The dynamics of communication in
global virtual software development
teams
- A case study in the agile context during
the Covid-19 pandemic
Master’s Thesis 30 credits
Department of Business Studies
Uppsala University
Spring Semester of 2020
Date of Submission: 2020-06-03
Abstract
The increase of globalization of business led to the creation of global virtual software development teams in which communication plays an important role. The lack of recent studies raises the need to investigate communication in the context of agile and the current Covid-19 pandemic. This study identifies in literature three factors present in global virtual software development teams: technology, culture, and trust. It aims to explore how they influence internal communication. A qualitative method is conducted on a case study by combining data collected from semi-structured interviews and observations. The case study is a global virtual software development team which implemented the agile Scrum methodology. The findings provide a deep understanding of the positive and negative influences of the factors on communication, including their interrelations and context. Some of the main findings are: (I) the best set up for collaboration among global team members is the combination of synchronous and asynchronous communication through technological tools, (II) cultural diversity leads to differences in the ways members communicate, (III) trust is the precondition of working collaboratively and communicating effectively, (IV) global team members need to occasionally have in-person interactions to nurture their interpersonal relations, (V) the agile Scrum methodology influences internal communication positively, (VI) each communication factor is influenced by time; thus, the theoretical framework developed for this study is updated with this addition, (VII) the Covid-19 pandemic affects communication to a certain extent.
Keywords: Global Virtual Software Development Team, GVT, Internal Communication, Virtual
Communication, Communication Factors, Culture, Technology, Trust, Agile methods, Scrum methodology, Covid-19 pandemic.
Acknowledgments
This paper is the result of efforts and dedication which allowed me to not only use my academic knowledge but also gain a deeper understanding on a topic that I am passionate about. This is the completion of a personal achievement.
I would like to express my deepest appreciation to my supervisor Mrs. Cecilia Pahlberg who provided me with encouragement, constructive feedback, and patience throughout the duration of this Master project.
I am extremely grateful to each member of the global virtual team case that gave me valuable insights and put in the effort and time for this study. Thank you for your important contribution! I would also like to thank my fellow students who have contributed with treasured inputs along the way.
Last but not least, the completion of my thesis would not have been possible without the support of my beloved ones.
Table of Contents
1 Introduction ... 1
1.1 Background ... 1
1.2 Problem statement ... 2
1.3 Research question and contribution ... 2
2 Literature Review ... 3
2.1 Organizational communication ... 3
2.2 Definition and characteristics of global virtual software development teams ... 4
2.3 Communication in global virtual software development teams ... 5
2.3.1 The use of technology and the computer-mediated communication ... 6
2.3.2 Cultural diversity ... 7
2.3.3 Intercultural communication ... 9
2.3.4 Interpersonal trust ... 9
2.3.5 Interpersonal relations ... 11
2.4 Agile project management in software development ... 11
2.4.1 Agile software development ... 11
2.4.2 Agile Scrum methodology and internal communication ... 12
2.5 Communication theories in project management ... 14
2.5.1 Social Information Processing Theory ... 14
2.5.2 Communication accommodation theory ... 15
2.6 Theoretical framework ... 16
3 Methodology ... 17
3.1 Research and Methodological Approach ... 17
3.2 Research design ... 17
3.2.1 Single case study ... 17
3.2.2 Case company ... 18
3.3 Methods and data collection techniques ... 18
3.3.1 Semi-structured interviews ... 18
3.3.2 Interview guide ... 20
3.3.3 Observations ... 20
3.4 Data analysis procedure ... 21
3.5 Research quality ... 22
3.6 Method limitations ... 22
3.7 Ethical considerations and reflections ... 23
4 Empirical Findings ... 23
4.1 Team overview and formation ... 23
4.2 The implementation of Scrum methodology in the team case ... 23
4.3 Technology factor ... 24
4.3.1 Challenges and opportunities of ICT ... 25
4.3.2 Advantages and disadvantages of CMC ... 26
4.3.3 Build relationship online ... 27
4.4 Culture factor ... 27
4.4.1 Cultural diversity ... 27
4.4.2 Intercultural communication ... 28
4.4.3 Interactions in a culturally diverse GVT... 29
4.5 Trust factor ... 30
4.5.1 Interpersonal trust ... 30
4.5.2 Interpersonal relations ... 31
4.6 Agile setting ... 31
4.6.1 Scrum and internal communication ... 31
4.6.2 Scrum methodology and the GVTs’ factors on communication ... 32
4.6.3 The effect of Covid-19 pandemic on communication ... 33
5 Discussion ... 34
5.1 Technology factor in GVTs ... 34
5.2 Culture factor in GVTs ... 36
5.3 Trust factor in GVTs ... 37
5.4 GVTs in the agile setting ... 39
5.4.2 The effect of the Covid-19 pandemic on communication ... 40
5.4.3 The updated theoretical framework ... 41
6 Conclusion... 42
6.1 Concluding remarks and research contribution ... 42
6.2 Future research ... 44
References ... 46
Appendix A: Operationalization of the theoretical framework ... 52
Appendix B: Interview Guide ... 53
Appendix C: Technological tools ... 54
Appendix D: Scrum sprint process ... 54
Appendix E: Examples from the Coding Table... 55
Appendix F: Examples of thematic analysis for observations ... 55
Abbreviations
CAT - Communication Accommodation Theory CMC - Computer-Mediated Communication GSD - Global Software Development GVT - Global Virtual Team
ICT - Information and Communications Technology SIPT - Social Information Processing Theory
1 Introduction
1.1 Background
In recent years, the increase of globalization of business has led software organizations to adopt the practice of global software development (GSD) to manage their projects (Herbsleb and Moitra, cited in Holmström et al., 2006; Jusoh et al., 2018). In GSD, software development projects are handled by global virtual teams (GVTs), which are generally described as “culturally diverse, geographically dispersed, electronically-communicating workgroups” (Daim et al. 2012, p. 199). In GVTs, communication plays an important role since it allows sharing information, documentation and opinions among team members. In this way, communication supports and increases the results of team efforts (Sarker et al., 2011).
There are several advantages for adopting GVT, including the opportunity to invest in capital resources in foreign countries, to hire the most talented experts and, at the same time, to decrease development costs and time to market (Bird et al., 2009; Lane and Ågerfalk, 2009). However, besides the benefits of GVTs in global software development, team members face some challenges in establishing effective communication and coordination mechanisms (Shah, Raza, and UlHaq, 2012). In fact, communication can become critical in GVTs due to the geographical and temporal distances and the resulting lack of shared understanding (Piccoli, Powell and Ives cited in Sarker et al., 2011).
In software projects, it might be difficult, among all parties involved, to clearly define in advance the requirements and specifications of the scope for the implementation of the software development (Paasivaara and Lassenius, 2006). Furthermore, due to the current competition in the market and the continuous change of the business environment, projects involving innovation and software development require high flexibility and agility (Azanha et al., 2017). For these reasons, GSD teams prefer to use agile project management which is “an iterative and incremental approach to delivering requirements throughout the project life cycle” (Apm.org.uk, 2019). Agile software development involves communication, trust, flexibility, empowerment and collaboration as central values and behaviors of any project (Agile Alliance, 2001).
Currently, the world is facing a global health crisis leading to a social and economic disruption caused by the Covid-19 pandemic (UNDP, 2020). Organizations are encountering new challenges and are forced to change their ways of working and interacting among employees, customers, stakeholders and consumers. Managers and team members are forced to adapt their working approach by interacting fully virtually in response to the constraints imposed by the Covid-19 pandemic (Korolov, 2020). Those GVTs from software development with members working in several offices located worldwide, might also be affected by the pandemic. This is due to the fact that, for instance, they are working fully remotely without the access to a physical
office. Therefore, all organizations have to be agile in responding to this new challenge and find new ways for the employees to work and communicate effectively.
1.2 Problem statement
Global virtual software development teams are surrounded by a challenging environment characterized by a volatile business environment (Holmström et al., 2006); agile project management provides solutions to software teams through its methodologies and practices to react to those challenges (Shrivastava and Date, 2010; Mishra and Mishra, 2009; Pikkarainen et al., 2008). However, GVTs have to overcome other issues related to communication among team members due the spatial, temporal and culture distances (Ågerfalk and Fitzgerald, 2006; Gill, Bunker and Seltsikas, 2012). Effective communication is crucial for software development teams to ease knowledge transfer among team members quickly, to allow team members to apprehend requirements from clients, and to give support to team members in their development activities (Pikkarainen et al., 2008; Barlow et al., 2011). Many studies have shown that communication among the members of a team plays an important role since it can determine the success or failure of a project (Barlow et al., 2011; Castka, 2001).
In literature, there are few studies that explore communication among agile software development teams (e.g. Pikkarainen et al., 2008). Pikkarainen et al. (2008) investigated the impact of agile practices on communication in software development co-located teams. The researchers suggested, for future research, to investigate the effects of agile practices on communication in more complex contexts, for instance in GVTs from software development. Moreover, there are several studies concerning the phenomenon of communication in GVTs from software development (Daim et al., 2012; Holmström et al., 2006). However, these studies do not take into consideration the setting in which the team is working. There are no recent studies concerning communication in GVTs in the agile context. Moreover, currently there is a new challenge, the Covid-19 pandemic, in the agile context whose implications on the daily work and communication of a GVT have yet to be explored. The lack of recent studies together with the suggestion to investigate communication in a more complex context arise the need to explore communication nowadays in the context of agile.
1.3 Research question and contribution
The main purpose of this study is to contribute to the knowledge of today’s internal communication in global virtual software development teams who are working in the agile context and, thus, using agile methods to manage their projects.This context was chosen because agile management provides practices and principles that are beneficial for software development since it promotes communication among team members. Therefore, this study provides an insight into the understanding of some factors which positively and negatively affect internal
communication in global virtual teams from software development in the context of agile. Those factors concerning communication in GVTs are drawn from theory in Chapter 2. As part of the understanding of the influences of those factors, the ways on how they interconnect and are handled to improve internal communication are explored. Moreover, since a new challenge in the agile context just emerged, the Covid-19 pandemic, the effects on communication and the daily work in GVTs are explored.
To address this scope, this study raises the following research question:
RQ: How do factors of global virtual software development teams influence internal communication in the context of agile?
The findings of this study provides a contribution to both practical and theoretical knowledge. They contribute to practice by providing software development teams and managers with a deeper understanding of the dynamics of virtual communication in GVTs. Moreover, a better overview of the use of agile practices in the process of boosting communication and collaboration in teams is given. Theoretical knowledge is enriched by the identification of factors and concepts related to communication in global virtual software development teams. Additionally, by investigating the implications of the Covid-19 pandemic on communication and daily work in GVTs, the study aims to also provide a practical contribution on communication challenges and benefits for other GVTs and collocated teams.
2 Literature Review
2.1 Organizational communication
Organizational communication has been defined as the “nerve system that makes organizations and organizational units cohere and permit their members to coordinate all work efforts” (Kratzer, Leenders and Van Engelen 2009, p. 676). In the organization, the internal communication occurs that is often associated with the employee communication (Tkalac Verčič, Verčič, and Sriramesh, 2012; Velentzas and Broni, 2014). In this context, employees can benefit from several types of communication (Korkala and Maurer cited in Yagüe et al., 2016). Specifically, communication has been classified in several types and techniques: informal versus formal and synchronous versus asynchronous (ibid.). Informal communication can be defined as spontaneous communication, naturally occurring, and results from the individuals’ needs such as the necessity of both having a sense of belonging and connectedness within a social group and feeling safe in an environment (Kratzer, Leenders and Van Engelen, 2009). On the contrary, formal communication is imposed or mandated communication “representing the legitimate authority of the organization and is reflected by the organizational chart” (Aldrich, 1976 cited in Kratzer, Leenders and Van Engelen 2009, p. 676). Regarding the synchronous communication, it can be described as the instant communication between two individuals and it can occur
face-to-face or through the phone, chat rooms and video-communication channels (Hockmann, Knoell and Leiss, 2009; Journell et al., 2014). In this case, when a sender is transmitting a message to a receiver, they “are using communication devices at the same time enabling an immediate ‘conversational’ response to be given, as if they were in the same room at the same time” (Cox and Hollyhead 2008, p. 906). In contrast, the asynchronous communication occurs at different times. This happens through a technology system (computer and communication network) such as emails, online forums and short message service (SMS) text messaging (Cox and Hollyhead, 2008; Hockmann, Knoell and Leiss, 2009).
2.2 Definition and characteristics of global virtual software development teams
In today’s global business environment, software project development has increasingly become geographically distributed, leading to a decrease of in-person interactions among the developers (Cusumano, 2008). Most of the software development companies are adopting virtual teams for their global software development projects (Oshri, Kotlarsky, and Willcocks cited in Denhere, Hörne and Van der Poll, 2015). Virtual software development is the “development of software that involves development teams residing in different countries collaborating and performing a variety of tasks, such as planning, controlling, coding, testing, implementation and maintenance, in order to develop new software products and achieve common project goals” (Tjørnehøj et al. 2014, p. 114).
GVTs are facing several difficulties due to the spatial, temporal and socio-cultural distances that are believed “to challenge project processes such as communication, coordination and control” (Damian and Lanubile 2002, p. 1). Spatial distance refers to the geographical dispersion which separates team members across different locations (Ågerfalk et al., 2008). This distance cuts out the possibility for having in-person communication and consequently, it is difficult for team members to create cohesion (Hossain, Bannerman and Jeffery, 2011). Temporal distance is a measure of temporal dispersion experienced by team members that want to interact (Ågerfalk et al., 2008). Temporal distance means that team members can work in different time zones and time shifts reducing the possibility of synchronous communication which can negatively influence the collaboration and the overall results (Hossain, Bannerman and Jeffery, 2011). Social-cultural distance is described as the measure of a team member’s understanding of another member’s morals and the rules of behavior and the effort required by all team members to comprehend the national and organizational cultures of the different office locations (Ågerfalk et al., 2008). Culture can affect how individuals interpret a specific situation and how they react to it (Kotlarsky and Oshri, 2005 cited in Holmström et al., 2006); and this may differ depending on the diverse backgrounds of team members. Therefore, GVTs can be defined as “a temporary, culturally diverse, geographically dispersed, electronically communicating work group” (Jarvenpaa and Leidner 1999, p. 792) that, despite their spatial and temporal distances, works
together towards the same goals and objectives (Batarseh, Daspit and Usher, 2018). This is possible thanks to globalization and technology developments that allow teams to go beyond the geographic and cultural divisions and work collaboratively (McLarnon et al., 2019).
Organizations started to use virtual teams to solve the problem of the lack of sufficient skilled developers at the location where they operate; to overcome this issue they hire competent software developers in different geographical areas of the world, often at a lower cost (Denhere, Hörne and Van der Poll, 2015). In this way, each member of the virtual teams brings their own competence and expertise to the collective knowledge of the organization that is needed to successfully perform group problem-solving tasks (ibid.). Moreover, when companies expand their businesses across borders, it is common for them to use global virtual teams (Gheni et el., 2016). Furthermore, GVTs in software development extensively rely on computer-mediated communication (CMC) technology which allows geographically dispersed team members to communicate efficiently and work collaboratively (Jarvenpaa and Leidner, 1999). For this reason, the communication between GVTs can be described as virtual communication. However, the undefined environment of online collaboration, together with asynchronous communication might lead to delays and disappointments in the interactions (McLarnon et al., 2019). In fact, it was found that technology-mediated communication can decrease teams’ decision-making effectiveness and hinder project success (Driskell, Radtke, and Salas, 2003; O’Neill et al., 2016).
2.3 Communication in global virtual software development teams
Communication is crucial for software development teams to ease knowledge transfer fast among members, to allow members to comprehend requirements from customers, and to support team members in performing the tasks in the most efficient way (Pikkarainen et al., 2008). Communication frequency and communication quality have been found to be precursors for the GVT effectiveness and team performance (Chang, Hung and Hsieh, 2014; McLarnon et al., 2019). In fact, communication frequency refers to “how much communication occurs among team members'' whereas communication quality refers to “the degree to which the communication among team members is accurate and understood” (Marlow, Lacerenza, Salas, 2017, p. 578). Communication quality has been defined as “the extent to which communication among team members is clear, effective, complete, fluent, and on time” (González-Romá and Hernández cited in Marlow, Lacerenza, Salas 2017, p. 578). A study found that communication should increase in order to improve activity and interaction among team members (Cummings, Espinosa, and Pickering, 2009). It is important to emphasize that frequent communication can be difficult in GVTs due to the fact that members are geographically, temporally and culturally separated (McLarnon et al., 2019). The absence of frequent communication can raise issues related to team coordination, lowering the achievements and the results (ibid). When analyzing the studies on GVTs, some concepts which positively and negatively influence communication
emerged and are further described in the next subsections. There is extensive literature concerning GVTs, therefore the most frequent concepts have been taken into account.
2.3.1 The use of technology and the computer-mediated communication
“Technology is at the core of virtual teams … without internet, email, video conference and audio bridges, virtual teams can't even exist” (Daim et al. 2012, p. 200). In fact, it is possible to consider the advancement of the internet and technology as the “foundations of the GVT” and communication, that is an essential part of all team-working, as their “cement” (Morgan, Paucar-Caceres and Wright 2014, p. 6). Members of a GVT are able to communicate and share information across space, time, cultures and organizational boundaries mainly through information and communications technology (ICT) (Daim et al., 2012). However, communicating through ICT can show some difficulties since it requires understanding the benefits and the limitations of technology and how to develop online relationships (Grosse, 2002). For example, the textual chat is ideal for informal and spontaneous communication (Daim et al, 2012). However, studies done by Daim et al. (2012) show that compared to emails, chats can often make people perceive more the spatial and temporal distance because of the synchronous nature of the chat. The members of GVTs have to comprehend how to choose the more appropriate communication channel for specific tasks and, at the same time, they “need to be sure to balance distance work with face-to-face communication” (Grosse 2002, p. 22). Team members can use several technologies to communicate with each other; among those there are e-mail, instant messaging, and video calls and teleconferencing (Ågerfalk and Fitzgerald, 2006; Fiore, 2003). Sometimes technical problems, connected with CMC, can occur such as delayed or failing audio or video signals and slow internet speed (Krawczyk-Bryłka, 2016). This affects virtual collaboration negatively since it raises misunderstandings and delayed feedback among team members. However, it was found that online collaboration can boost team members’ creativity and satisfaction (Magnusson, Schuster and Taras, 2014).
Furthermore, non-verbal communication such as facial expressions, hand gestures and the quality or the tone of the voice are often missing in CMC and it might reduce team collaboration (Fiore, 2003; Hussain, 2018). On the one hand, non- verbal communication is very important in order to fully understand the meaning of a statement. On the other hand, the lack of non-verbal communication decreases miscommunication due to cultural differences (Daim et al., 2012). In fact, there are different meanings of non-verbal communication depending on the respective cultural origins (Sudhiir and Sudhiir, 2016).
Moreover, since the information flow exchanged in virtual teams is technology-mediated, the rich visual, auditory and social array of cues, which are commonly used in teams’ interactions, are attenuated in distributed settings (Fiore et al., 2003). This leads to ambiguity and misinterpretations. For example, hand gestures were found to be valid cues to communicate
significant information among team members in certain task-related activities (ibid.). However, in distributed environments, gestures can be misunderstood due to the lack of context where that cue is normally used (ibid.). Furthermore, a study by Ambadar (cited in Fiore et al., 2003) showed that face perception has to be well reproduced to avoid losing important information during the interaction.
CMC involves both asynchronous interactions through “a collaborative workspace, as well as e-mail” and synchronous interactions such as “desktop videoconferencing, shared workspace, chat and other features” (Ebrahim, Ahmed and Taha 2009, p. 2656). Asynchronous communication has been criticized for not being a good way of communicating, but it was found that it can be beneficial for collaboration in GVTs. For example, when it comes to brainstorming or making progress on a task, team members usually wait for meetings. However, if the team members work asynchronously, they can freely decide when to make their contributions and express their opinions (Malhotra, Majchnak and Rosen, 2007). This allows all members of the team with different backgrounds to express their ideas at their own pace. Therefore, combining both synchronous and asynchronous communication will provide the best set up for team members, leading to an increase of their performance and personal satisfactions (ibid.).
2.3.2 Cultural diversity
Culture plays a major role in GVTs in software development because those teams are formed by individuals who come from countries with different cultures. There are many definitions of culture which generally embrace shared values, beliefs, expectations, customs and rituals. For this study, the following definition developed by Spencer-Oatey (2004, p. 4) was considered: “culture is a fuzzy set of attitudes, beliefs, behavioral norms, and basic assumptions and values that are shared by a group of people, and that influence each member’s behavior and his/her interpretations of the ’meaning’ of other people’s behavior”.
Culture is a multifaceted construct that can be examined from several perspectives; among those there are the national culture and the organizational culture (Shachaf, 2008). National culture refers to ethnic, racial, gender, and other demographic characteristics (Dafoulas and Macaulay, 2002), while organizational culture is defined as “a collection of shared key values, symbols, meanings, beliefs, assumptions, and expectations among organizational members'' (Sriramesh, Grunig and Buffington 1992, p. 591). Organizational culture has been considered as a crucial factor affecting internal communication (Men and Jiang, 2016). According to Hofstede (2011, p. 1), organizational cultures “reside in visible and conscious practices: the way people perceive what goes on in their organizational environment”. Additionally, some independent dimensions, which outline most of the assortment in organization practices, were found (Hofstede, 2011). Among those dimensions, there is the “process-oriented versus results-oriented” which means that “process-oriented cultures are dominated by technical and bureaucratic routines, results-
oriented by a common concern for outcomes” (Hofstede 2011, p. 20). Another organizational culture dimension is the “open systems versus closed systems” which refers to the “common style of internal and external communication, and to the ease with which outsiders and newcomers are admitted” (ibid.). While identifying this dimension, the researcher found that organizational cultures are influenced by national culture differences. According to Daim et al. (2012), national differences of individuals influence their behavior embedded in their own culture.
Regarding national cultural studies, they stress that the cultural diversity may develop communication challenges among people from different cultural backgrounds. One of these studies was done by Geert Hofstede who developed in 1980 the five-dimensional model of culture by focusing on the following categories: power distance, collectivism-individualism, masculinity-femininity, uncertainty avoidance and long term-short term orientation (Hofstede, 1994). Power Distance is “the degree to which the culture believes that institutional and organizational power should be distributed unequally and the decisions of the power holders should be challenged or accepted” (Lustig and Koester 2010, p. 114). For instance, those who come from cultures with a larger power distance, like India, tend to avoid public confrontations and often wait to be told what to do (Binder, 2007). On the other hand, team members coming from a smaller power distance, such as Sweden, expect to be treated as equal (ibid.). The Uncertainty Avoidance dimension describes “the extent to which members of society feel uncomfortable with ambiguity” (Hofstede 2011, p. 10). In fact, based on the level of this dimension, people are more or less open to accept tight supervision of their work, to accept risks and solve difficult cases (Lustig and Koester, 2010). Consequently, project managers should consider the background of the team members when assigning tasks and giving instructions to achieve a more effective result (Binder, 2007). Cultural values, therefore, highly influence the way people interpret different situations, affecting their decisions (Hofstede cited in Wildman and Griffith, 2014). The other dimensions are: Individualism versus Collectivism which relates to “the integration of individuals into primary groups”; Masculinity versus Femininity that refers to “the division of emotional roles between women and men”; Long Term versus Short Term Orientation that considers “the choice of focus for people's efforts: the future or the present and past” (Hofstede 2011, p. 8). Project managers need to consider all the characteristics of cultural dimensions when planning activities, communicating, discussing projects and personal progresses with the team members (Binder, 2007). It is important to emphasize that culture is complex and these few dimensions are not able to provide a full explanation of that complexity (Shachaf, 2008). However, recognizing the cultural characteristics gives the tools to the project manager to identify and handle potential communication challenges and to each team member to be aware of cultural differences and communicate efficiently.
2.3.3 Intercultural communication
Intercultural communication occurs when people from different cultures interact with each other and it is referring to the effect on communication behavior caused by this interaction (Arasaratnam, 2013; Chitakornkijsil, 2009). In fact, virtual teams are characterized by cultural diversity (Krawczyk-Bryłka, 2016), which can be both challenging and rewarding with different values and communication styles offering valuable solutions by combining complementary abilities (Smith and Blanck, 2002). Particularly, one of the main advantages of cultural diversity in a GVT is the presence of “diversity of thought, sharing and innovation that comes together, through dispersed groups”. In addition, it was found that cultural diversity in virtual teams brings better decision-making since team members can use different capabilities and knowledge whilst developing collaborative solutions that facilitate the acceptance of new ideas (Shachaf, 2008; Zakaria et al., 2004). However, by being geographically dispersed GVTs are facing cultural differences which can negatively affect the overall success of team performance (Daim et al., 2012). In fact, members of a team can have diverse verbal and non-verbal communication styles and make decisions differently depending on their cultural background (Gudykunst and Ting-Toomey cited in Shachaf, 2008).
Moreover, language, which is closely connected with culture (Kashima and Kashima, 1998), can present some barriers to cultural expectations and understandings that have a negative impact on team performance (Daim et al., 2012). According to a study by Shachaf (2008), these differences created difficulties in understanding both written and spoken language. Consequently, team members have to “invest more time and effort in encoding and decoding messages'' and the cost of interaction, in terms of time and energy, for both senders and receivers is higher compared to team members with the same cultural background (Shachaf 2008, p. 134). In addition, Chitakornkijsil (2009) emphasized that the main barriers to intercultural communication become ignorance and lack of knowledge when a team member is not open to understand another team member’s culture. Therefore, communicating effectively requires understanding each other’s cultures and being willing to make efforts.
2.3.4 Interpersonal trust
Trust and communication are two distinct behavioral constructs; nonetheless in GVTs, they often “play out together” (Sarker et al. 2011, p. 284). Morgan and Hunt (1994, p. 25) suggested that communication is “a major precursor of trust … [since] it fosters trust by assisting in re- solving disputes and aligning perceptions and expectations”. Trust in a team context is the mutual belief to have the competence and the willingness to engage together towards the same goal (Boies, Fiset and Gill 2015). In GVTs, building interpersonal trust among team members can be complex because of the substantial dependence on a collaborative relationship with technology (Chang, Hung and Hsieh, 2014). Therefore, trust in virtual teams is achieved by several types of
communication, using mainly technological tools to mutually engage (Sarker et al., 2011). There are several factors which can boost trust such as the presence of a positive team climate, the possibility of frequent team communication and the opportunity for personal competence growth (Holton, 2001). Furthermore, it was suggested that “shared social norms, repeated interactions, and shared experiences” ease the creation of trust (Jarvenpaa and Leidner 1999, p. 792).
Mutual trust is important for building connected teams where each member is putting effort to provide a solid contribution to the workgroup and take decisions with good intentions (Zaccaro and Bader, 2003). Therefore, interpersonal trust can be defined as “the degree of intention or willingness to depend on other virtual team members” (Chang, Hung and Hsieh 2014, p. 1321). However, members in GVTs can find barriers that impede the development and improvement of interpersonal trust because of the spatial, temporal, and cultural distances. Several studies emphasized that in-person communication is irreplaceable for both building trust and regaining back the lost trust (Jarvenpaa and Leidner, 1999). Therefore, in the context of virtual teams, the lack of face-to-face encounters, and thus communication cues, can have a negative impact on interpersonal trust. For instance, members can find difficulties in having quick responses and maintaining long-term relationships (Daim et al., 2012).
The Social Identity Model of Deindividuation Effects (SIDE) theory suggests that “in the absence of individuating cues about others, as is the case in CMC, individuals build stereotypical impressions of others based on limited information” (Lea and Spears cited in Jarvenpaa and Leidner 1999, p. 793). However, some studies indicate that communication exchanges among team members through the electronic space over time leads also to trust development (Jarvenpaa and Leidner (1999), Daim et al, 2012). According to a case analysis conducted by Jarvenpaa and Leidner (1999), communication behavior that eases trust, during the initial stage of the formation of a team, is “social communication” and “communication of enthusiasm”. The members of the team can help develop this trust by handling technical uncertainties and taking initiatives (Daim et al., 2012). In the next stages of the maturity of the team, the communication behaviors, which are needed to maintain trust in a group, are “predictable communication and substantial and timely responses” (Jarvenpaa and Leidner 1999, p. 807). Ridings et al. (cited in Sarker et al. 2011, p. 284) found that communication “in the form of responses to posts of remote members, sharing of personal information, and so forth can help to increase the trustworthiness of the individual”. Moreover, in each team there is often a manager who has a key role in building and maintaining mutual trust through a clear explanation of responsibility and role to individuals. Furthermore, their work is to constantly follow the people in the activity to secure the right performance and to achieve the goals (Daim et al., 2012).
2.3.5 Interpersonal relations
As stated by Wildman and Griffin (2014), informal interactions highly contribute to building trust and, consequently, to developing strong relationships. Therefore, interpersonal trust is helping team members to establish durable relationships. However, when this is not possible, it becomes more difficult for team members to bond and to build interpersonal relations (Wildman and Griffin, 2014). In fact, GVTs can face challenges to build effective interpersonal relationships among team members since they are geographically dispersed (Daim et al., 2012). Moreover, they rely on technologies which can lead to misunderstandings (ibid.). Team diversity can be an obstacle in developing solid relationships between the members; a study suggests that “the greater the team member diversity, the more time will be required for team members to form strong bonds” (DeSanctis and Poole cited in Jarvenpaa and Leidner 1999, p. 793). However, this is not always the case, and some teams may build strong relationships and trust regardless of the heterogeneity and short time spans (ibid.).
2.4 Agile project management in software development 2.4.1 Agile software development
Software organizations are part of highly dynamic businesses as they have to constantly adapt to the unexpected changes in the market, new customer needs and advances in technology (Lycett cited in Pikkarainen et al., 2008). To respond to those challenges, software teams have been using agile project management to handle their software development projects (Atlassian, 2020). Agile software development methodologies are “based on iterative development, where requirements and solutions evolve through collaboration between self-organizing cross-functional teams” (Shrivastava and Date 2010, p. 10). They are based on the Agile Manifesto which was collaboratively written in 2001 by some developers who used the word “agile” as an “umbrella term” referring to numerous iterative and incremental methods (Hoda, Noble and Marshall 2011, p. 2). According to the Agile Manifesto, agile development “emphasizes individuals and interactions over processes and tools, working software over comprehensive documentation, customer collaboration over contract negotiation, and responding to change over following a plan” (Agile Alliance, 2001). Several organizations consider agile methods as a way to overcome the problems they face in software development, such as the fact that the time for the development of the software is too long, the costs are too high and there are quality issues upon delivery (Holmström et al., 2006). Therefore, agile methods, such as Scrum, are adopted to increase the speed of the software development, and thus improving communication and collaboration within members of agile teams, as well as among agile teams, customers and business units (Schwaber and Beedle 2002; Pikkarainen et al., 2008).
Furthermore, as mentioned above, user requirements of a project can change over time and agility allow the software teams to efficiently and effectively react and incorporate those new
requests during the project life cycle (Alzoubi, Gill and Al-Ani, 2016). Additionally, due to the geographically dispersion of members in software development teams, the visibility of the project status can be limited (Shrivastava and Date, 2010). By using agile methods, however, which are based on short interactions and where the development is done in cycles, it is feasible to identify possible issues from the initial stages of the project (ibid.). Moreover, internal communication is positively influenced by agile principles since it allows members to keep track of the short-term goals (ibid.). Agile principles can also increase trust between diverse cultures involved in the process by frequent communication (Paasivaara and Lassenius, 2006). It has been discovered that the “quality of software and communication is improved, and communication and collaboration are more frequent than before because the Scrum methodology is used in the project” (Paasivaara, Durasiewicz and Lassenius 2009, p. 196). In fact, among the positive effects of improved communication in this context there is a higher software quality and a reduction of number of defects (Hanakawa and Li cited in Hummel, Rosenkranz and Holten, 2013). Additionally, it was found that motivation of team members increased by using agile methods (ibid.).
In agile software development, as also emphasized above, communication within the development as well as with the customers plays a central role; in particular, informal and spontaneous communications are promoted (Yagüe et al., 2016). Customers and developers should collaborate on a daily basis and the information regarding the project should be shared by informal communication tools “such as face- to-face conversation rather than through formal documentation, plans, and models” (Hummel, Rosenkranz and Holten 2013, p. 344). Moreover, it has been shown that “the higher the communication frequency, the more productive the project is” (Braithwaite and Joyce cited in Hummel, Rosenkranz and Holten 2013, p. 347). Therefore, agile in GVTs, by promoting frequent communication and offering useful practices, has been demonstrated to be valuable for a project’s quality, performance and overall for its success (Shrivastava and Date, 2010; Mishra and Mishra 2009; Pikkarainen et al., 2008).
2.4.2 Agile Scrum methodology and internal communication
One of most popular agile methods is the Scrum methodology which is “an iterative, incremental and rapid process for software project management” (Holzmann and Panizel 2013, p. 69). Scrum facilitates the teams to respond rapidly and effectively to change by providing techniques and practices for project management (Sliger, 2011). Frequent communication among team members is promoted through Scrum activities, such as daily stand-up meetings, to ensure that all the software development activities are in place (Schwaber and Beedle, 2002). It has been discovered that the “quality of software and communication is improved, and communication and collaboration are more frequent than before because the Scrum methodology is used in the project” (Paasivaara, Durasiewicz and Lassenius 2009, p. 196). In fact, among the positive
effects of improved communication in the agile context there is a higher software quality and a reduction of number of defects (Hanakawa and Li cited in Hummel, Rosenkranz and Holten, 2013). Additionally, it was found that motivation of team members increased by using this agile method (ibid.).
Communication is considered one of the most valuable aspects of Scrum methodology (Holmström et al., 2006), which provides four process activities called Scrum ceremonies named Sprint planning, Daily Scrum, Sprint review, Sprint retrospective as shown in Table 1.
Table 1. Scrum ceremonies adapted by Schwaber and Beedle (2002)
As demonstrated by the Scrum methodology, a higher communication frequency and communication quality are involved since team members are constantly reminded to discuss the project and work together (Paasivaara et al., 2009). Several studies have found positive correlation between the Scrum ceremonies and communication among team members (Pikkarainen et al., 2008; Wijnands and Dijk, 2007; Ramesh et al., 2006; Paasivaara, Durasiewicz and Lassenius, 2009). For instance, the Daily Scrum has been found to constitute the most important tool for communication in Scrum as it has a positive effect on internal communication (Pikkarainen et al., 2008). In fact, during the Daily Scrums, communication has been described as the “glue” which connects all team members (Ramesh et al. cited in Hummel, Rosenkranz and Holten, 2013, p. 348). Moreover, it is important that all team members participate in the meetings in order to prevent misunderstandings (Wijnands and Dijk, 2007). This practice has also a positive impact on team communication frequency since it encourages one-to-one communication after the meetings (Paasivaara, Durasiewicz and Lassenius, 2009). Concerning the sprint planning, it is crucial both to share information about features and requirements with customer representatives and to ease informal communication among the team members (Pikkarainen et al. 2008). Finally, the sprint review and the sprint retrospective have been found to play an important role since they both encourage communication, sharing of experience and feedback between teams (Paasivaara and Lassenius 2010).
2.5 Communication theories in project management
Communication influences the relationship among the members in project teams since it improves sharing of opinions and avoids misunderstandings, which might lead to the failure of a project (Andrade Rodríguez, 2017). Some communication theories have been identified in relation to improving the success of project management. Among those, there is the Social Information Processing Theory (SIPT) which provides a better understanding on how online relationships can be developed, and the Communication Accommodation Theory (CAT), which explains that when individuals communicate with another person they adapt to the person that they are interacting with (ibid.).
2.5.1 Social Information Processing Theory
SIPT is an interpersonal communication theory that was developed by Walther (1992). It describes how online interpersonal communication can develop “without nonverbal cues and in which way people develop and manage relationships in a CMC environment” (Walther 1992, p. 56). According to this communication theory, “given time and opportunity to interact, relationships between individuals can form in online environments” (Olaniran, Rodriguez and Williams 2012, p. 45). Walther (1992) drew up six theoretical propositions based on five assumptions, summarized by Sumner and Ramirez (2017, p. 4) as follows:
1. “Online communication takes longer to facilitate personalized relational communication 2. The presence and tone of relational communication depends on the number of messages
exchanged
3. The nature of relational communication differs between initial interactions and later interactions
4. The changes in relational communication take longer to develop via CMC than via in-person interaction
5. During later periods of interaction, CMC and in-person communication are of similar relational tone, assuming communicators have had ample time to exchange messages and form impressions”.
When it comes to interacting in online settings, the development of an interpersonal relationship may need more time to develop compared to in-person relationships (Jarvenpaa and Leidner, 1999). However, online interactions are a crucial communication tool among people who come from different backgrounds and are placed in different locations, especially in the current times (Lewis, 2007 cited in Andrade Rodríguez, 2017). In fact, globalization has led to the development of project teams composed by members from diverse countries (ibid).
There are several studies which argue that online communication deteriorates the ability of individuals to communicate. However, the SIPT argues that online interpersonal relationships “may demonstrate the same relational dimensions and qualities as in-person relationships over time” (Walther 1992, p. 57). In fact, technologies used in CMC increase the pace of communications by changing the perception of time and space and reducing their differences
(Tsatsou, 2009). However, as emphasized by Tsatsou (2009), CMC through ICT alters the dimension of time in the work setting and in the social interactions since it generates rigid time patterns for scheduling activities.
Nevertheless, the SIPT affirms that “online communication can be more effective than in-person communication” (Andrade Rodríguez 2017, p. 48). Information through CMC is transferred quicker and contemporary to all the members of the team regardless of the time or place of communication (ibid.). However, it was argued that through CMC less information is transferred in comparison with in-person exchanges (Spears and Lea, 1992; Walther, 1995). This is because CMC systems need more interactions to transfer the same amount of information (Olaniran, Rodriguez and Williams, 2012).
2.5.2 Communication accommodation theory
CAT was developed by professor Howard Giles in the 1970s and it highlights the adjustments people undertake while interacting (Gallois, Ogay and Giles, 2005). According to the CAT, when people communicate, they “adjust their vocal patterns and speech to accommodate the needs of the recipient of the message” (Mooz, Forsberg, and Cotterman cited in Andrade Rodríguez 2017, p. 46). Moreover, beside the adjustment through verbal communication, people can modify their gestures according to the person they are communicating with (Gallois, Ogay and Giles, 2005). The theory focuses on how people “minimize” communication differences by taking into account the context, the language, and the identity (Andrade Rodríguez 2017, p. 46). “The context and identity determine the language and manner of communication” (ibid.). Moreover, CAT can be used to increase the quality of shared information when members of intercultural teams communicate during decision-making processes (Aritz and Walker, 2010).
Two types of accommodation process are considered, namely convergence and divergence (Gallois, Ogay and Giles, 2005). The converge process occurs when people adapt to the other person’s way of communication in order to decrease the social differences (ibid). In this context, the “downward convergence” is when people can use several ways to grab others’ attention by adopting the same gestures or words and by talking about common subjects (Giles and Soliz 2014, p. 159). On the contrary, the divergence process uses the “upward convergence” showing an opposite approach to communication where people tend to approach others by using complex wording to mark a threshold between the speakers (ibid.). The recipients accept positively the speech convergence since they attribute a positive intent to the resultant behavior whereas they unfavorably evaluate the speech divergence since they perceive it with a negative intent (ibid.). According to the CAT, “people adjust speech, vocal patterns, and gestures to help promote mutual understanding in communication” (Gallois and Giles, cited in Brown 2018, p. 56). In the organization context, CAT can be taken into account to communicate effectively among the employees with different cultural backgrounds.
2.6 Theoretical framework
In the current times, there is the need to explore the role of communication in GVTs in the agile context. Therefore, an analysis of what defines GVTs in software development teams in literature has been scrutinized. Additionally, the agile management context has been described together with Scrum practices and its impact on communication. It has been shown that agile practices are beneficial for promoting communication in GVTs. However, some factors of GVTs can challenge communication among team members. From the literature review the most frequent concepts of GVTs emerged namely ICT, CMC, cultural diversity, intercultural communication, interpersonal trust and interpersonal relations. In the context of project management, two theories have been found: SIPT and CAT. Studies have shown that these characteristics and theories affect communication in GVTs. Therefore, they have been considered as factors that affect internal communication in global virtual software development teams. GVTs are geographically dispersed and their internal communication is mainly virtual. As some concepts concern the same areas, they have been grouped into three main factors: Culture, Technology and Trust. For each factor, the concepts found in literature were divided. It was found in the literature that these concepts influence internal communication in GVTs positively and negatively, for this reason the plus and minus signs were added. Moreover, the three factors can influence each other according to literature; thus, double arrows were added to highlight their interrelations that ultimately affect communication.
A theoretical framework based on different scientific articles for this study has been created, as shown in Figure 1, by taking into consideration the main communication factors of GVTs in the agile context. Moreover, in the agile context the Covid-19 pandemic is taken into account since it is affecting teams in the way of working and communicating.
3 Methodology
3.1 Research and Methodological Approach
The aim of this study is to get a deeper understanding of how factors of GVTs impact communication among the members of a global virtual software development team in the context of agile. An abductive approach, which is a combination of inductive and deductive research, was considered the most suitable selection for this study. According to the abductive approach, new information can be created and combined with existing knowledge (Gioia, Corley and Hamilton, 2012). In this study, existing knowledge provided a deeper understanding about the existing factors affecting communication in GVTs from software development.
The research approach of this study is developed using a qualitative method in order to understand the phenomenon of communication in GVTs from software development in the agile context. This approach was selected because it is considered to be appropriate for studies that explore social phenomena thanks to its flexibility in the structure and in the results (Azungah, 2018). Qualitative research tries to comprehend a specific research issue or topic from the perspectives of the people it involves and thus it is effective “in obtaining culturally specific information about [their] values, opinions, behaviors, and social contexts” (Mack et al. 2005, p. 1). The main advantage of qualitative approach is its capability to offer textual descriptions about the ways in which individuals experience a specific research problem (ibid.). Information provided by the descriptions considers the “human side” of the issue that is “the often contradictory behaviors, beliefs, opinions, emotions, and relationships of individuals” (ibid.). In this case, textual descriptions provide information on a team members’ similar and differing opinions and experiences towards communication.
3.2 Research design
3.2.1 Single case study
In order to analyze internal communication in a GVT, a single case study approach was considered applicable. Moreover, the study aims to give a better understanding of factors of communication in the context of agile which have not been fully explored in literature. According to Yin (2018), a case study is considered a good approach to gain a deeper knowledge of a phenomenon that has not been well-explored. In this case, communication needs to be explored in the context of agile, which is the setting where global virtual software development
teams work nowadays. Moreover, in this context a new challenge appeared, Covid-19 pandemic which has not been explored yet.
Furthermore, according to Yin (2018), case studies are appropriate when the RQ require an extensive and in-depth explanation of social phenomena. Additionally, case study research is “a comprehensive method that incorporates multiple sources of data to provide detailed accounts of complex research phenomena in real-life contexts” (Morgan et al. 2016, p. 1060). In fact, the more the RQ aims to explicate contemporary situations, such as “how” or “why” some social phenomena act, the more the case study approach is suitable (Yin 2018, p. 33). For these reasons, a case study was selected for investigating the social phenomenon of communication.
3.2.2 Case company
The case company is a global and multicultural company with more than three-hundred employees, twenty different nationalities and localization capabilities. It is a leading provider of products and services for content moderation of online marketplaces worldwide. Content moderation is used to filter, through specific criteria, the content generated by users on web platforms. The case organization is a Swedish company that has a global distributed organization. The participants of the study are members of a global virtual software development team who work at the case company using agile methods, in particular the Scrum methodology. The company was considered suitable for this paper as the team case is dispersed in different locations and the team has been implementing agile methods for a significant period of time. In fact, the team has implemented the agile Scrum methodology since 2017 and it is distributed in three cities in different countries: Stockholm in Sweden, Paris in France and Iasi in Romania. The Scrum team consists of ten people with the Scrum Master based in Stockholm. It is important to emphasize that the opinions and experiences of the team members are influenced inevitably by the agile context where they work in.
3.3 Methods and data collection techniques
Case studies usually combine data collection techniques such as interviews and observations (Eisenhardt, 1989). On one side, interviews are undertaken with individual informants whereas the data collected with the observations are from a group of individuals (Goodwin and Horowitz, 2002). Although interviews and observation take diverse approaches, they share the same purpose which is gaining a thorough understanding of the cases that are investigated, in particular the way individuals act and reason (ibid.). Therefore, the data was collected by conducting interviews and observations in order to study the phenomenon of communication more in-depth.
3.3.1 Semi-structured interviews
Semi-structured interviews have been considered as the suitable method for data collection in this study since they allow to gain an understanding of a complex phenomenon (Bryman and
Bell, 2015). This method allows the interviewers to be more flexible when asking questions in comparison to a more structured approach. In fact, the interviewers have the possibility to ask supplementary questions in accordance to the responses (ibid.). Before the actual interview, the author conducted informal meetings with each member of the team to familiarize with them, to introduce the topic of the research and to give a short presentation of the interviewer. During the actual interview, the purpose of the study and the structure of the interview were explained. All the informants agreed to record the interviews after the author informed and ensured that all recordings would be treated confidentially and that their names would have been kept anonymous.
The language used in the interviews was English in order to have a shared understanding since the interviewees have multiple nationalities and also because it is the team’s working language. The interviews were conducted by calls or video calls on a platform called Slack that the team uses to communicate with each other. The interviewees are ten people and they were contacted virtually since they are dispersed geographically. The team is composed by the Scrum Master, Product Owner, and the development team which is divided into back-end developers, development and operations (DevOps) engineers, front-end developers and a designer. Table 2 gives an overview of the informants who have been assigned a letter to ensure anonymity when quoting or referring to their opinions in the next sections. The opinions and experiences of each member to the team gave an insight related to the research topic and what factors affect internal communication in this context. Moreover, interviews were done with all the components of the team, including people in a managerial position (Scrum Master and Product Owner), to gain a complete overview of the communication phenomenon from the perspective of managers and the developers. Informant A is also the manager of the team.
Table 2. Overview of the interviewees
Participants Role Based in Interview Time
Informant A Scrum Master Sweden Call 60 min
Informant B Product Owner France Call 60 min
Informant C Development team member with some
responsibilities as Product Owner France Video call 50 min Informants D Development team member Sweden Video call 25 min Informant E Development team member Romania Video call 30 min
Informant F Development team member France Call 35 min
Informant H Development team member Sweden Video call 25 min Informant I Development team member Sweden Video call 30 min Informant L Development team member France Video call 30 min Informant M Development team member France Video call 30 min
3.3.2 Interview guide
In order to address the RQ, How do factors of global virtual software development teams influence internal communication in the context of agile?, the theoretical framework was broken down into three factors: Technology, Culture and Trust and each factor incorporates specific concepts. The factors provided a foundation for the primary data collection and were used to develop the questions for the interview guide. The interview questions were formulated with the scope of gathering a deeper understanding on how team members from software development communicate and how the factors Technology, Culture and Trust are influencing their perception of internal communication and interrelations between those factors that ultimately affect communication. Since these factors are immersed in the agile context, explicit questions were addressed on this matter; in particular, to understand the opinions on the link between Scrum and internal communication and its potential connections with the communication factors. Therefore, in order to prepare the questions for the interviews, an operationalization of the theoretical framework has been done, as shown in Appendix A. The complete interview guide is displayed in Appendix B. The first part of the interview is composed of four general questions regarding the role of each member in the team and their experience in GVTs in the context of agile. Team members that have managerial roles (Scrum Master and Product Owner) have been asked different general questions to understand the story of the team over the years and how they have implemented Scrum methodology. The second part has been divided in three sections. The third part concerns the agile context more specifically the participants’ opinions on the Scrum ceremonies linked to internal communication and its factors and the influence of Covid-19 pandemic on their daily work.
3.3.3 Observations
Observations were conducted to analyze how the team members interact with each other and to investigate the factors that can affect their internal communication. The case company gave the possibility to the author of this study to participate in the Scrum ceremonies during one sprint (two weeks) by using the Slack platform. According to Mack et al. (2005, p. 14), observations are useful “for gaining an understanding of the physical, social, cultural, and economic contexts in which study participants live; the relationships among and between people, contexts, ideas, norms, and events; and people’s behaviors and activities”. Observations enable researchers to analyze what people, in this case the members of a global virtual software development team, actually do “rather than what they say they do” (Caldwell and Atwal cited in Morgan et al. 2016, p. 1061). Moreover, with this method the researcher of this study approaches participants in their own setting trying to acknowledge what it means to be an “insider” whilst being “an outsider” (Mack et al. 2005, p. 13).
Furthermore, the data gathered through the observations are used to check “the participants’ subjective reporting of what they believe and do” (Mack et al. 2005, p. 14). It was also possible to join a meeting on the implications of Covid-19 pandemic for the team members. The team members expressed their feelings and concerns and discussed how to tackle the Covid-19 pandemic. It is important to emphasize that when the observations have been done, it was the middle of March 2020 and it was the second week that the team was working fully remotely due to the Covid-19 pandemic. While conducting the observations at the case company, the author of the study made careful and objective notes. During the meeting, the author took short and quick notes in order to not distract the members of the team in the meetings. After the meeting, those notes were extended and completed. The notes were divided into “my observations” and “my reflections” in order to distinguish between what was happening and what reflections were made related to that.
The researcher of this study played the role of observer as participant. That means that the researcher was known and recognized by the participants (Babchuk, 1962). Moreover, the participants were informed of the aim of the study. The interactions during the observations between the researcher and the participants were limited in order to keep a neutral role. While conducting the observations, the author had the theoretical framework in mind allowing to consider relevant aspects of communication among members of GVTs. Thus, the observations helped to extend the empirical findings and the existing knowledge of GVT communication. However, some concepts of the communication factors, such as culture diversity and interpersonal trust, were hard to detect during the observations.
3.4 Data analysis procedure
The coding methodology for the data collected through semi-structured interviews in this study was inspired by the methodology developed by Corley and Gioia (2004). The process, adapted to this methodology, was classifying responses into specific categories. According to the Gioia methodology, there are 1st order of analysis, 2nd order analysis and aggregate dimensions (Gioia, Corley and Hamilton, 2012). In the 1st order of analysis, there are themes which adhere closely to the participants’ responses, including similarities and differences. In the 2nd order analysis, the themes are developed into concepts that might help the researcher to describe and explain the observed phenomenon (Gioia, Corley and Hamilton, 2012). Those concepts are associated with the ones under each communication factor in the theoretical framework. The 2nd order concepts are further distilled into aggregate dimensions of technology, culture, trust and agile context aligned with the theoretical framework.
This coding was chosen for the strong data structure, which not only helps the researcher to visualize the data but also provides a graphic representation of how the data have been analyzed from the raw data, the responses of the team members, to the concepts and dimensions (Gioia,
