attention and attitude towards the change can be redirected, which can ultimately influence the whole change process. Furthermore, to investigate the change agents’ sensemaking of DM, their sensegiving activities can be examined through the concept of change conversations (Barrett, Thomas & Hocevar, 1995; Ford & Ford, 2009; Quinn, 1996).
Ford and Ford (2009) illustrate how conversations lie at the heart of change and suggest that intentional change unfolds through four types of conversations: initiative, understanding, performance, and closure. They further argue that, if managed correctly, these forms of
conversation can move a change forward. At the beginning of a change process, the first type of conversation is initiative. This conversation is used to suggest new ideas, visions, goals, directions, and futures and to create attention to what is or should be done. These conversations aim to direct people’s attention and initiate conversations about the need for and urgency of change (Ford & Ford, 2009). The conversation of initiative can fail if the conversation is targeted at the wrong people, for instance, people who are in a position where they do not see themselves as having the skills, power, or authority to move the change forward (Ford & Ford, 2009; Iveroth & Hallencreutz, 2016).
The second conversation aims to create an understanding and enhance the exploration of a proposed change (Ford & Ford, 2009). The conversation can be based on, for instance, an event, a proposal or a plan, about which the change agent wants to open up for discussions.
The change agent is also open for questioning and analyses, which may be channeled by underlying reasoning and assumptions, in order to trigger dialogue and deep reflective conversations about the change. The mission is to achieve a more common ground of meaning, language, and shared context for the change, as well as to increase confidence in the availability of further development of opportunities and courses of action. The change agents need to “sell the idea” and get people to like it (Iverroth & Hallencreutz, 2016). A common pitfall for these types of conversations is that managers underestimate the importance of creating a common understanding with clear roles and structures, thus they do not spend enough time on it. The result is that change resistance will emerge because people are confused since they have not vented, discussed, or shared their feelings, thus cannot make sense of the situation (Ford & Ford, 2009).
The third conversation is that of performance, which focuses on getting to action and establishing accountability (Ford & Ford, 2009). This conversation is usually raised when a manager wants to achieve something and enforces this by making requests using, for instance, deadlines, time, quality, and cost. In a successful conversation, the other party listens and makes promises of this achievement, thus an agreement and accountability for action are established (ibid). These types of conversations usually fail if leaders are unclear about who is accountable for what or take for granted that the employees already know that.
(Iveroth & Hallencreutz, 2016).
The final type of change conversation is closure, which is intended to establish and summarize the change (Ford & Ford, 2009; Iveroth & Hallencreutz, 2016). This type of conversation calls for acknowledgment and celebration of what has been achieved, with the aim to motivate people to let go of the past and move forward by acting on the new opportunities enabled by the change (ibid). If leaders fail to acknowledge the progress or accomplishments, the employees can feel that their contributing efforts are unappreciated or that they didn’t make a difference (Ford & Ford, 2009). In conclusion, these four types of conversations are often present simultaneously or jump between different phases of change as they are dynamic in social interactions (ibid).
3.4 Theoretical Model
The theoretical framework of this thesis contains the role of change agents (Beckhard &
Harris, 1987; Caldwell, 2003; Caldwell, 2006; Iverroth, 2010) and the process theory of sensemaking (Weick, 1995), which includes sensegiving (Gioia & Chittipeddi, 1991) and change conversations (Ford & Ford, 2009). The level of analysis is based on top and HR managers acting as change agents since they have the mandate to translate and implement DM into organizational practices (Byrd, 2018). Furthermore, their mental models are investigated through the application of sensemaking, sensegiving and change conversations.
Studying the mental model of the change agents is relevant, as it provides the foundation for how the process of DM is initiated, communicated, and executed, despite renown difficulties related to the conceptual complexity of diversity. Moreover, studying the implementation process retrospectively can provide findings of different challenges and the use of change conversations at different stages of the process.
Figure 2. Theoretical model
3.5 Theoretical Limitations
The process theory of sensemaking is usually applied to real-life investigations as the essential idea is to understand how individuals continuously construct reality through social interaction (Weick, 1995). As this study aims to investigate the sensemaking of past experiences during the process of implementing DM, the theory is applied using a retrospective approach. Retrospective sensemaking, is used to explain past experiences that are reconstructed in the present, as an act of recollection (Petersen, 2019). However, due to the continuous nature of the sensemaking theory, the aspect of temporality can be problematic as recollection of past events can be influenced by present events, future goals or other heuristics (Kahneman & Tversky, 1979; Kahneman & Miller, 1986). Thus, past experiences are malleable and equivocal because the sensemaker continuously changes their perception due to current or future frames (Weick, 2001; Gioia & Chittipeddi, 1991). Nevertheless, Weick, Sutcliffe and Obstfeld (2005) argue that retrospective sensemaking is inevitable when making sense of changes that are results of actions. The quote; “how can I know what I’m seeing until I see what it was” describes how one makes sense of the character of an act in its aftermath (ibid).
Retrospective sensemaking has also been applied to specific case studies investigating organizational change. For instance, a comparative case study was conducted to address the relational challenges in internal R&D (Research & Development) collaborations (Einola et al., 2017). The data was collected through interviews with customers and suppliers, of which the authors applied the concept of retrospective sensemaking to understand their perspective of past challenges and actions during the process. Their study has a number of similarities to this current study, which argues for the applicability of this theory. Firstly, both studies focus on a specific organizational change process, R&D collaborations and DM. Secondly, their case companies were early to engage in R&D offshore relationships, thus providing them with an opportunity to study the entire change process retrospectively using sensemaking theory. The same condition of early adaptation is applied to our case company in the sense of DM. Thirdly, the findings are based on the interpretations of actors that were involved in the change process, through retrospective sensemaking of past experiences. Fourth, both studies aim to investigate complex change processes that include cognitive evaluations of social interaction, which according to Weick (1995) can benefit from sensemaking.
4. Method
This section presents the reasoning behind the choice of research design, the characteristics of the employed qualitative research approach and how the empirical data was analyzed.
4.1 Research Design
The purpose of this study is to explore the change process of DM implementation to highlight related challenges and how they are managed. Bryman and Bell (2015) state that organizational change should be reviewed using a qualitative method to mirror the continually unfolding experience of strategies, while creating a contextual understanding of an event. Moreover, qualitative methods are often able to show events and patterns over time.
Thus, focusing more on the process, which includes how an organization’s past can change its present reality, while also providing an explanation for its current state (ibid). Therefore, a qualitative research design was used for this study.
Social science research commonly involves both deductive and inductive reasoning processes, moving iteratively between them during the research process refers to abduction (Eriksson & Kovalainen, 2015). A deductive approach was used for the operationalization of the interview guide since it was based on theory and designed to answer the research question. The themes for data analysis were identified and generated with an inductive approach since it was based on empirical raw data, and not theoretical concepts. The study therefore used an abductive analysis method. This choice was made to avoid limiting the raw data and provide a less rich description while still maintaining a connection to the interview guide and research question (Nowell et al., 2017).
By exploring how change agents make sense of DM, manage its challenges, and how they act to communicate this to the rest of the organization, an exploratory approach is used.
Exploratory research is often used when the aim is to discover factors in the early stages of research of a new topic, which focuses on ‘how’ and ‘why’ questions (Myers, 2013). The study was conducted on one case company, to retrospectively understand the challenges of DM and how they were managed. Furthermore, to explore the sensemaking of DM from ‘the inside’, through the eyes of the people involved in the implementation of it to illuminate the managerial challenges (Eriksson & Kovalainen, 2015).
Cox and Hassard (2007) argue that interpreting and understanding the past with retrospective research can contribute to generating potential for new futures. Furthermore, Weick (1995) argues that sensemaking is an ongoing process which occurs when individuals make retrospective sense of past situations and creations. Thus, making a retrospective perspective suitable for this study. Moreover, choosing a retrospective approach provides data from changes over a longer period of time, which would not have been possible otherwise due to the scope of this study.
4.2 Data Collection
Qualitative research and methods focus on meaning in context, thus creating a need for data collection that is sensitive to underlying meanings and understanding how individuals interpret and give meaning to their experiences (Merriam & Tisdell, 2015). In light of this, the collection of empirical data was done through semi-structured qualitative interviews.
Bryman and Bell (2015) state that semi-structured interviews, where participants are asked to reflect on processes leading up to or after an event, can create an understanding for managers' experiences of change processes. Furthermore, in business research, interviews have been typically used as the primary data source in case studies (Eriksson & Kovalainen, 2015).
However, the collection of data is not focused on individual perceptions and opinions, but rather on an organizational level and the collective understanding of top and HR managers.
Due to the Covid-19 pandemic the interviews were conducted online through Microsoft Teams. Cassell (2015) states that conducting interviews through similar online tools has many advantages, such as having access to visual cues while still conducting the interview at a distance. Moreover, unstructured informant meetings were conducted in a preparatory study to provide insights and inspiration for the formulation of the problem statement.
4.2.1 Preparation Work and Informants
Preparation work was conducted for two separate reasons. First, to identify and formulate a relevant problem statement within the research area of DM, which was achieved with the support of two informants. Two separate online meetings were conducted with a sustainability expert and a diversity expert, both working for PwC, which is one of the world’s largest audit and consulting firms (Tadros & Wootton, 2020). The meetings were conducted in an unstructured and exploratory manner which is commonly used to explore a general area of interest, allowing the perceptions of the informants to guide the conversation
(Cassell, 2015). While a small number of broad questions were prepared to help guide the conversation if needed (Appendix A), the informants were encouraged to speak freely and open up about their perceptions of sustainability and diversity, what challenges they have faced, and predictions for the future (Saunders, Lewis & Thornhill, 2012).
The second part of the preparation work was motivated by that in order for case studies to explain a complex phenomenon within a real context, interviews are often complemented with an assortment of documents (Gerring, 2007). Thus, in order to gain more knowledge of the context and to compile more background information of the case company, a review of the case company’s annual reports was made. The annual reports contained additional information of the company’s general sustainability efforts and DM initiatives. By reviewing the annual reports over the last decade, a noticeable increase in disclosed D&I efforts could be observed, indicating a shift from passive to active engagement in DM. Hence, the study’s scope and time-frame could be devised.
4.3 Case Study
A case study requires an in-depth analysis of a case, also described as a social or theoretical phenomenon in its context. Case studies aim to answer how or why something happens, thus providing insights into processes, interactions and dynamics (Ciesielska, Jemielniak & Iiwa, 2018). A group, an individual, a process, or social relationships can all be considered ‘cases’
and therefore be subject for research (ibid). In this study, the implementation of DM and its challenges through the eyes of change agents is considered the case, thus making it a single case study (Lapan, Quartaroli & Riemer, 2012). Case studies can be used to explain the dynamics of certain historical periods, thus making it an appropriate method for exploring process-related change in organizations (Eriksson & Kovalainen, 2015). Moreover, case studies are deemed useful when the research topic is broad, complex, and when context is of high importance, which is true for organizational ethical orientations, making it an appropriate method for research in DM (Dul & Hak, 2008).
According to Gerring (2007) there are several characteristics for a case study such as, (1) a small population, (2) the research is holistic and therefore enforces a comprehensive examination of a phenomenon, (3) a qualitative research design, (4) examining a specific phenomena, (5) the topic is diffuse (case and context are difficult to distinguish), (6) a naturalistic or real-life context evidence gathering method, (7) it employs triangulation.
While characteristics 1-5 can be applied to this study, due to the Covid-19 pandemic, criteria 6 and consequently 7, could not be fulfilled. While including more primary sources in the form of observations would have increased rigour and credibility through triangulation (Lapan, Quartaroli & Riemer, 2012), it was not possible since all the respondents were working from home. Instead, interviews of key informants were the sole data collection method, in combination with document analysis during the preparation work.
In this study one case company was explored, which will be further explained below. This is preferable when examining causal mechanisms rather than effects, since the purpose is not to find out what effects DM can have, but rather the challenges that arose during the implementation process (Gerring, 2007). External validity may be questioned when conducting case studies due to the small population used for the identified phenomena.
However, small population cases often excel in regards to internal validity (ibid). In addition, Silverman (2020) states that case studies of smaller populations can have great significance for the understanding of social processes by examining the contextual dimension in-depth that influences a phenomenon.
4.3.1 The Case Company
When deciding what company would be suitable for this research, two criterias were used.
Firstly, the company had to have over five hundred employees. This dimension of relevance is motivated by the fact that companies of that size have to disclose their efforts towards a more inclusive and diverse workplace by law (Riksdagsbeslut: 2016/17:CU2). Secondly, the company had to have disclosed D&I efforts a few years back in time, since it otherwise would not be possible to examine the implementation process of DM retrospectively.
The subject for this case study is a global large cap, listed corporation with headquarters in Sweden. The Company has been anonymized by request from the respondents, thus it will be labelled The Company in this study. The relevancy for this specific company is firstly motivated by their size. Secondly, the company provides diversity and non-discrimination reports on their website and annual reports, providing a transparent view into their employee statistics, both regarding gender and age. Furthermore, The Company specifically states that they aim for an inclusive company culture that promotes cultural and gender based diversity on all organizational levels. Lastly, the case company has a wide business including
manufacturing, distribution, sales, R&D, marketing and HR. Therefore, they employ a variety of people with different educational, socio-economic backgrounds with ranging positions.
4.4 In Depth Semi-Structured Interviews
Saunders et al. (2012) recommend semi-structured interviews for studies in which the reasoning behind decisions, attitudes and opinions are relevant to understand. Furthermore, in studies with a large quantity of interview questions that are of complex or open-ended nature.
Therefore, not only providing information regarding ‘what’ and ‘how’, but also providing an emphasis on ‘why’. Furthermore, semi-structured interviews are suitable for an exploratory approach (Cassell, 2015). The format of the semi-structured interviews are theoretical, due to the questions being thematically organized around different theoretical concepts that have been identified in the literature review and theory section (ibid).
Sreejesh, Mohapatra and Anusree (2014) state that semi-structured interviews provide some flexibility for the interviewer while keeping them limited to the researcher’s topic of interest.
The flexibility allows follow-up questions, or “probing”, providing the opportunity of diving deeper into the experiences of the respondents while also encouraging them to provide relevant details. Respondents may use words or ideas in certain ways, making it possible to probe their meanings and add significance and depth to the collected data, while also providing a path into areas that were not previously considered relevant yet in reality can aid in fulfilling research objectives (Saunders et al., 2012).
4.4.1 Selection of Respondents
Myers (2013) states that finding ‘key informants’ is important when conducting a case study.
These are characterized by being the most knowledgeable about the topic of interest within the organization, while also having decision-making authority for it. In case studies, the selection of respondents are often done through purposive sampling to ensure relevance, meaning that they are strategically chosen (Lapan, Quartaroli & Riemer, 2012; Cassell, 2015;
Silverman, 2020). The criteria for selecting respondents was that they had been involved in the decision-making and/or implementation of diversity initiatives in the case company, thus representing the population that is studied. Respondent 3, who has the highest responsibility for managing this process in the case company, assisted the sampling of the other respondents. Due to her experiences and insights in both the DM process and the company’s general structure, her assistance further ensured a strategic selection of key informants.
The information collected from the interviews determined the sample size of respondents. In qualitative research, there are no set rules when it comes to having a certain number of participants, it rather depends on what the study aims to explore (Ravitch & Carl, 2020).
Furthermore, the sample size of this study was decided by roles that were accessed within the case company, the information they could provide and the credibility of that information.
Hence, a small sample of key informants is suitable since the goal is not to generalize but to explore a complex phenomenon by deliberately, ethically and thoroughly answering the research question, in order to achieve a deeper understanding of a complex phenomenon that is contextualized and valid (Ravitch & Carl, 2020).
Table 1. Overview of Respondents.
Respondent Position Tenure
(current role) Management scope Interview duration
Respondent 1 President Europe 7 years Top management 41 min
Respondent 2 Senior Vice President (SVP) Legal Affairs & Head Legal
Council
6 years Top management 56 min
Respondent 3 Vice President (VP)
Organizational Development 6,5 years Top management &
HR management 1h 52min
Respondent 4 Culture & Communications
(C&C) Manager 4 years Top management &
HR management 45 min
Respondent 5
Human Resource (HR) Manager Distribution
Subsidiary 3 years Top management &
HR management 43 min
4.4.2 The Interview Guide
The interview guides were similar to all the respondents with some adaptation to their respective roles. For instance, the interviews with respondents solely in the top management focused more on the decision-making process than the implementation process, while the interviews with respondents in both top and HR management focused on both. The interview guide was constructed so that each interview would take approximately 45 minutes, including both the predetermined questions and supplementary questions. One of the interviews, with Respondent 3, was appreciably longer even though the interview guide was equivalent to the others. The reason being that Respondent 3 had the possibility to present business material from management training programs, the Code of Conduct and more thoroughly go through examples from the entire process of implementing DM.
Every interview began with a general introduction that included; a short summary of the study, an explanation of the interview’s layout, and some ethical considerations which will be explained further in a separate section. The initial questions of every interview aimed at introducing the respondent with regard to their position and for how long they had held their current position in the organization. The respondents also answered when they perceived the process of active DM started, in order to provide the study with a scope and time frame and strengthen the observations made from their annual reports. The actual interview guide was constructed with reference to previous research and relevant theories, in order to stay consistent with the study’s research question and purpose.
4.5 Theoretical Relevance
In order to investigate what challenges appeared during the period of DM implementation, and how they were managed, the process was explored. Supported by the process theory of sensemaking (Weick, 1995; Iveroth & Hallencreutz, 2016) with specific focus on change agents (Iverroth, 2010) and their conversations of change (Ford & Ford, 2005), noted challenges during the process from the organization’s initiation of DM to the current state of DM was charted. The role of change agents is motivated through previous research on DM that has elevated the significant role of, especially the HR department, but also the organization’s top management, in the translation and implementation of DM efforts (e.g.
Byrd, 2018; Dobbin, 2009; Foster & Harris, 2005).
The theory of sensemaking and sensegiving can provide a basis for understanding the top management and HR managers’ mental model of DM, which potentially affects how they communicate through change conversations with employees and other stakeholders, hence, influence the acceptance of change (Armenakis & Harris, 2002; Bernerth, 2004). Studying the mental model of the top and HR managers is thus relevant, as it provides the foundation for which the process of DM is initiated, communicated and executed, despite renown difficulties related to the conceptual complexity of diversity. Furthermore, with support in the sensemaking theory, the decision-making of change agents are highly dependent on cues from employees or other stakeholders (Iveroth & Hallencreutz, 2016), thus examining the conversations of change can shed light on potential challenges caused by resistance or scepticism during the process.
4.5.1 Operationalization
To illustrate the theoretical relevance further, as well as the confirmability of the interview design, all interview questions were operationalized with reference to suitable aspects of the chosen theories as well as extractions from previous research. The operationalization is based on four themes; change agents, sensemaking, sensegiving & change conversations and the retrospective change process. Table 2 below, presents how the interview questions are related to theory, a method that is suggested by Bryman and Bell (2015), as it can be helpful when analyzing the results. The purpose of the questions is also presented to justify their relevance in attaining the information necessary for answering the studys’ research question. The interview guide originated from these four themes, which allowed flexibility and adaptability to every single respondent, keeping the reflexive nature of the interviews (Alvesson, 2003).
Furthermore, the guide allowed the interviews to be controlled within the study’s framework of purpose.
Table 2. Operationalization of interview questions
Theme Reference Main Interview Questions Purpose
Change Agents
Kirton (1980); Storey (1992); Lawler (1986);
Cummings & Worley (2008); Mullins (2018)
- In your role as X, what part have you played in the implementation process?
- How do you experience your level of power in the implementation process?
- How does the decision-making process work, from initiative to implementation?
Contributes to the understanding of how the change process was initiated, managed and controlled, from a perspective of executive power.
Also provides insights in challenges and measures taken
during decision-making processes.
Sensemaking
Iveroth (2010); Weick (1995); Starbuck &
Milliken (1988)
- What is your general perception of the ongoing DM process?
- What are the main drivers of DM, according to you?
- What are the main challenges of DM, according to you?
- What is the purpose of diversity management, according to you?
- Where do you think you get this understanding from?
- Has this perception changed during the process of DM?
Contributes to a deeper comprehension of top and HR
managers’ mental models of DM and how they are constructed. Provides insights into the drivers and motivational
factors for engaging in DM, furthermore, increases the understanding of how they might affect the change process.
Moreover, gives insights into the main challenges of actors of
change in DM.
Sensegiving &
Change Conversations
Gioia & Chittipeddi (1991); Ford & Ford
(2009)
- How have you communicated this understanding to the
employees/other stakeholders?
- How are diversity initiatives introduced throughout the levels and departments of the organization?
- How is information about initiatives spread throughout the organization?
- How do you establish compliance and accountability for diversity policies?
- How do you communicate the level of progress?
- Do you perceive DM to be established in the everyday operation?
- How do you notice this?/How can it be improved?
Contributes to the understanding of how top and
HR managers communicate their mental model of the DM to
the employees or other stakeholders in order to push the
change process forward.
Through supplementary questions, specific challenges and responsive actions related to
responding cues from employees/other stakeholders
can be surveyed.
Retrospective Change Process
Weick, Sutcliffe &
Obstfeld (2005); Einola et al.,(2017)
- How did the process of DM start?
- What were the initial actions?
- What were the reactions and how did they reach you?
- How did the process progress?
- How did the process become a separated “movement”?
(e.g. language use, symbols, events, activities?)
- Did you notice any resistance?
- How did you respond to resistance?
- Could you give some examples of events, challenges or
accomplishments you’ve experienced during the process?
Provides a retrospective overview of the change process from start to present state. Gives
insight into the challenges and strategies of the different phases
of change. Gives insight into how top and HR managers retrospectively make sense of
past experiences and actions, which contributes to the construction of their current
mental models of DM.
4.6 Data Analysis and Coding
One of the main difficulties with qualitative research is the vast amount of data that is collected, in this case through interview transcripts, and the challenge of finding analytic paths (Bryman & Bell, 2015). As suggested by Bryman and Bell (2015), it is important to transcribe and code as soon as possible to sharpen the understanding of the material and avoid the feeling of data overload. Thus, the authors transcribed the interviews directly after they were conducted and started coding immediately when all the transcriptions were done.
Since the interviews were conducted in Swedish, the coding and transcription were done in the original language to avoid concerns of translation accuracy due to linguistic differences (Cassell, 2015). Quotes used to present the empirical findings were carefully translated into
English to ensure a correct interpretation. To increase credibility, the data was analyzed by both of the authors to contribute to a deeper engagement with the data (Nowell, et al., 2017).
The data analysis was based on the studys’ empirical findings with the aim to answer the research question. A thematic analysis (TA) method was applied due to its flexibility to be modified for the needs of specific studies and usefulness for summarizing large data sets (Cassell, 2015). Furthermore, it is argued that TA is useful when examining the perspectives of different research participants and highlighting similarities and differences (Nowell, et al., 2017). TA is a method for identifying, analyzing, and interpreting patterns of meaning, or themes, within qualitative data from both small and large data sets. (Clarke & Braun, 2017).
The basis of the systemic procedures of finding analytic paths was provided by generating codes and themes from the transcribed data. Themes are the larger patterns of meaning which are built up by codes, the smallest unit of analysis that captures data relevant for the research purpose. Clarke and Braun (2017) emphasizes the organic approach of TA to coding and development of themes, thus giving the researcher a more active role and therefore providing more flexibility. Themes are then underpinned by an organizing concept, the research question. However, the aim is not to summarize the data into different themes, but rather to identify and interpret key features of the data guided by the research question while also providing a framework for organizing and reporting analytic observations.
Table 3. Coding scheme and examples of coding
Research Question
Main Themes
Empirical Themes
Codes Quote Examples
Mobilization
Decision- Making
Management
“I know from experience that it’s almost impossible to drive change forward unless the management is behind the process”
Bottom-Up
“I believe in building from the bottom, so the employees can express their ideas and concerns, not just the management”
Unity Adaptation
“You always have to think about the target group you are addressing”
What are the challenges of implementing
diversity management and
how are they managed during
the ongoing change process?
Inclusion
“It is hard to reach everyone, especially when you have your own production”
Creating a Shared Meaning
Drivers
Justice “It is not fair unless everyone is given equal opportunities.”
Business Performance
“We want the best person in the right position, that is the goal”
Communication
Formal
“We communicate goals and statistical measures in ‘business updates’”
Casual
“I believe the biggest change happens in the corridor rather than in a meeting”
Evaluation
Qualitative
“The qualitative goals have to do with norms, attitudes and power balance”
Quantitative
“The quantitative goals are usually relatively simple to conduct. For example, we want to increase the share of women or people of foreign descent in executive positions.”
The Diversity
Paradox
Initiatives
Policies
“We never use quotas for the final recruitment; however, we do try to prioritize the
under-represented group.”
Education
“It is hard not to fall into definition traps when creating material for education.”
Norms
Stereotypes
“The norm is a minority, namely, a white, middle aged man, straight, upper-middle class with no physical dysfunction.”
Human difference appreciation
“Our personal qualities, wishes and capabilities vary more between individuals than between genders”
