Qualitative research deductive and inductive approaches to data analysis

(1)

Qualitative Research Journal

Qualitative research: deductive and inductive approaches to data analysis Theophilus Azungah,

Article information:

To cite this document:

Theophilus Azungah, (2018) "Qualitative research: deductive and inductive approaches to data analysis", Qualitative Research Journal, Vol. 18 Issue: 4, pp.383-400, https://doi.org/10.1108/QRJ- D-18-00035

Permanent link to this document:

https://doi.org/10.1108/QRJ-D-18-00035

Downloaded on: 30 November 2018, At: 00:49 (PT)

References: this document contains references to 116 other documents.

To copy this document: [email protected]

The fulltext of this document has been downloaded 121 times since 2018*

Users who downloaded this article also downloaded:

(2016),"Sample size for qualitative research", Qualitative Market Research: An International Journal, Vol. 19 Iss 4 pp. 426-432 <a href="https://doi.org/10.1108/QMR-06-2016-0053">https://

doi.org/10.1108/QMR-06-2016-0053</a>

(2011),"Purposeful Sampling in Qualitative Research Synthesis", Qualitative Research Journal, Vol. 11 Iss 2 pp. 63-75 <a href="https://doi.org/10.3316/QRJ1102063">https://doi.org/10.3316/

QRJ1102063</a>

Access to this document was granted through an Emerald subscription provided by emerald- srm:387340 []

For Authors

If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service information about how to choose which publication to write for and submission guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.

About Emerald www.emeraldinsight.com

Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online products and additional customer resources and services.

Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.

Downloaded by Göteborgs Universitet At 00:49 30 November 2018 (PT)

(2)

Qualitative research: deductive and inductive approaches

to data analysis

Theophilus Azungah

Department of Management Studies, Bolgatanga Polytechnic, Bolgatanga, Ghana

Abstract

Purpose–The purpose of this paper is to explain the rationale for choosing the qualitative approach to research human resources practices, namely, recruitment and selection, training and development, performance management, rewards management, employee communication and participation, diversity management and work and life balance using deductive and inductive approaches to analyse data. The paper adopts an emic perspective that favours the study of transfer of human resource management practices from the point of view of employees and host country managers in subsidiaries of western multinational enterprises in Ghana.

Design/methodology/approach–Despite the numerous examples of qualitative methods of data generation, little is known particularly to the novice researcher about how to analyse qualitative data. This paper develops a model to explain in a systematic manner how to methodically analyse qualitative data using both deductive and inductive approaches.

Findings –The deductive and inductive approaches provide a comprehensive approach in analysing qualitative data. The process involves immersing oneself in the data reading and digesting in order to make sense of the whole set of data and to understand what is going on.

Originality/value–This paper fills a serious gap in qualitative data analysis which is deemed complex and challenging with limited attention in the methodological literature particularly in a developing country context, Ghana.

KeywordsQualitative, Emic interviews documents, Deductive, Inductive Paper typeViewpoint

Introduction

The paper explains the rationale for applying the qualitative approach using interviews and documents as the main sources of data in a naturalistic underexplored Ghanaian context to gain a better insight and appreciation of how western multinational enterprises (MNEs) transfer their human resource management (HRM) practices to their subsidiaries in Ghana.

This is a country and a context that has been underrepresented in academic research in international HRM. It is often argued that qualitative data analysis is complex, laborious and time consuming with limited theoretical attention (De Casterleet al., 2012). The process can be frustrating particularly to novice researchers in the absence of clear-cut guidelines on how to analyse qualitative data using particular methods. This paper proposes and discusses in a systematic manner how to analyse qualitative data using both deductive and inductive approaches. The author developed a model to explain how interview and documentary data are analysed using deductive and inductive methods.

The qualitative approach highlights“contextual understanding”with an emphasis that researchers usually seek to understand the“behaviour, values, beliefs, and so on in terms of the context in which the research is conducted”(Bryman and Bell, 2011, p. 411). Using a qualitative approach of inquiry, the paper seeks to combine context sensitivity and rigorous explanations into how and why western MNEs standardise or localise their HRM practices within the cultural and institutional contexts of Ghana. In fact, quantitative approaches have dominated the International Business (IB) field with qualitative methods in a supporting role. Some scholars interpret this as a loss to the field of IB, because qualitative methods can shed light on the complex cultural and institutional settings of MNEs

Qualitative Research Journal Vol. 18 No. 4, 2018 pp. 383-400

The current issue and full text archive of this journal is available on Emerald Insight at:

www.emeraldinsight.com/1443-9883.htm

383 Qualitative research

(3)

(Birkinshaw et al., 2011; Doz, 2011; Gephart, 2004; Piekkari and Welch, 2006; Welch et al., 2011). Due to the nascent state of literature on HRM practices in Africa (Mellahi and Mol, 2015), I argue that qualitative research, in particular, is useful to explore in greater detail how HRM practices are transferred to subsidiaries located in Ghana. It is argued that qualitative research may be more suitable for carrying out research in emerging markets, since some instruments designed and applied in the developed markets may possibly not be appropriate in emerging market contexts (Hoskisson et al., 2000). Thus, a qualitative approach is required in order to provide a rich descriptive, exploratory and explanatory study of the transfer of institutionally distant practices through the emic perspective.

Qualitative research is therefore more likely to provide meaningful contextualisation and clarity to the research questions and concepts within the Ghanaian social setting rather than hypothesising and testing variables. Qualitative research provides insights that are difficult to produce with quantitative measures by providing thick, in-depth descriptions (Gephart, 2004) of HRM practices in real-life contexts that management and employees ascribe to in the Ghanaian socio-cultural context.

To be in a position to give a vivid descriptive detail of standardisation and localisation of HRM practices of subsidiaries of western MNEs, it would be appropriate to adopt an emic perspective or insider view from the standpoint of employees and managers about how these opposing forces interact. Thus an insider view “employs the meanings in use” (Gephart, 2004, p. 455) relating to HRM practices in local affiliates of western MNEs to explain the experiences of management and employees in their routine work lives.

Why qualitative research?

Qualitative researchers regard the world as“complex, dynamic, interdependent, textured, nuanced, unpredictable, and understood through stories […] (they) distrust generalisations and are most comfortable immersed in the details of a specific time and place”(Patton, 2015, p. 13). It employs texts, words and talk to develop concepts that facilitate understanding of social phenomena (Gephart, 2004; Yauch and Steudel, 2003). It “studies, documents, analyses, and interprets how human beings construct and attach meanings to their experiences”(Patton, 2015, p. 13). Qualitative research has a“humanistic focus”(Gephart, 2004, p. 455) and sensitive to detail and research setting (Bryman and Bell, 2015; Doz, 2011) and as a result endeavours to understand typically a small number of participants’ worldviews rather than concrete realities of objects (Erlingsson and Brysiewicz, 2013) and testing of hypothesis on large samples (Geloet al., 2008). The voice of the researched is valued and occupies a central position in the research process (Ebbs, 1996).

To emphasise the importance of the qualitative line of inquiry, Morgan and Smircich (1980, p. 498) argued that as the:

[…] social world constitutes some form of open-ended process, any method that closes the subject of study within the confines of a laboratory […] does not do complete justice to the nature of the subject.

The qualitative strategy provides authentic setting (Aguiniset al., 2010; Turneret al., 2015) and permits the researcher, an opportunity to delve into and uncover the underlying assumptions, beliefs and values (Yauch and Steudel, 2003) shaping HRM practices in subsidiaries of western MNEs in Ghana.

The approach is less structured, more open-ended and flexible allowing participants to voice issues considered salient to them. Research participants have an opportunity to reveal their perspectives about the phenomenon under investigation without the researcher imposing any predetermined concepts and opinions on them. The researcher employs constructs and meanings in use by social actors to explain their interactive experience of social reality. The flexibility and open nature of qualitative research is likely to uncover new

384 QRJ 18,4

(4)

phenomena in IB worthy of further inquiry and thereby enriching the discipline (De Massis and Kotlar, 2014; Doz, 2011; Sinkovicset al., 2008; Yauch and Steudel, 2003).

It is possible to illuminate meanings and contextual influences on people’s perspectives and actions within a certain milieu (Buckleyet al., 2014; Given, 2008; Maxwell, 2013; Ritchie et al., 2014; Yauch and Steudel, 2003) employing qualitative methods.“Up-close qualitative types of methodologies” are appropriate for researching phenomena that require “deep contextual understanding to be meaningful”(Birkinshawet al., 2011, p. 575). Klingebiel and Stadler (2015, p. 197) concluded that“in-depth studies of decidedly African phenomena often hold greater promise than traditional surveys, at least as an initial step towards understanding contexts better”. Some important issues relating to HRM practices still remain not well understood in the Ghanaian context and as a result, this justifies the use of qualitative approaches.

Birkinshawet al.(2011, p. 573) observed that“qualitative methods can once again play a critical role to interpret and understand the complex plurality of contexts”. In Ghana, for instance, MNEs may need to acquire land for various purposes and it is interesting to note that about 80 per cent of land is customarily owned by families, clans and traditional authorities, such as chiefs (Kasanga and Kotey, 2001). These informal institutions are likely to have some influences on HRM practices in subsidiaries that have acquired land. Mason (2002, p. 3) observed that qualitative research produces“rounded and contextual understandings on the basis of rich, nuanced and detailed data”. The choice of qualitative approach is therefore considered appropriate for the purposes of researching subsidiaries of western MNEs.

Maxwell (2013, p. 30) explained that“a major strength of qualitative research is in getting at the processes that led to the outcomes, processes that experimental and survey research are often poor at identifying”. With qualitative methods, the researcher was able to probe in-depth for explanations about how HRM practices are standardised or localised and the reasons for doing so, from the perspectives of employees and managers experiencing the phenomena. The approach, therefore is “uniquely suited to ‘opening the black box’ of organisational processes”(Doz, 2011, p. 583) Thus, through the qualitative approach, the author gained rich insight and nuance for understanding HRM practices in western subsidiaries in Ghana.

Inquiry from an emic perspective

Peterson (2001) explained the terms“emic”and“etic”which are increasingly being used in IB research were originally coined by Pike (1967). Each perspective has basic underlying assumptions, values and beliefs about reality. Pike (1967) originally coined emic and etic terminologies derived from the suffixes of the nomenclatures phonemic and phonetic, classifications for linguistic analysis. Phonetics refers to the study of speech sound and analysis while phonemics denotes the distinctive meanings and structure of a particular language (Peterson and Pike, 2002; Pike, 1967; Sinkovicset al., 2008). It is suggested that

“in a strict linguistic sense, the phonetic implies an objective description, fully specified in all possible dimensions, and the phonemic implies a description which is based upon the categories employed by the people under study”(Buckleyet al., 2014, p. 309).

An etic orientation perceives reality as objective, culture free and suitable for hypothesis testing and generalisations (Buckleyet al., 2014; Sinkovicset al., 2008). Social phenomenon is independent of the researcher (Evered and Louis, 1981). It is associated with the universalists who argue in the context of IB that state-of-the-art organisational practices are valuable, context free and universally applicable (Mendoncaet al., 1999). However, Adler and Adler (1987) disagreed with the etic assertion and concluded that all human knowledge is influenced by the values and interpretations of the researcher.

Emic perspectives are often associated with subjective/idiographic/qualitative/insider terms (Buckleyet al., 2014; Evered and Louis, 1981; Sinkovicset al., 2008). An emic approach

385 Qualitative research

(5)

focusses on researching social phenomenon within a particular cultural context and understanding this phenomenon as members of that cultural context understand it (Schaffer and Riordan, 2003). The researcher adopts an emic perspective that favours the study of transfer of HRM practices from the point of view of employees and host country managers within their work context.

An emic view point tends to avoid imposing the researcher’s constructs on research participants and focusses on understanding the insider’s contextualised experiences, viewpoints, perceptions, meanings and interpretations of social phenomena (Evered and Louis, 1981). This perspective is suitable for this research because it supports the researcher’s philosophical, ontological and epistemological stance that phenomenon is socially constructed and can be interpreted from the perspectives of research participants. In line with Pfeffer (1981, p. 8), the researcher seeks to understand how social phenomena – HRM practices in subsidiaries of western MNEs – are“perceived, interpreted and legitimated” from the perspectives of employees and host country managers.

In-depth emic approach with a focus on participants’interpretations has the potential of expanding researchers’knowledge horizons about emerging, interesting phenomena in IB.

Thus, it is suggested that fieldwork enables researchers to gain“a deep understanding of the meanings that research participants attribute to the contextual factors in the field rather than just reading about them” (Michailovaet al., 2014, p. 148). As a result, this research seeks to appreciate and share the experiences of host country employees and managers using interviews and documents as sources of data.

Methods of data collection

The research question serves as the lodestar of this study whilst the literature review provides foundation for legitimising the research questions and justifying the current study as one that adds something novel to the scholarly conversations (see Levy and Ellis, 2006) on qualitative research and data analysis. The researcher developed two interview guides for both managers and employees. Due to the unexplored nature of HRM practices in western subsidiaries in Ghana, this research employed semi-structured interviews to bring to light how employees and managers in subsidiaries of western MNEs in Ghana understood and interpreted transferred HRM practices. It is argued that interviews are appropriate tools for the collection of data in the Ghanaian setting. Oppong (2017, p. 3) explained that interviews are culturally sensitive and that Ghanaians and, for that matter, many Africans are“more willing to provide information through oral narration, traditional way of imparting and sharing knowledge, than having to answer structured questions that include writing and experimentation”. Eight subsidiaries in the manufacturing, service and extractive sectors were selected for this study because these areas of the economy account largely for the foreign direct investment in Ghana (Sutton and Kpentey, 2012).

Codes/pseudonyms are used for both interview participants and the subsidiaries.

An interview guide was used to safeguard consistency (Brewsteret al., 2015) based on the research questions and systematic review of the literature. The interview guide ensures that the interview addresses themes identified in advance in the literature as crucial to the research questions (Kvale, 1996; Mason, 2004). It provides a relatively high level of flexibility and structure than the unstructured conversational interviews (Patton, 2015). A structure facilitates the organisation and analyses of the interview data (Bowen, 2009).

As a result of the nascent nature of HRM practices of western subsidiaries in Ghana, open-ended semi-structured interviews were utilised. Unlike forced choices in surveys, open-ended inquiry enhances the accuracy of retrospective reports because participants are at liberty to state whether they remember certain events or not (Miller et al., 1997).

The interviewee has a fair degree of freedom as to what to say, how much to say and how to express it (Rowley, 2012). Furthermore, semi-structured interview allows research

386 QRJ 18,4

(6)

participants to give a vivid description of their feelings, experience and views about the importance of a topic–in this case HRM practices of western subsidiaries in Ghana. Also, semi-structured interviews are flexible and offered the researcher an opportunity to approach different interviewees in varied ways while still covering the same questions.

Semi-structured interviews correspond with the qualitative research design for answering the questions “how” and “why”. Semi-structured interviews are associated with the ontological and epistemological stance that reality is socially constructed and interpreted in line with the worldviews of participants (Mason, 2004) which can be applicable to managers and employees working even in the same organisation (Heintz, 2002). The flexibility in semi-structured interviews makes it possible to cross-check and validate information from previous interviews (Bryman and Bell, 2015).

Information obtained from semi-structured interviews is often criticised for generating inconsistent data that cannot be compared between cases. Mason (2004, p. 4) challenged this criticism as inaccurate because semi-structured interviewing “uses a logic where comparison is based on the fullness of understanding of each case, rather than standardisation of the data across cases”. Comparing cases is not based on how uniform the data are across cases but based on comprehensive understanding of the HRM practices prevailing in each case subsidiary.

Conducting interviews

Interviews can take the form of“face-to-face interactions in which researchers typically pose questions that respondents answer” (Gephart, 2004, p. 458). They are avenues through which multiple realities are constructed (Stake, 1995). Simons (2009, p. 43) observed that interviews permit research participants to“reveal more than can be detected or reliably assumed from observing a situation”. Kvale (1983, p. 174) explained that qualitative researchers “gather descriptions of the life-world of the interviewee with respect to interpretation of the meaning of the described phenomena”. Interviews are a“construction site for knowledge”(Kvale, 1996, p. 14) with mutual learning simultaneously taking place between the interviewee(s) and the interviewer.

Conducting this research in the researcher’s own country made it possible to relate easily to the cultural context and the research participants. The researcher acquired cultural competence of the Ghanaian context having worked and lived in this setting his entire life.

This made it possible to understand the“cultural nuances and conceptions”of the interview participants (Karra and Phillips, 2008, p. 553). The researcher’s cultural know-how made it possible to pursue the interview questions with finesse. Karra and Phillips (2008, p. 553) argued that without the“necessary linguistic and cultural competence, an interviewer is likely to either abandon a line of questioning too early or to carry on and offend the interviewee”. However, Karra and Phillips (2008, p. 554) cautioned researchers who are natives of the research context that being close to the“the topic of study may make certain features seem more salient and important than they really are”. In light of this, the researcher adopted a reflexive stance by paying attention to“the most insignificant detail” and organising interview data“in a case study database”to enhance the transparency and reliability of the research (Gibbert and Ruigrok, 2010, p. 17).

The interviews allowed the researcher to probe for underlying values, beliefs and assumptions of participants shaping their interpretations of HRM practices. The interviews also allowed participants a voice on issues deemed salient to them (Yauch and Steudel, 2003). It is further argued that“interviews are well suited for capturing behaviours that have taken place in an authentic context”(Turner et al., 2015, p. 5) and to learn unique perspective from participants. The method highlights the context where transfer of HRM practices takes place, in order to succinctly explicate employees’ “understandings, interpretations, experiences and interactions” (Mason, 2004, p. 3). Additionally, through

387 Qualitative research

(7)

observations during interviewing, researchers can read participant’s body language and intonation which may serve as useful sources of data. Nonetheless, in order to help minimise memory loss arising from participants’ inability to recall facts and details about the phenomenon under investigation, this paper further studied and examined documents with relevant information for the research. Documents such as company bulletins, newspapers, flyers, mission and vision statements, company brochures, and information from company websites were studied and used to triangulate interview data. For purposes of confidentiality and to pre-empt the possibility of identifying the case companies, direct quotes from company websites are paraphrased.

The author adopted the funnel approach of interviewing, building of trust, rapport and presenting questions in an unbiased manner in order to minimise biases from both the interviewer and interviewee (Hutchings, 2005). Questions were structured beginning with name, position, professional experience, education leading to company history and finally focussing on HRM practices. In each organisation, a key person such as the human resource manager or director who is knowledgeable about the organisation in terms of its structure, policies and HRM practices was interviewed. Employees of participating subsidiaries were also interviewed along similar lines but questions about strategic policies concerning human resource practices were not included in lower level employee interviews. The interview for each manager lasted approximately 1.5 h whilst those of employees lasted for between 50 min to 1 h.

Certain salient reasons explain why both host country managers and employees were interviewed. Gamble and Huang (2009, p. 1686) observed that empirical studies on the transfer of HRM practices in MNEs are often criticised for “their “thin”, monochrome texture, with a reliance upon survey “snapshots” and/or interviews with a handful of managerial staff”. For instance, in the IB literature, examples of empirical studies confined to managerial level include: middle to senior management (Morris et al., 2009; Yu and Meyer-Ohle, 2008), top management and human resource managers (Björkman and Lu, 1999; Bjorkman and Lu, 2001). In a review of HRM studies in China between 1998 and 2007, Cooke (2009, p. 9) observed that“at least two-thirds of the empirical studies have collected their data from managers, and most of them relied on managers as the sole source for empirical data”. Using managerial staff as the sole source of empirical studies may mean that“the voice of employees is largely unheard” (Cooke, 2009, p. 9). Relying on a single participant may limit the authenticity of the interviews, if there are no other sources to cross-check the veracity of the information relayed (Patton, 1990). Michailova and Minbaeva (2012, p. 68) concluded that relying on a single participant per MNE or subsidiary top managers is not only ontologically problematic but also implies that “espoused values or ‘wishful thinking’ are captured”. Also, there is the need for research to “move beyond opinion obtainable among a population potentially part socialised to western mores–looking below the senior managerial surface to solicit views among the‘average’ employee, for whom the prospect of hybrid human resource management may remain more elusive”(Azolukwam and Perkins, 2009, p. 66).

Additionally, interview participants might not recall correctly past events due to memory failures leading to distortions (De Massis and Kotlar, 2014; Golden, 1992; Nutt, 1986). It is suggested that using multiple participants and“well-informed interviewees”with different perspectives might mitigate these problems (De Massis and Kotlar, 2014, p. 19) and enhance the confidence and trustworthiness of the findings (Graebner, 2009).

There is also a need to distinguish between HRM practices intended by a subsidiary and those actually implemented by interviewing managers and employees within and outside the HRM department (Khilji and Wang, 2006). For instance, in their study of the banking sector in Pakistan based on interviews, questionnaires and documents, Khilji and Wang (2006) found among other things that human resource practices could not simply be found from a single

388 QRJ 18,4

(8)

respondent in a subsidiary such as the human resource manager. Also, there was a lack of alignment between implemented HRM practices and intended HRM practices (Khilji and Wang, 2006). Thus, management knowledge and intentions of human resource practices may differ from what prevails in practice. As a result, the lived experience and knowledge of human resource practices of employees might differ from management (Zhuet al., 2013).

Managers may also use the interview as a site for identity work. According to Alvesson (2003), interviews could be used by the interviewee as a site for identity work by exhibiting one’s public relations aspects. Van Maanen (1979, p. 542) also noted that interviewees may also tend to put forward a “manufactured image” of routinised activities of the studied group to the fieldworker. The researcher therefore assumed a reflexive stance during the interviews by trying to make sense of every word and gestures used and to establish good rapport with the research participants (Millset al., 2006).

The use of multiple participants allows the researcher to cross-validate the information supplied by different sources (Bryman and Bell, 2015; Creswell, 2009; Patton, 2015).

Application of multiple sources of data–employees, managers and documents–increases credibility and provides a more holistic perspective on human resource practices in western subsidiaries in Ghana.

The author applied within-method triangulation utilising open-ended, semi-structured interviews as the major source of data on standardisation and localisation of human resource practices, with documents functioning as triangulation and supplementary sources of data (Corley and Gioia, 2004). He used documentary evidence to verify or question the details provided by research participants. In this way, the weaknesses of interviews usually associated with memory loss of participants (De Massis and Kotlar, 2014) and distortion of information due to the presence of interviewer (Grinnell and Unrau, 2011) make the unobtrusive nature of documents (De Massis and Kotlar, 2014; Ravitch and Carl, 2016) compensate for these shortcomings. Triangulation helps to minimise the researcher’s biases and increases confidence in findings ( Jick, 1979; Turner et al., 2015). It is particularly relevant in developing countries such as Ghana, where power distance is high, to use multiple sources of data. It is argued that persons in managerial positions in high power distance countries are more likely to monopolise information flow within an organisation with subordinates finding it difficult to air their views.

Rational for studying and analysing documents

Documents are regarded as key sources of data for qualitative research (Bryman, 2012;

Bryman and Bell, 2015; Patton, 2015). They are considered “the fabric of our world” (Love, 2003, p. 83) and“contain text (words) and images that have been recorded without a researcher’s intervention”(Bowen, 2009, p. 27). Hughes and Goodwin (2014, p. xxvi) defined documents as“typically involving the written word, pictures or sound recordings which, while they have the potential to serve as informants on human experiences in relation to specific life events, historical periods, social changes […] were not necessarily created for the purpose of documenting these”. Thus, unlike interviews, documents are created independently without a researcher’s intervention (Merriam and Tisdell, 2015).

Stake (1995, p. 68) stated that documents function as“substitutes for records of activity the researcher could not observe directly”. Documents are stable and discreet sources of data (De Massis and Kotlar, 2014; Merriam and Tisdell, 2015) but accessibility to organisational documents remains a challenge (Bryman and Bell, 2015). Some organisational documents that may be accessible and within the public domain are reports to shareholders, mission, vision and value statements. The researcher found it difficult to gain access to other documents such as minutes of meetings and employee handbooks. As a result, he relied on documents such as flyers, company bulletins, mission and vision statements and organisational websites.

389 Qualitative research

(9)

In addition to problems of accessibility, documents“exist in great numbers and in many forms”and there is a need to evaluate the authenticity and accuracy of documents (Bryman and Bell, 2015; Merriam and Tisdell, 2015). McCulloch (2004, p. 42) suggested that

“the author, the place and the date of writing all need to be established and verified”.

There is need to ascertain the “authenticity, credibility, accuracy […] preponderance of evidence that the documents will provide”(Bowen, 2009, p. 33). However, establishing the authenticity of these documents may be very challenging and in some cases impossible because these documents are from web sources. Merriam and Tisdell (2015, p. 176) observed that“the issue of authenticity about dealing with online documents and identities is much more complicated”. As a result, the researcher relied on company documents in line with Bryman (2012) who suggested that documents emanating from companies are likely to be authentic. Reviewing and analysing relevant contextual documents (Patton, 2015; Ravitch and Carl, 2016) enhanced understanding of the phenomenon under investigation by providing additional data for cross-checking and verifying first person accounts (Forster, 1994; Patton, 2015; Ravitch and Carl, 2016) of human resource practices in subsidiaries.

Despite problems of accessibility and authenticity, documents play a crucial role in the research process. They provide a description of “historical process and developments in organisations and can help in interpreting informants’rewriting of history in later verbal accounts”(Forster, 1994, p. 148).

Also, since organisational documents already existed and were not generated at the request of a particular research (Hughes and Goodwin, 2014; Merriam and Tisdell, 2015), a crucial criterion adopted for selecting documents include relevance to the research questions (Merriam and Tisdell, 2015). The research questions guided the entire research design.

Documents do not suffer interviewer and interviewee effects. They exist without the influence of a researcher (Brym and Lie, 2012; De Massis and Kotlar, 2014; Ravitch and Carl, 2016). The interviewer and interviewee effects on research findings can be attenuated by supplementing documentary data to that of interviews.

Nevertheless, because organisational documents were not generated at the instance of this research, they are often in a fragmented, inaccurate form for the research purpose. Documents might not encompass actual events and processes of an organisation (Forster, 1994; Merriam and Tisdell, 2015). Atkinson and Coffey (2004, p. 58) concluded that researchers:

[…] should not use documentary sources as surrogates for other kinds of data. We cannot, for instance, learn through records alone how an organisation actually operates day-by-day. Equally, we cannot treat records-however “official”-as firm evidence of what they report […].

This recognition or reservation does not mean that we should ignore or downgrade documentary data. On the contrary, our recognition of their existence as social facts (or constructions) alerts us to the necessity to treat them very seriously indeed.

The variety of sources of information offers the opportunity to look at the phenomenon under investigation holistically with an open mind (Bowen, 2009). If the information contained in the documents is similar to that of the interview scripts, then the confidence and trustworthiness of the qualitative information would have been established. Further probing was conducted through phone calls and e-mails to verify information in cases where there was divergence in information contained in interview scripts and documents.

Data analysis

It is often argued that qualitative data analysis is complex, laborious and time consuming with limited theoretical attention. The process can be frustrating particularly to novice researchers in the absence of clear-cut guidelines on how to analyse qualitative data using particular methods (de Casterle et al., 2012). This section discusses systematically the process of data analysis using both deductive and inductive approaches.

390 QRJ 18,4

(10)

The researcher employed both deductive and inductive approaches in analysing the data. The deductive approach uses an organising framework comprising of themes for the coding process (Bradleyet al., 2007; Braun and Clarke, 2006; Burnardet al., 2008; Miles and Huberman, 1994). The framework, often referred to as a start list (Miles and Huberman, 1994), is applied in the analysis in anticipation that certain core concepts are in the data (Bradleyet al., 2007; Thomas, 2006). The inductive approach, on the other hand, involves working exclusively from the participant experiences that drive the analysis entirely.

According to Thomas (2006, p. 238), the inductive analysis refers to “approaches that primarily use detailed readings of raw data to derive concepts and themes”. It entails going through data line by line thoroughly and assigning codes to paragraphs or segments of texts as concepts unfold (Bradleyet al., 2007; Curryet al., 2009) relevant to the research questions (Thomas, 2006). It is a recursive process that involves moving back and forth between data analysis and the literature to make meaning out of emerging concepts (Neeley and Dumas, 2016) and was used to capture“the most empirically grounded and theoretically interesting factors”(Schüssleret al., 2014, p. 147). In the inductive analysis,

“although the findings are influenced by the evaluation objectives or questions outlined by the researcher, the findings arise directly from the analysis of the raw data, not froma priori expectations or models”(Thomas, 2006, p. 239).

The data analysis process started with organising both interview and documentary data for each subsidiary. As an initial step, the researcher immersed in the data reading and digesting in order to make sense of the whole set of data and to understand“what is going on”(Morse, 1999, p. 404) through reflexivity, open-mindedness and following the rationale of participants’narratives. After several readings of the transcripts, he began to identify key concepts and themes using the research questions as the lenses (Braun and Clarke, 2006;

Eloet al., 2014).

The researcher adopted a sequential approach to analyse the data, by initially applying a deductive approach. With the deductive approach, initial codes are drawn from the existing literature on the topic of inquiry or what is known about the phenomenon of inquiry aided by the research aims, research questions and interviews questions.

The approach assumes that certain core concepts are in the data based on knowledge of the extant literature on the topic (Bradleyet al., 2007; Thomas, 2006). Data were coded into categories using a start list. Subsequently, the inductive approach was used to derive themes from the data (Thomas, 2006). The sequential approach using deductive and inductive analysis resulted in two data sets of categories which were reduced to manageable sizes by collapsing similar categories into higher order ones. From the integrated data set, data were cross-tabulated in order to examine and identify relationships among categories (see Suter, 2012).

Deductive data analysis

The research applied a deductive approach by “basing analysis on pre-existing theory” (Gale et al., 2013, p. 3), that is institutional theory in this instance. HRM practices are examined against the background of a multiplicity of cultural and normative institutional challenges that western MNEs face when they seek to transfer their preferred human resource practices to the Ghanaian context (Figure 1).

The “keyness” of a theme depends on “whether it captures something important in relation to the overall research”(Braun and Clarke, 2006, p. 82). In order to make full sense of the data, a start list ofa priori categories was generated in line with previous research (e.g. Hutchingset al., 2013; Miles and Huberman, 1994; Zikic and Richardson, 2016)–as illustrated in Table I – using the research questions. Using the topic recruitment and selection as an example, a start list is generated from issues mentioned in response to interview questions in relation to the research questions.

391 Qualitative research

(11)

The start lists provided initial focus in identifying certain key aspects of the data that directly relate to the research questions. The first column in Table I presents sub-questions from the interview guide that relate to the first and second research questions shown in Column 2. The start lists are shown in Column 3 while the fourth column explains the rationale for the start list. Interview transcripts are entered as text files to NVivo software version 11 to code the data using participants’ own expressions, terms and phrases (Gioiaet al., 2013; Petriglieri, 2015). Data were deductively analysed to develop clusters of data. The application of headquarters’rules and regulations, manuals from headquarters to subsidiaries signify standardisation of human resource practices while issues portraying power distance, selective patronage, centralised decision making, influence of chiefs or politicians and favouritism reflect localisation.

Organisation of interview and documentary data

Reading, digesting and making sense of the data

Open coding Start list

Categorisation Data coding according to

392 QRJ 18,4

(12)

Inductive data analysis

Having developed clusters of data from the deductive analysis, an inductive analysis was also conducted in line with Thomas (2006) by engaging in detailed readings of data not only to gain a holistic understanding of what was said (see Galeet al., 2013), but also to ensure that all important aspects of the data are captured (see Charmaz, 2014; Galeet al., 2013).

The researcher read texts line by line thoroughly and assigned codes to paragraphs or segments of texts relevant to the research questions. Key concepts and themes were identified using the research questions as the lenses. Deriving themes from the raw data using the inductive approach pre-empts the possibility of a researcher forcing a predetermined result (Bradleyet al., 2007; Braun and Clarke, 2006; Glaser, 1992).

The detailed analysis of the eight cases with a description of“what is going on”in each case (Wolcott, 1994, p. 16) resulted in two distinct mini-case reports of the eight subsidiaries on standardisation and localisation of HRM practices comprising 36 and 21 pages, respectively. The summaries of the case studies facilitated the next phase of an inductive analysis which primarily focussed on identifying“essential features and the systematic description of interrelationships among them –in short, how things work” (Wolcott, 1994, p. 12). The research employed cross-case analysis methods (Eisenhardt, 1989; Graebner, 2009), using tables to compare and contrast data by themes between cases (Galeet al., 2013; Graebner, 2009; Miles and Huberman, 1994). Overlapping categories were identified and further refined and reduced by clustering (see Burnard et al., 2008).

The researcher grouped data that conveyed similar meaning to higher-level nodes through triangulation of data from both interviews and documents to generate first-order descriptive codes (Neeley and Dumas, 2016; Vuori and Huy, 2016). Having established the first-order categories, the next phase of the analysis involved identifying links among the first-order categories in order to group them into theoretical distinct second-order themes (Dacin et al., 2010; Lawrence and Dover, 2015). This phase of the analysis was equally iterative, moving back and forth between the first-order descriptive codes and the evolving patterns in the data, until conceptual patterns were developed for second-order themes. Thus, while the first-order descriptive codes comprise mostly of

“informant-centric terms”, the second-order categories employ “researcher-centric

Example sub-questions from interview

questions Relates to RQ Category (start list) Rationale for category Is the headquarters

(HQ) directly involved in the implementation of HRM practices?

RQ1: Do subsidiaries of western MNEs standardise their HRM practices in Ghana?

If yes, why and how?

Influence of HQ (expatriates, rules, regulations)

For controlling and ensuring that HRM practices align with corporate practices What mechanisms does

HQ use to implement its HRM practices?

RQ1: Do subsidiaries of western MNEs standardise their HRM practices in Ghana?

If yes, why and how?

Mechanisms (manuals, visits by HQ personnel)

Guiding HRM practices

Does the Ghanaian culture influence your HRM practices? How?

RQ2: Do subsidiaries of western MNEs localise their HRM practices in Ghana? If yes, why and how?

Informal institutions (power distance, selective patronage and centralised decision making)

For legitimising HRM practices of subsidiaries

Describe how you were employed

RQ1 and RQ2 Mode of employment

(company websites, tests, relations working in the company)

Signalling either standardisation or localisation

Table I.

Category start list for recruitment and selection

393 Qualitative research

(13)

concepts, themes and dimensions”(Gioiaet al., 2013, p. 18). According to Van Maanen (1979, p. 540), informant-centric terms are the “facts” of the research. They are the interpretations of the“routinized practical activities actually engaged in by members of the studied organisation” (Van Maanen, 1979, p. 542). Thus, informant-centric terms represent informants’interpretations of their lived experience of human resource practices in subsidiaries. Researcher-centric terms, on the other hand, are the researcher’s interpretations of participants’ interpretations of their lived experience of the study’s phenomenon. Van Maanen (1979, p. 541) noted that they“represent what could be called interpretations of interpretations”.

In the final phase of the analysis, the researcher determined how the second-order themes interacted with and related to one another within a larger context in order to collapse these into aggregate theoretical dimensions (Dacinet al., 2010; Lawrence and Dover, 2015; Neeley and Dumas, 2016; Petriglieri, 2015).

In the process of analysing data, a researcher may intuitively search for data that confirm his/her personal experience and beliefs and fail to notice data that contradicts personal values (Smith and Noble, 2014). Thus, in order to minimise the incidence of personal bias and enhance the trustworthiness/rigour of the analyses (Burnard et al., 2008; Miles and Huberman, 1994; Pope et al., 2000), a colleague researcher, who worked with oil multinational companies in Nigeria and is familiar with MNEs’ operations in Africa, checked the coding schemes. He posed critical questions about the processes of data analysis which urged the researcher to look for additional ways of interpreting the data (Lauckner et al., 2012). He explained how employees in Africa, for instance, understood compassionate leave to cover emergency cases involving extended family members which in the MNE context is limited to the immediate family and further that payment of health care is interpreted to mean a caring organisation. These insights helped to refine the explanations and to expand the coding scheme. This independent review provided further perspectives and additional credibility to the analysis (Baker and Nelson, 2005; Gioiaet al., 2013).

Conclusion

The paper draws on the extant methodology literature to demonstrate how analysis of raw data from interview transcripts and organisational documents progressed towards the identification of core themes that captured the phenomenon of study. It sheds light on the steps involved in thematic analysis and develops a conceptual model to describe an approach that demonstrates rigour within a qualitative research using deductive and inductive approaches to achieve rich interpretive data analysis. Even though qualitative data analysis depends largely on the skills and ability of researchers to conceive, conceptualise and interpret data for successful analysis, it is anticipated that the model and processes applied in this analysis can be replicated and assist other students and researchers apply rigour and transparency in data analysis methods.

A methodological limitation pertaining to the study is subjectivity. For instance, qualitative data collection, analysis and interpretations are subjective and liable to researcher’s biases (Bryman and Bell, 2015; Denzin and Lincoln, 2011). The researcher brings his/her values into the research process. Thus, in order to minimise bias, the author triangulated interviews with documents in order to cross-check and validate the evidence.

Also a potential limitation of this research is the inherent retrospective bias of interviewees concerning past events (see Yorket al., 2016). The author sought to minimise these biases which often occur in cross-sectional data collection by triangulating interview data with documents, but future research might address this issue by examining individuals’perceptions across several points in time.

394 QRJ 18,4

(14)

References

Adler, P.A. and Adler, P. (1987),Membership Roles in Field Research, Sage, Newbury Park, CA.

Aguinis, H., Werner, S., Abbott, J.L., Angert, C., Park, J.H. and Kohlhausen, D. (2010),“Customer-centric science: reporting significant research results with rigor, relevance, and practical impact in mind”,Organizational Research Methods, Vol. 13 No. 3, pp. 515-539.

Alvesson, M. (2003), “Beyond neo-positivists, romantics, and localists: a reflexive approach to interviews in organizational research”,Academy of Management Review, Vol. 28 No. 1, pp. 13-33.

Atkinson, P. and Coffey, A. (2004), “Analysing documentary realities”, in Silverman, D. (Ed.), Qualitative Research: Theory, Method and Practice, 2nd ed., Sage, Thousand Oaks, CA, pp. 77-92.

Azolukwam, V.A. and Perkins, S.J. (2009), “Managerial perspectives on HRM in Nigeria: evolving hybridization?”,Cross Cultural Management: An International Journal, Vol. 16 No. 1, pp. 62-82.

Baker, T. and Nelson, R.E. (2005),“Creating something from nothing: resource construction through entrepreneurial bricolage”,Administrative Science Quarterly, Vol. 50 No. 3, pp. 329-366.

Birkinshaw, J., Brannen, M.Y. and Tung, R.L. (2011),“From a distance and generalizable to up close and grounded: reclaiming a place for qualitative methods in international business research”, Journal of International Business Studies, Vol. 42 No. 5, pp. 573-581.

Björkman, I. and Lu, Y. (1999), “The management of human resources in Chinese-Western joint ventures”,Journal of World Business, Vol. 34 No. 3, pp. 306-324.

Bjorkman, I. and Lu, Y. (2001), “Institutionalisation and bargaining power explanations of HRM practices in international joint ventures – the case of Chinese-Western joint ventures”, Organization Studies, Vol. 22 No. 3, pp. 491-512.

Bowen, G.A. (2009), “Document analysis as a qualitative research method”,Qualitative Research Journal, Vol. 9 No. 2, pp. 27-40.

Bradley, E.H., Curry, L.A. and Devers, K.J. (2007), “Qualitative data analysis for health services research: developing taxonomy, themes, and theory”,Health Services Research, Vol. 42 No. 4, pp. 1758-1772.

Braun, V. and Clarke, V. (2006),“Using thematic analysis in psychology”,Qualitative Research in Psychology, Vol. 3 No. 2, pp. 77-101.

Brewster, A.L., Curry, L.A., Cherlin, E.J., Talbert-Slagle, K., Horwitz, L.I. and Bradley, E.H. (2015),

“Integrating new practices: a qualitative study of how hospital innovations become routine”, Implementation Science, Vol. 10 No. 1, p. 168.

Brym, R. and Lie, J. (2012),Sociology: Pop Culture to Social Structure, Wadsworth Cengage Learning, Belmont, CA.

Bryman, A. (2012),Social Research Methods, Oxford University Press, Oxford.

Bryman, A. and Bell, E. (2011),Business Research Methods, Oxford University Press, Oxford.

Bryman, A. and Bell, E. (2015),Business Research Methods, Oxford University Press, Oxford.

Buckley, P.J., Chapman, M., Clegg, J. and Gajewska-De Mattos, H. (2014), “A linguistic and philosophical analysis of emic and etic and their use in international business research”, Management International Review, Vol. 54 No. 3, pp. 307-324.

Burnard, P., Gill, P., Stewart, K., Treasure, E. and Chadwick, B. (2008),“Analysing and presenting qualitative data”,British Dental Journal, Vol. 204 No. 8, pp. 429-432.

Charmaz, K. (2014),Constructing Grounded Theory, Sage Publications, London.

Cooke, F.L. (2009),“A decade of transformation of HRM in China: a review of literature and suggestions for future studies”,Asia Pacific Journal of Human Resources, Vol. 47 No. 1, pp. 6-40.

Corley, K.G. and Gioia, D.A. (2004),“Identity ambiguity and change in the wake of a corporate spin-off”, Administrative Science Quarterly, Vol. 49 No. 2, pp. 173-208.

Creswell, J.W. (2009),Research Design: Qualitative, Quantitative, and Mixed Methods Approaches, Sage, Thousand Oaks, CA.

395 Qualitative research

(15)

Curry, L.A., Nembhard, I.M. and Bradley, E.H. (2009),“Qualitative and mixed methods provide unique contributions to outcomes research”,Circulation, Vol. 119 No. 10, pp. 1442-1452.

Dacin, M.T., Munir, K. and Tracey, P. (2010),“Formal dining at Cambridge colleges: linking ritual performance and institutional maintenance”,Academy of Management Journal, Vol. 53 No. 6, pp. 1393-1418.

De Casterle, B.D., Gastmans, C., Bryon, E. and Denier, Y. (2012),“QUAGOL: a guide for qualitative data analysis”,International Journal of Nursing Studies, Vol. 49 No. 3, pp. 360-371.

De Massis, A. and Kotlar, J. (2014),“The case study method in family business research: guidelines for qualitative scholarship”,Journal of Family Business Strategy, Vol. 5 No. 1, pp. 15-29.

Denzin, N.K. and Lincoln, Y.S. (2011),The Sage Handbook of Qualitative Research, Sage Publications, Thousand Oaks, CA.

Doz, Y. (2011), “Qualitative research for international business”,Journal of International Business Studies, Vol. 42 No. 5, pp. 582-590.

Ebbs, C.A. (1996),“Qualitative research inquiry: issues of power and ethics”,Education, Vol. 117 No. 2, pp. 217-222.

Eisenhardt, K.M. (1989), “Building theories from case study research”, Academy of Management Review, Vol. 14 No. 4, pp. 532-550.

Elo, S., Kääriäinen, M., Kanste, O., Pölkki, T., Utriainen, K. and Kyngäs, H. (2014),“Qualitative content analysis”,Sage Open, Vol. 4 No. 1, pp. 1-10.

Erlingsson, C. and Brysiewicz, P. (2013), “Orientation among multiple truths: an introduction to qualitative research”,African Journal of Emergency Medicine, Vol. 3 No. 2, pp. 92-99.

Evered, R. and Louis, M.R. (1981),“Alternative perspectives in the organisational sciences:‘inquiry from the inside’and‘inquiry from the outside’ ”,Academy of Management Review, Vol. 6 No. 3, pp. 385-395.

Forster, N. (1994),“The analysis of company documentation”, in Cassell, C. and Symon, G. (Eds), Qualitative Methods in Organisational Research: A Practical Guide, Sage, London, pp. 147-166.

Gale, N.K., Heath, G., Cameron, E., Rashid, S. and Redwood, S. (2013),“Using the framework method for the analysis of qualitative data in multi-disciplinary health research”,BMC Medical Research Methodology, Vol. 13 No. 1, pp. 117-117.

Gamble, J. and Huang, Q. (2009),“The transfer of organisational practices: a diachronic perspective from China”, The International Journal of Human Resource Management, Vol. 20 No. 8, pp. 1683-1703.

Gelo, O., Braakmann, D. and Benetka, G. (2008),“Quantitative and qualitative research: beyond the debate”,Integrative Psychological and Behavioural Science, Vol. 42 No. 3, pp. 266-290.

Gephart, R.P. (2004), “Qualitative research and the academy of management journal”,Academy of Management Journal, Vol. 47 No. 4, pp. 454-462.

Gibbert, M. and Ruigrok, W. (2010),“The‘what’and‘how’of case study rigor: three strategies based on published work”,Organizational Research Methods, Vol. 13 No. 4, pp. 710-737.

Gioia, D.A., Corley, K.G. and Hamilton, A.L. (2013),“Seeking qualitative rigor in inductive research notes on the Gioia methodology”,Organizational Research Methods, Vol. 16 No. 1, pp. 15-31.

Given, L.M. (Ed.) (2008),The Sage Encyclopaedia of Qualitative Research Methods, Sage Publications, Thousand Oaks, CA.

Glaser, B.G. (1992),Basics of Grounded Theory: Emergence vs Forcing, Sociology Press, Mill Valley, CA.

Golden, B.R. (1992),“Research notes. The past is the past–or is it? The use of retrospective accounts as indicators of past strategy”,Academy of Management Journal, Vol. 35 No. 4, pp. 848-860.

Graebner, M.E. (2009),“Caveat venditor: trust asymmetries in acquisitions of entrepreneurial firms”, Academy of Management Journal, Vol. 52 No. 3, pp. 435-472.

Grinnell, R.M. Jr and Unrau, Y.A. (2011), Social Work Research and Evaluation: Foundations of Evidence-based Practice, Oxford University Press, New York, NY.

396 QRJ 18,4

(16)

Heintz, M. (2002),“East European managers and Western management theories: an ethnographic approach of Romanian service sector enterprises”,Journal for East European Management Studies, Vol. 7 No. 3, pp. 279-297.

Hoskisson, R.E., Eden, L., Lau, C.M. and Wright, M. (2000), “Strategy in emerging economies”, Academy of Management Journal, Vol. 43 No. 3, pp. 249-267.

Hughes, J. and Goodwin, J. (2014),Documentary and Archival Research: Sage Benchmarks in Social Research Methods, Vols 1-4, Sage Publications, London.

Hutchings, K. (2005), “Examining the impacts of institutional change on knowledge sharing and management learning in the people's Republic of China”,Thunderbird International Business Review, Vol. 47 No. 4, pp. 447-468.

Hutchings, K., Michailova, S. and Harrison, E.C. (2013), “Neither ghettoed nor cosmopolitan”, Management International Review, Vol. 53 No. 2, pp. 291-318.

Jick, T.D. (1979),“Mixing qualitative and quantitative methods: triangulation in action”,Administrative Science Quarterly, Vol. 24 No. 4, pp. 602-611.

Karra, N. and Phillips, N. (2008),“Researching‘back home’international management research as auto-ethnography”,Organizational Research Methods, Vol. 11 No. 3, pp. 541-561.

Kasanga, R.K. and Kotey, N.A. (2001), Land Management in Ghana: Building on Tradition and Modernity, University of Ghana, Accra.

Khilji, S.E. and Wang, X. (2006),“ ‘Intended’and‘implemented’HRM: the missing linchpin in strategic human resource management research”, The International Journal of Human Resource Management, Vol. 17 No. 7, pp. 1171-1189.

Klingebiel, R. and Stadler, C. (2015),“Opportunities and challenges for empirical strategy research in Africa”,Africa Journal of Management, Vol. 1 No. 2, pp. 194-200.

Kvale, S. (1983),“The qualitative research interview”,Journal of Phenomenological Psychology, Vol. 14 No. 1, pp. 171-196.

Kvale, S. (1996),Interviews: An Introduction to Qualitative Research Interviewing, Sage Publications, Thousand Oaks, CA.

Lauckner, H., Paterson, M. and Krupa, T. (2012),“Using constructivist case study methodology to understand community development processes: proposed methodological questions to guide the research process”,The Qualitative Report, Vol. 17 No. 25, pp. 1-22.

Lawrence, T.B. and Dover, G. (2015), “Place and institutional work creating housing for the hard-to-house”,Administrative Science Quarterly, Vol. 60 No. 3, pp. 371-410.

Levy, Y. and Ellis, T.J. (2006),“A systems approach to conduct an effective literature review in support of information systems research”,Informing Science: International Journal of an Emerging Transdiscipline, Vol. 9 No. 1, pp. 181-212.

Love, P. (2003),“Document analysis”, in Stage, F.K. and Manning, K.K. (Eds),Research in the College Context: Approaches and Methods, Brunner-Routledge, New York, NY, pp. 83-97.

McCulloch, G. (2004),Documentary Research in Education, History and the Social Sciences, Routledge, London.

Mason, J. (2002),Qualitative Researching, Sage Publications, London.

Mason, J. (2004),“Semi-structured interview”, in Lewis-Beck, M.S., Bryman, A.E. and Liao, T.F.F.

(Eds), The Sage Encyclopaedia of Social Science Research Methods, Sage Publications, Thousand Oaks, CA, pp. 1020-1021.

Maxwell, J.A. (2013),Qualitative Research Design: An Interactive Approach: An Interactive Approach, Sage Publications, Thousand Oaks, CA.

Mellahi, K. and Mol, M.J. (2015),“Africa is just like every other place, in that it is unlike any other place”,Africa Journal of Management, Vol. 1 No. 2, pp. 201-209.

Mendonca, M., Kanungo, R.N. and Aycan, Z. (1999),“Culture: the forgotten factor in human resource management”,The Journal of Business Perspective, Vol. 3 No. 2, pp. 1-7.