Abstract
The explicit consideration of Research Paradigms in International Human Resource Management, the title of this Special Issue, helps us in analysing and systematising the field to show how research in international human resource management is typically conducted, what preferred perspectives prevail and which approaches have been rather neglected so far. In this introduction, we map the field, and after defining the contours of international human resource management, we use the distinction between positivism and interpretivism to outline implications for the goals of international human resource management studies and associated ontological and epistemological assumptions. Next, we analyse research methods, ways to construct research questions, researchers’ roles, sampling procedures, data collection techniques, key theoretical contributions, focus on context in theory construction, quality indicators of data analysis and evaluation criteria in each of the two key research paradigms when studying international human resource management issues. In so doing, we offer a framework for the contributions made to this Special Issue, including literature reviews focusing on the systematisation of international human resource management research, using various paradigm lenses and specific methods. We sincerely hope that the notions, typologies and contributions included in this Special Issue, all based on extensive literature reviews, will help advance research in international human resource management.
Keywords
Introduction
International human resource management (IHRM) has grown substantially as a field of study and has become more multifaceted in the way conceptual and empirical contributions are designed (see, for example, Cooke et al., 2019; Pudelko et al., 2015; Stahl et al., 2012; for another Special Issue dedicated to this topic in the GHRM (German Journal of Human Resource Management), see Festing et al., 2013). As outlined by Pudelko et al. (2015), IHRM is not a field in its infancy anymore as stated for a long time based on a statement by Laurent (1986) in the early phase of IHRM research. We substantiate this point in the introduction to our Special Issue on Research Paradigms in International Human Resource Management, based on a variety of (recent) literature reviews in IHRM which help us to define and illustrate this field of research. Drawing on a philosophy of science perspective, through this Special Issue of the GHRM, we aim to analyse and systematise the field to show how research in IHRM is typically conducted. In line with other researchers who have also addressed this topic from different perspectives (Bonache, 2019; Cooke, 2018; Farndale et al., 2017b), we believe this is an important way to highlight developments in the field and classify contributions so that further research and research strategies in IHRM can be advanced. Together with its rich literature basis, this introduction, as well as the included papers, should help orient scholars embarking on their journey to investigating the fascinating field of IHRM and give support and guidance for important epistemological decisions to be taken on the way.
In this introductory article, we first address the notion of research paradigms in general and with respect to IHRM. Then, we present positivism and interpretivism as two key research paradigms and systematically outline the implications for studying IHRM (for an overview, see Table 1 later in this article). Taking this approach, we create the framework for this Special Issue. Finally, we introduce the papers included herein and clarify their contribution within the context of research paradigms in IHRM, that is, within the presented framework.
Research paradigms and study dimensions in IHRM.
Literature review on research paradigms in IHRM
Research paradigms
The notion of the research paradigm is influential and relevant because of Thomas Kuhn’s book The Structure of Scientific Revolutions. After being published first in 1962, the book had such an impact that it had to be re-printed several times in the following years. Today, it is considered a ‘classic’ tome, with its influence reaching beyond the limited circle of scholars interested in the theory and philosophy of science. In fact, his name is often found in journalist articles within the mass media, like the one published in The Guardian by Steven Poole (31 May 2018), who in a review on a book by the film-maker Errol Morris refers to Kuhn and his paradigms to explain the era of post-truth, fake news and the Trump administration’s invocation of ‘alternative facts’. It is thus logical that in management in general, and in IHRM in particular, Kuhn is also a highly influential author. If we combine the search terms ‘Thomas Kuhn’ and ‘Management Research’ in Google Scholar, we obtain more than 6010 citations (8 December 2019). Kuhn’s concept of the paradigm is grounded in a descriptive and historical analysis of the natural sciences, from which he concluded that only the mature natural sciences are paradigm-based, and social sciences are still in a pre-paradigmatic, pre-scientific phase due to their many approaches (Kuhn, 1962). However, the social sciences are in a state in which several paradigms permanently coexist and in which their rivalry generates fruitful discussions about the foundations of their approaches and methods. In the following, we will show how this multi-paradigmatic state enriches research endeavours in IHRM.
International human resource management
In this Special Issue, we do not focus on the application of Kuhn’s paradigms to management research in general but to the specific domain of IHRM (Cooke et al., 2019; Pudelko et al., 2015; Stahl et al., 2012), which represents an important part of HRM research (Markoulli et al., 2017). IHRM is defined as ‘. . . understanding, researching, applying and revising all HR activities in their internal and external contexts as they impact the processes of managing HRs in organizations throughout the global environment to enhance the experience of multiple stakeholders’ (Schuler and Tarique, 2007: 718). Research in this area addresses three different streams: HRM in the multinational context, comparative HRM and cross-cultural HRM (Dowling et al., 2017; see also Farndale et al., 2017b; Pudelko et al., 2015; Reiche et al., 2019a).
HRM in the multinational context looks at the way HRM is conducted in multinational enterprises (MNEs), for example, the extent of global standardisation and the local adaptation of HRM practices in the interplay between geographically dispersed units such as headquarters and subsidiaries (for a review, see also Brewster et al., 2016). On the contrary, various kinds of international work represent a major research field when HRM in the multinational context is concerned (for an overview on working internationally, see McNulty and Brewster (2019), including, for example, expatriation and the growing field of migration; for a consideration of global work in both, IHRM and international business perspectives, see Reiche et al., 2019b). For decades, a strong research focus has been on expatriate management/global mobility (for literature reviews, see Andersen, 2019; Caligiuri and Bonache, 2016; Kraimer et al., 2016; for a respective research handbook, see McNulty and Selmer, 2017). According to a literature review by Cooke et al. (2019), most papers on IHRM have been published on HR practices in MNEs between 2000 and 2014, with other topics including country of origin/domicile effects in multinational corporations (MNCs), and intra-organisational knowledge and strategy flows.
The second stream of research, comparative human resource management, includes the investigation of similarities and differences between HRM practices in different countries and regions (for a review, see Brewster et al., 2016; for a respective research handbook, see Brewster et al., 2018) and questions to what extent these converge or diverge (see, for example, Farndale et al., 2017a). Relevant data on international comparative HRM practices often stem from CRANET, which has provided international comparative empirical evidence of organisational policies and practices in comparative human resource management across the world since 1989 (Brewster et al., 2004; Dewettinck and Remue, 2011). Special emphasis in this research stream is given to the institutional perspective (Gooderham et al., 2019) as an explanatory framework.
The third research stream in IHRM, cross-cultural HRM, combines research from the fields of cross-cultural psychology and intercultural management studies with HRM (Dowling et al., 2017). For a long time, this research stream was inspired by national cultural values (e.g. based on Hofstede, 1980; House et al., 2001) or cultural distance, and it investigated the impact on HRM policies and practices (for a discussion, see Pudelko et al., 2015). More recent culture-related research provides more opportunities to address the cultural context in more differentiated ways (Gelfand et al., 2017), by questioning the assumption of nationality being a proxy for culture (Taras et al., 2016) and pointing, for example, to individuals engaging partially and plurally with culture, as described by polyculturalism (Morris et al., 2015), or to cultural tightness or looseness (Gelfand et al., 2006; for an application to the HRM system, see Farndale and Sanders, 2017). Furthermore, in addition to the dominant positivist views of culture, the importance of interpretivism is emphasised in cross-cultural IHRM research (Romani, 2019). Applying these perspectives to IHRM potentially enriches this research stream, and Farndale et al. (2017b) see it as standing distinct from the other two research streams, while Vaiman and Brewster (2015) explicitly address the particularities of cultural (Reiche et al., 2012) and institutional explanations (Wood et al., 2012) in IHRM. However, for example, Peterson and Van Iterson (2015) call for a combination of both perspectives when explaining IHRM phenomena.
Prevailing paradigms in IHRM
In order to understand the prevailing paradigms in the multifaceted IHRM research field, and to frame this Special Issue, we come back to Kuhn’s paradigms. It is important to point out that the notion of a paradigm, as defined by Kuhn, is not clear and does not have a single meaning. Margaret Masterman, in a workshop at the London Business School, distinguished 21 senses of the word as used by Kuhn in his book (Masterman, 1970), albeit Kuhn admitted the ambiguity of the term and tried to clarify it by classifying two different meanings (Kuhn, 1979). The first refers to the disciplinary matrix in a field: the set of fundamental theories, instruments, values and epistemological assumptions that all members of a scientific community share at a given time. This would include assumptions regarding the most pertinent problems to tackle, the most appropriate methods to be used, what an acceptable solution to the problems would look like and so on. The second meaning is a subset of the first one and refers to a set of ‘exemplars’ or particular scientific problems that have been solved by means of those theoretical assumptions – and which appear in textbooks on the discipline (Kuhn, 1979).
Drawing on Kuhn’s definition of a paradigm, we use such a term in IHRM to refer to the key theoretical and epistemological assumptions regarding the ‘right way’ to do research within the discipline. So understood, it is common to follow the seminal classification of Burrell and Morgan (1979) and describe four basic paradigms in the area: functionalist, interpretive, radical humanist and radical structuralist. Two of the papers in this Special Issue classify IHRM research contributions based on systematic literature reviews according to similar distinctions of research paradigms (see the papers by Primecz as well as Kornau, Frerichs and Sieben). Other authors, like Sandberg (2005), reduce existing research paradigms down to a choice between the positivist (i.e. the functionalist in Burrell and Morgan’s (1979) typology) and the interpretivist.
For the purposes of this introduction, we follow this second position, and thus we highlight the importance of paradigms in IHRM by describing and explaining how different it is to work and to do research in IHRM when depending on either a positivist or an interpretivist research position. We focus on the different goals and assumptions as well as on the ways in which IHRM issues are investigated. These differences are included in Table 1, and thereafter they are explained in more detail.
Positivism and interpretivism as the two key research paradigms in IHRM
Goals and basic assumptions
Positivism is a research paradigm that ‘seeks to explain and predict what happens in the social world by searching for regularities and causal relationships between its constituent elements’ (Burrell and Morgan, 1979: 5). It is the most widespread approach in the social sciences (Lee, 1991; Welch et al., 2011) and also, to date, in IHRM research – as shown by the contributions made by Primecz, and Kornau, Frerichs and Sieben in this Special Issue. Following this paradigm, it is common to assume that things are known by models and relations among variables and that the role of researchers is to provide (deductive-nomological) explanations through generalisable causal mechanisms that can be measured empirically (Cascio, 2012; Edmondson and McManus, 2007). This emphasis on verifiable and causal relationships leads Gimbel (2016) to define positivism as the position that repudiates ontology (what things are in themselves, their nature or essence) in favour of epistemology (what can be known and tested). This stance regarding ontology is not shared by all scholars characterising positivism. In fact, some social science researchers (Guba and Lincoln, 1994) suggest that positivists are committed to realist ontology, which means that entities and facts in the research domain exist and that their existence is objective and mind-independent. In any case, they all share the view according to which the task of science is to provide a true representation of that reality by means of models and causal laws (Kolakowski, 1993).
Interpretivism is an approach based on philosophical phenomenology (Sandberg, 2005). It is a relatively rare perspective in the area of IHRM and qualitative research, although some work inspired by this paradigm is published in top-tier management journals (e.g. Caprar, 2011; Sandberg, 2000). Its objective is to make sense of the meanings and subjective intentions of particular individuals in a given context, without imposing a priori analytic categories (Babones, 2016; Gephart, 2004). This attempt to understand the world from the perspective of its participants cannot be carried out without the interpretative effort of the researcher, hence the ‘interpretivism’ in the name of this methodology. In contrast to positivism, which assumes that there is an objective world that can be represented by concepts and propositions, the interpretive phenomenological approach asserts that we do not live in a singular social reality but rather in multiple socially constructed views of social reality (Babones, 2016; Lee, 1991; Yanow and Schwartz-Shea, 2015).
Positivism and interpretivism assume different answers to a number of essential questions and issues regarding the ‘right’ way to conduct research in IHRM. We have classified them through the nine categories described below, but it is important to highlight, however, that these categories are not the only ones that allow us to differentiate the way of studying IHRM within the two paradigms. Indeed, we could very well use others (e.g. the preference for single vs multiple research settings, etc.), but the point would be basically the same: positivist and interpretive paradigms have a very different understanding of how to do ‘good’ quality research within the area.
How to study IHRM
Research method
It is common to oppose two fundamental research methods, namely, quantitative and qualitative, and the choice depends on the overall goal and basic assumptions of the research project. Edmondson and McManus (2007: 1155) speak about a methodological fit, which ‘. . . refers to internal consistency among elements of a research project – research question, prior work, research design, and theoretical contribution’ (for the importance of fit specifically in the context of qualitative research, see also Gehman et al., 2018).
Some authors associate positivism with quantitative methods (e.g. Babones, 2016), ‘. . . allowing for singling out causal relationships on the assumption of a cause-effect relationship between variables’ (Della Porta and Keating, 2008: 26). Typical methods include statistical analysis and experiments (Della Porta and Keating, 2008), and although most quantitative methods have been in place for a long time, some interesting research tools have been newly applied to IHRM. An example is the use of vignettes to investigate favouritism in recruitment and selection (Hotho et al., 2018; for an overview, see Wallander, 2009). As will be outlined in detail in the literature reviews in this Special Issue by Primecz; Kornau, Frerichs and Sieben; as well as Redondo, Fabra and Martin, the association between positivism with quantitative methods is a widely spread approach in IHRM.
Despite the above-mentioned association between positivism with quantitative methods, there is also a strong tradition of qualitative positivism (Prasad and Prasad, 2002), which has been named as such because it uses non-quantitative methods within largely positivist assumptions (Johnson et al., 2006). Epistemologically, qualitative positivist research focuses on searching, through non-statistical means, for regularities and causal relationships between different elements of the reality, before summarising identified patterns into generalised findings (Guba and Lincoln, 1994). Qualitative positivism is the dominant qualitative methodology in HRM research (Harley, 2015) and has a long tradition in IHRM research (see, for example, Festing, 1997). Eisenhardt (1989) and Yin (2018) are the two most influential authors in this area, and it is difficult to find qualitative works (even some that claim to work with non-positivist paradigms) that do not refer to them to justify their methodology. Part of their success lies in offering researchers a series of guidelines which are sufficiently specific for conducting rigorous (positivist) qualitative research (Piekkari et al., 2009). Nothing equivalent can be found in the non-positivist paradigms, whose researchers face a very fragmented and an even contradictory panorama of recommendations relating to how to carry out their studies and how to evaluate their contributions (Prasad and Prasad, 2002; Welch and Piekkari, 2017). This lack of guidelines has been addressed by the more recent interpretivist qualitative approach suggested by Gioia et al. (2013) (see also Aguinis and Solarino, 2019; Flick, 2018a), and it will be further discussed below in the context of that research paradigm. According to Eisenhardt and Graebner (2007), qualitative (positivist) studies use empirical evidence from one or more organisations to build theory and hypotheses inductively. These hypotheses can then be tested quantitatively (Eisenhardt, 1989), thereby revealing the degree to which the relationship between variables identified in the qualitative study can be extended to the population as a whole. Quantitative and qualitative methods are thus perfectly compatible in positivism, to the point where today there is a tendency in many management and human resource studies to combine them in what is defined as a ‘hybrid method’ (Edmondson and McManus, 2007). In line with this idea, data triangulation, a combination of using different research methods in a complementary way, is emphasised, although this is not without challenges (Della Porta and Keating, 2008). An example of using different research methods in the context of IHRM is found in a comparative study in an Asian context conducted by Collins et al. (2017). The authors applied a mixed-methods approach consisting of a quantitative approach with an experimental character, followed by a qualitative approach, in order to discover different contextual factors that may influence the postulated relationship between human factors on knowledge-sharing within cross-national virtual teams.
In interpretivism, the relationship between quantitative methods is more problematic. As mentioned previously, interpretivism is based prominently on phenomenology. This philosophical position was developed by the German philosopher Edmund Husserl (1900–1901), who in spite of his initial mathematical training argued that the methods of natural science are inadequate for the study of social reality. As interpretivism ‘. . . aims at understanding events by discovering the meaning human being attribute to their behaviour and the external world’ (Della Porta and Keating, 2008: 26), the focus is on meanings and studies explicitly considering contextual variables instead of trying to decontextualise (Cooke, 2018). Methods typically applied include the analysis of texts and discourses (Della Porta and Keating, 2008), so the qualitative methodology is dominant. However, as mentioned above, qualitative methodology should not be equated with qualitative positivism as introduced above. In the overview by Gehman et al. (2018) on three important qualitative approaches to theory-building, outlining the differences between the approaches by Eisenhardt (1989), Gioia et al. (2013) and Langley (1999), Gioa points out that he sees himself as a ‘pure interpretivist’ (Gehman et al., 2018: 291). Moreover, he pursues a clearly different approach than, for example, Eisenhardt (1989), whose work is assigned to the positivist qualitative approach. As mentioned above, Gioia et al. (2013) introduce strong rigour in interpretivist qualitative research when emphasising the importance of the development of a well-documented data structure that is built on first-order codes, second-order themes and aggregate dimensions (Gehman et al., 2018; Gioia et al., 2013). Examples of employing this approach to HRM, highlighting the particularities of specific firm, industry or national contexts, can be found in Harsch and Festing (2019, 2020) and Muratbekova-Touron et al. (2018).
Constructing research questions
Positivist researchers typically hold two essential assumptions regarding the ‘right’ way to construct research questions (Alvesson and Sandberg, 2011): ‘Gap spotting’ is the way to justify a study, and this study should build upon and extend the state of the art. The key focus is on explaining phenomena in organisations by searching for contextual and organisational variables that cause specific actions, starting with a thorough literature review and the elaboration of hypotheses (Swanson, 2005). For example, Farndale et al. (2017a) study the effects of institutional factors, caused by different types of market economies, and the nationality of the organisation (foreign vs domestic) as contextual factors in the variation and alignment of HRM practices.
From the interpretive perspective, there has been some discussion over whether the gap-spotting approach is the right way to formulate research questions, with problematisation being proposed instead (Alvesson and Sandberg, 2011). This directly challenges the two assumptions mentioned above and suggests that the research questions must identify and challenge existing assumptions in order to disrupt an established body of literature and construct new and original theories. For example, Caprar (2011: 608) chose an ethnographic approach to analysing host-country nationals in Romanian subsidiaries of US American MNCs, and based on his empirical findings and the patterns identified, he contributed to enhancing the explanation of ‘the assumed localness of HCNs’ by emphasising the interplay between organisational and host-country culture.
Researcher’s role
The separation of facts from values is one of the main theses of positivism (Chalmers, 2013) when uncovering truth and fact through primarily quantitative studies (Swanson, 2005). The researcher is an ‘outsider’, while reality is ‘objective’ and subjectivity is irrelevant (Kolakowski, 1993), which means that the researcher should remain detached, objective and value-free (Lee, 1991). According to this idea, therefore, the role of the researcher in interpretive studies is problematic.
For example, in an interpretive study on international compensation in a workers’ cooperative (Bonache and Zárraga-Oberty, 2019), the authors agreed to join the working team responsible for redesigning a compensation system for expatriate staff, on an unpaid basis, defining two conditions necessary for their cooperation: first, they would have unrestricted access to data, and second, they would proceed with the rigour and pace expected of academic work rather than of management consultancy. This type of agreement meant that they were able to put themselves in the cooperative members’ shoes and understand their standpoint on the issue. It is clear that this agreement does not respond to the positivist view on the researcher’s role; however, in this instance, it did allow the researchers to develop rapport and empathy (i.e. understand the true experience of the research purpose), and this is precisely the type of attitude and relationship between researchers and research participants that interpretivism requires of the former of these two groups (Sandberg, 2005). Instead of being problematic, as assessed by positivism, the role that they assumed in the research was paradigmatic, albeit within an interpretivist paradigm. The role of the insider status of researchers in IHRM is also the subject of one of the contributions of this Special Issue, since Collins and McNulty provide insights into the challenges and the impact of the researcher’s insider status on various issues in expatriate management research.
Sampling procedures
Sampling is necessary because a study of all elements of interest is not possible, especially in an international setting: ‘The general issue of sampling is how to select cases or examples from a wider population . . . so that the research in the end can make statements that apply not just to the individual participants’ (Flick, 2018c: 175).
In positivist quantitative research, researchers use statistical (probabilistic and non-probabilistic) sampling in order to create a test group that is representative for a population (Cascio, 2012; Flick, 2018c). A common approach used in probability sampling strategies is a random sample, where each participant has the same probability of being selected for the study, which is not the case in non-probabilistic sampling. Probabilistic strategies include simple random, systematic, multi-stage random, stratified random and cluster sampling, while non-probabilistic sampling uses judgement, quota, convenience and snowball systems: ‘Of the two, probability samples tend to be more costly and more accurate, tend to take more time, are more universally accepted, and usually are better in terms of the generalizability of results than are non-probability samples’ (Cascio, 2012: 2358). For probability sampling, the definition of a population is critical – and this may present a challenge in IHRM research with its global scope. Therefore, if, due to cost or time constraints, probability sampling is not possible for a certain IHRM research project, the authors should clearly argue why a non-probabilistic sampling strategy has been used and also illustrate the limitations of this approach (Cascio, 2012).
In both positivist and interpretive qualitative research, however, theoretical or purposeful sampling approaches are required (Patton, 1990), but there is no formalised sampling routine; instead, it is about defining substantial criteria (Flick, 2018c). Purposeful sampling (Flick, 2018c) makes no pretence at attempting to obtain a random sample representative of a large and diverse population, as happens in statistical sampling (Aguinis and Solarino, 2019). Instead, it identifies and selects information-rich cases related to the phenomenon of interest, in which ‘one can learn a great deal about issues of central importance to the purpose of the research’ (Patton, 1990: 169). Examples of sampling strategies include extreme case, typical, maximal variation, critical or convenience case sampling, and often, sampling decisions are not taken ex ante but in the course of the investigation, and they tend to follow the criterion of substance than formality (Flick, 2018c). Interpretive researchers speak also of opportunistic sampling, or convenience sampling (Caprar, 2011), which can be observed in the comparative study by Horwitz et al. (2006), who study the application of different HR practices to manage knowledge workers in two culturally distant countries, namely, Singapore and South Africa.
Data collection techniques
Data collection techniques in positivist studies include quantitative elements such as questionnaires often using Likert-type scales, secondary data analysis, experiments, quantitatively coded documents and qualitative approaches focused on grounded theory testing (Swanson, 2005). The dominant data collection technique for quantitative studies is surveys (for questionnaire design, see Rowley, 2014). This is also true for IHRM research; however, due to the international dimension herein, special emphasis should be given to the particularities of research designs across national and cultural boundaries (for further information, see, for example, Byrne and Van de Vijver, 2010; Harkness et al., 2003; Raghuram et al., 2017). Cascio (2012) points to the importance of translation, conceptual, functional and metric equivalence. Furthermore, the hitherto neglected experimental data collection technique is introduced and discussed in one of the contributions of this Special Issue, based on a systematic literature review in the context of expatriation. While Noethen and Alcazar (2019) emphasise the potential of improving our (causal) understanding of expatriate management, they also outline the current limited uses and the challenges of using experiments in IHRM.
As the qualitative research methodology is a leader in interpretivist research, typical data collection techniques include interviews, focus groups, document and archival data analysis, participant observation and ethnography, often with the goal of grounded theory development (see, for example, Swanson, 2005). Interviews (Brinkmann and Kvale, 2018; Flick, 2018c; Rowley, 2012) as an often pursued data collection technique in IHRM research have specific challenges, due to the international business context. For example, Marschan-Piekkari and Welch (2004) point to language issues in cross-cultural interviewing, and Marschan-Piekkari et al. (2004) outline the specific challenges of the multinational organisational context. The contribution made by Mahadevan in this Special Issue analyses the usefulness of ethnographical research in IHRM (for an elaboration of this method, see also Coffey, 2018; for ethnographical challenges in international business, see Chapman et al., 2004; for overall recommendations on qualitative research design, see Flick, 2018a).
Constructing theory and (key) theoretical contribution
Theory is understood in positivism as ‘a collection of assertions, both verbal and symbolic, that identifies what variables are important and for what reasons, specifies how they are interrelated and why, and identifies the conditions under which they should be related or not related’ (Campbell, 1990: 65). The way to construct theory in quantitative positivist research is on a step-by-step basis, following ‘. . . a sequential process in which hypotheses are formed and data are collected and then analysed’ (Edmondson and McManus, 2007: 1163). Studies typically focus on the antecedents or the side effect(s) of a phenomenon and add new, independent, moderating or mediating factors. The focus in positivist quantitative research is on theory testing, and the form of knowledge researchers aspire to develop is ‘natural laws’, or in post-positivism ‘probabilistic laws’ (Della Porta and Keating, 2008: 23). For instance, the publications by Malek et al. (2015) and Selmer and Lauring (2015) focus on the antecedents of expatriate adjustment such as support by the parent company, the host-country nationals and language. Furthermore, they relate this to expatriate performance.
In a Special Issue of the Journal of International Business Studies, Birkinshaw et al. (2011) stress the need for more qualitative research in international business and respective theory construction. Following this notion, Welch et al. (2011) suggested a typology for theorising from case studies. Based on the dimensions of ‘causal explanation’ and ‘contextualisation’, they distinguish between inductive theory-building, interpretive sensemaking, natural experiments and contextualised explanations, and find in their literature review a strong leaning towards inductive theory-building, which reflects the positivist paradigm (in line with the above-indicated case study approach by Eisenhardt, 1989). Grounded theory (Glaser and Strauss, 1967; for research guidance, see Flick, 2018b) is often used for inductive theory-building and is the most important approach in this regard within qualitative research. It involves ‘a spiral of cycles of data collection, coding, analysis, writing, design, theoretical categorization and data collection’ (Flick, 2018c: 449). Thus, the research corresponds to an iterative cyclic learning journey (Edmondson and McManus, 2007). In line with Goodman (1978), theories resulting from emerging codes in empirical research
. . . are visions of the world. These visions undergo a continuous revision, evaluation, construction and reconstruction. According to this view, theories are not (right or wrong) representations of given facts, but versions or perspectives through which the world is seen. (Flick, 2018c: 127f)
For instance, in the area of self-initiated expatriates, researchers conducted semi-structured interviews for data collection, and these were thereafter analysed following a grounded theory approach in order to explore this relatively new research field on the factors determining why self-initiated expatriates were employed and why they choose employment in developing countries (Kumar and Chhokar, 2018). Bešić and Ortlieb (2019), in a multiple case study, draw on semi-structured interviews to theorise about management rationales behind the assignment of so-called ‘expatriates of host-country origin’ to subsidiaries in transition countries.
Interpretivism also states that our descriptions of reality are always coloured by our pre-understandings (Alvesson and Sandberg, 2011; Gimbel, 2016). Nevertheless, unlike qualitative positivism, we do not have to decide a priori which pre-understandings will be challenged in the study. If we were to do this, we would not be applying the phenomenological epoché (Husserl, 1900–1901), which involves remaining open to being challenged by the data by continually calling into question our pre-understandings (Alvesson and Sandberg, 2011). In this sense, the specific content of what is included in the literature review – within the interpretative phenomenological approach – is much more informed by the findings than is usually the case for positivism. This entails a rather different way of constructing theory. In comparison with the concepts and research questions derived from the literature, typical in positivism, research questions in interpretivism emerge from the field (Friedrichs and Kratochwil, 2009). Their sources are usually puzzles, surprises or tension between the expectations derived from existing literature and what the researcher encounters in his or her field experiences (Yanow and Schwartz-Shea, 2015). These puzzles cannot be fully anticipated. For example, in the study on international compensation mentioned above (Bonache and Zárraga-Oberty, 2018), the authors expected to find an international compensation management strategy very similar to the one described in the literature and experienced in other organisations (i.e. use of the equity principle to justify salary differences, internal differentiation in working conditions, tension between conflicting or contradictory objectives). During their fieldwork, they discovered – very much to their surprise – a different and alternative way of tackling and experiencing the challenge of designing and managing international compensation. This finding inspired them to address different research questions (i.e. to what extent do the ways in which members of workers’ cooperatives understand and handle the challenge of designing an international compensation system reflect the specific ways MNCs understand and manage that challenge?) and led them to reassess their initial assumptions and theories. From that initial puzzle, and making use of a sound and rigorous methodology that would ground the theory (Leitch et al., 2010), they eventually revealed a different and alternative way of designing expatriation packages and explaining and justifying disparities in salary. This new method included some contextual conditions missing in the existing literature (e.g. the social mission of cooperatives, their democratic character, the local culture, professional careers of worker-members), thus providing contextual knowledge (Della Porta and Keating, 2008) and making the initial puzzle less surprising and more of a ‘normal’ or ‘natural’ event. This gradual process from puzzles to final understandings is the paradigmatic way of generating theory within interpretivism.
The importance of context in theory construction
The consideration of context, especially the national and the cultural contexts in comparative and cross-cultural streams (see above), is a core focus in IHRM (Cooke, 2018; De Cieri and Dowling, 2012; Dowling et al., 2017; Jackson et al., 2014; Schuler and Tarique, 2007). The importance of the further investigation of context has been emphasised not only in IHRM (e.g. Cooke, 2018; Farndale et al., 2017b) but also in the wider fields of organisational (Johns, 2017), management (Bamberger, 2008) and international business research (e.g. Meyer, 2015; Michailova, 2011), and it has been subject to recent calls for papers in major HRM journals (e.g. Farndale et al., 2019). The consideration of context and its conceptualisation (see, for example, Johns, 2006) has an impact on the way explanations in IHRM are created (Cooke, 2018) and is thus related to the arguments elaborated in the last section, that is, the way theory is constructed and the key theoretical contributions. The discussion about context and research paradigms is not new, though. For example, Brewster (1999) has highlighted the distinction between universalist and contextualised paradigms in IHRM for a long time, calling for a stronger focus on the latter (see also Dewettinck and Remue, 2011) in the context of the CRANET network. Whetten (2009) further differentiates the discussion about contextualisation when distinguishing between ‘theories in context’, with the goal to apply existing theories to new contexts and contribute context-related information about the further development or limitations of these theories, and ‘theories of context’ focusing on developing new context-related theories.
In positivist research, context is often treated as an exogenous variable that has an impact on the subject to be studied – as an independent, a moderating/mediating or a control variable. For example, scholars build on results from cross-cultural management research, such as the national cultural value dimensions identified by Hofstede (1980) or GLOBE (House et al., 2001), when analysing particular HR practices. This is the case in a publication by Nadeem et al. (2018), which reviews studies on the efficiency of HR practices, specifically high-performance work practices, in different cultural contexts. The authors identified a total of 70 studies which used cultural value dimensions identified by the aforementioned authors as well as Trompenaars and Schwartz (Schwartz, 1994; Trompenaars and Hampden-Turner, 1997). Another option is that it is controlled for context (Johns, 2017) and that ‘acontextual’ research results are produced (see also Bamberger, 2008: 841; Cooke, 2018). One example of a study pursuing this approach in the context of IHRM is the work on gender-specific preferences in global performance management by Festing et al. (2015), where it is illustrated that these preferences hold across cultures. Even in qualitative positivist research, where an IHRM-related phenomenon is studied within its real context, the goal is to build models in which the key variables and relationships can be abstracted from the ‘idiosyncratic detail’ of individual cases (Eisenhardt and Graebner, 2007). In this sense, qualitative positivism treats and conceptualises context as a limitation (Welch et al., 2011).
Interpretivist research treats context as more than a control variable (Johns, 2017), and it focuses on the notion of ‘theories in context’ (Whetten, 2009), in order to contribute contextualised knowledge (see also Tsui, 2004; Welch et al., 2011), that is, localised knowledge. This is in line with the call by Bamberger (2008) for more context-oriented qualitative research, citing the work by Chreim et al. (2007) as a positive example and emphasising the importance of creating a higher acceptance of (well-conceptualised) interpretive studies in the scholarly community, thereby questioning the superiority of context-free knowledge generated in the universalist paradigm (Tsui, 2004; Welch et al., 2011; Whetten, 2009). Examples of studies in the context of IHRM are outlined in the contribution to this Special Issue on ethnography by Mahadevan, who underlines an emic approach to researching culture as compared to the etic approach which dominates in the positivist approach (for the emic/etic distinction in international business and IHRM, see Buckley et al., 2015; Dowling et al., 2017; Teagarden and Von Glinow, 1997).
Quality indicators of data analysis
Quantitative researchers following the positivist paradigm have concerns and recommendations regarding the quality of data analysis. The focus is on the link between prediction and explanation (Swanson, 2005). In qualitative positivism, it is assumed that, first, triangulation is a key component of qualitative research, and second, the quality of a case study can be judged by its ability to provide a single, converging explanation. The first assumption is shared with interpretivism. Triangulation entails looking at a phenomenon from different angles (Yin, 2018), and this is applicable both to positivism and interpretivism. That said, for positivism, the consequent benefit thereof is the development of converging lines of evidence (Edmondson and McManus, 2007).
Conversely, for interpretivism, triangulation ‘. . . serves also to identify different ways the phenomenon is being seen’ (Piekkari et al., 2009), which is why, in interpretivist studies, rather than finding the intercoder agreement, as suggested by positivist researchers (Yin, 2018), it is typical to see how the same phenomenon is experienced in a different way by different people. In other words, the quality of an interpretivist study should not be judged by its ability to reflect a single, converging explanation (as is typically done in qualitative positivism) but for its ability to show how the same phenomenon is experienced and viewed from a plurality of viewpoints and perspectives. Important quality criteria in this regard are trustworthiness and authenticity (Swanson, 2005; for a further discussion on quality in qualitative research, see Flick, 2018d).
Evaluation criteria
One of the most divergent aspects between the positivist and interpretivist traditions relates to the standards used to justify the quality of a study. Positivists do not hesitate to refer to internal, external and construct validity and reliability as essential criteria for assessing the rigour of field research (Campbell and Stanley, 1963; Swanson, 2005), and relying on the consensus related to these criteria, the task of investigators consists of specifying the elements and procedures used to produce a valid and reliable work (Aguinis and Solarino, 2019; Gibbert et al., 2008). In positivist qualitative studies, similar criteria for assessing quality have been suggested (Welch and Piekkari, 2017).
The situation faced by interpretivist researchers when coping with the challenge of including criteria to justify their study is much more complicated. In order to avoid the risk of their work being evaluated using positivist criteria (Johnson et al., 2006; Welch and Piekkari, 2017), it is usually advised that the authors explicitly specify the standards intended for evaluating the quality of the study (Bonache, 2019). The answer to the question about which standards should be used is not easy; for instance, the interpretivist literature, even when also insisting on rigour and a systematic approach in its studies (Gioia et al., 2013; Yanow and Schwartz-Shea, 2015), does not offer the unanimity that can be seen in positivism, and different authors offer different standards (Welch and Piekkari, 2017). However, they all agree that interpretative studies should not be judged by objectivity but by alternative criteria. In addition, they also believe that objectivity is a myth, and they insist that ‘the truth’ or ‘the correct way’ to justify or interpret organisational reality does not exist (Yanow and Schwartz-Shea, 2015). That said, this does not imply rejecting the idea of rigour in research. Making use of the Weberian distinction between subjective scientists and rigorous science (Gimbel, 2016), they assert that it is necessary to assume that although researchers are subjective and saddled with pre-understandings, interpretive research should aspire to rigour in the way the interpretative study is conducted (Gioia et al., 2013; Welch and Piekkari, 2017).
Consistent with this position, some interpretivist authors (e.g. Sandberg, 2005; Yanow and Schwartz-Shea, 2015) specify the assessment standards necessary to judge the quality of a non-positivist study. They stress that the process of interpretation must be conducted systematically and rigorously so that the study is characterised by communicative validity (i.e. what the researchers understand and interpret is consistent with what is meant by the individuals being studied), pragmatic validity (i.e. there is no discrepancy between what the subjects being studied say they do and what they actually do) and interpretative awareness (i.e. the interpretation reached relating to the experience lived by the individuals being studied does not just reflect the beliefs, values and biases of the researcher himself/herself). In order to achieve these standards, it is also important to adopt systematically a series of initiatives, which they specify in the manuscript. For example, in order to ensure interpretative awareness, researchers should include, in the different phases of the field study, initiatives such as phenomenological epoché (see above), treating all the statements as equally important or asking follow-up questions. These standards will probably not convince positivist scholars, but they are consistent with the interpretive epistemological paradigm.
Compilation of this Special Issue
The call for the GHRM Special Issue on Research Paradigms in IHRM was published at the 16th Conference on International Human Resource Management in Madrid and beyond. In total, we received 14 submissions, which we appreciated very much. This gave us the chance to select those contributions most focused on looking at IHRM from a philosophy of science perspective and advancing the field in that way. As it was not our goal to give room to single studies using certain paradigms or certain methods, but instead to take a meta-perspective, we had to exclude some papers, even though they were promising and will certainly be published in other outlets. Thus, six contributions were accepted for publication in this Special Issue.
Structure of the Special Issue
The first part of the Special Issue is devoted to papers explicitly focusing on the notion and variety of research paradigms and their relative importance in IHRM research, reflected by the results of systematic literature reviews. We decided to include three systematic literature reviews herein: the work by Primecz was the broadest with respect to the literature base with 1649 analysed contributions, all collected from 1990 to 2018; the second literature review by Kornau, Frerichs and Sieben was based on a coding of 203 peer-reviewed IHRM articles, published between 2011 and 2018; and third, a literature review by Redondo, Fabra and Martin focused more specifically on systematising quantitative research in IHRM and linking it to the relevant journals as publication outlets. Thereafter, the important and interesting question about the researcher as an insider is raised and discussed by Collins and McNulty. Finally, we include papers focusing on the application of specific methods in IHRM and which, of course, are also strongly related to the context of research paradigms. These include ethnography and experiments – methods that have not been used that much in the past but which have a lot of potential when aiming at advancing IHRM research. We introduce the papers included in this Special Issue in more detail in the next section.
Summaries of papers included in the Special Issue
The aim of the first paper by Henriett Primecz, titled ‘Positivist, constructivist and critical paradigms in IHRM and some future directions’, is to analyse to what extent the field of IHRM is characterised by the various social science paradigms. The foundation of this work consists of an analysis of four major journals publishing IHRM research over 30 years. The author concludes that positivist studies overbalance constructivist contributions and that critical perspectives on IHRM are largely missing. Future research directions as outlined by the author include more paradigm reflexivity, a stronger focus on constructivist and critical studies as well as new paradigmatic directions.
The second contribution by Angela Kornau, Ilka Frerichs and Barbara Sieben, namely, ‘An empirical analysis of research paradigms within IHRM: The need for more diversity’, provides an analysis of research paradigms used within the field of IHRM and emphasises a need for more diversity. Within a systematic literature review, they coded 203 peer-reviewed IHRM articles published between 2011 and 2018. Their theoretical lens is Siebens’ compass of management research (2007), reflecting systematisation derived from critical management research. Distinguishing between ideologically critical, poststructuralist, functionalist and interpretive perspectives, certain methods, theories and topics residing in IHRM are discussed. Their findings point to the dominance of functionalist and interpretive studies, while critical research perspectives are limited in the IHRM research field. In terms of topics, IHRM in MNEs and expatriates are prevalent, and macro-level theoretical approaches mainly include institutional perspectives. As a conclusion, the authors promote a more diverse and reflexive approach to studying IHRM and indicate the potential values of more research diversity in this field.
In the third literature review, titled ‘A new ranking of IHRM journals: What type of quantitative research do they publish?’ the focus is on quantitative research in IHRM. Raquel Redondo, Eugenia M. Fabra and Gloria Martín suggest a new ranking of IHRM journals and analyse what type of quantitative research they publish, following their questions on ‘when, where, what, and how?’ The authors looked at the period from 2010 to 2018. Based on a conjoint analysis of empirical quantitative IHRM studies, they identify how various topics have been investigated, leading to recommendations for improving the analysis and knowledge of certain IHRM issues and helping identify trends and research gaps.
Another contribution, which is related to the research paradigms as well as to methodological challenges, is the paper on the seldom addressed insider status of researchers, by Heidi Collins and Yvonne McNulty, titled ‘Insider status: (Re)framing researcher positionality in IHRM studies’. Based on personal experience, the authors describe a series of methodological and ethical challenges involved in being an insider in research projects. They offer a framework of researcher personae for reflecting upon a researcher’s positionality before, during and after data have been collected and analysed, and they provide recommendations for research in IHRM and beyond this field.
In her paper, not only representing the second methodology-focused part of the Special Issue but also explicitly making the link to the research paradigms, Jasmin Mahadevan focuses on the usefulness of ethnography for IHRM, understanding it as a multi-paradigmatic mind-set based on the principles of ethnography in anthropology. She differentiates between five interrelated strands, including structural-functionalist ethnography, interpretive ethnography, critical, postmodern and postcolonial ethnography, and outlines how these can be applied. She demonstrates that each specific perspective needs to be discussed and applied by considering the specific context of IHRM, and she draws implications for potential contributions to research in IHRM. This should provide a roadmap for future IHRM researchers to use ethnography as a guiding principle of their research.
Daniela Noethen and Rocio Alcazar, in their paper titled ‘Experimental research in expatriation and its challenges: A literature review and recommendations’, focus on experimental research in expatriation. The basis is again a systematic literature review, which resulted only in an output of 17 papers over a period of 20 years of IHRM research, indicating a lack of evidence for causality in expatriate management research. With experiments and quasi-experiments, this paper points to often overlooked research methodologies that can improve our (causal) understanding of expatriate management. However, the authors also indicate challenges and opportunities that are unique to experimental research in expatriation: challenging data access, global sample dispersion, restricted manipulability of variables, and the cultural boundedness of constructs and interpretations. Eventually, the paper outlines how research in expatriation can be advanced – and what implications this might have for the decision making of practitioners managing international assignees.
Concluding remarks
The guest editors sincerely hope that this Special Issue will be of great interest to the IHRM research community, especially those who are starting their journey investigating IHRM. It should provide them with a clearer understanding of the choice of their research strategy and its implications for advancing the field of IHRM while avoiding problems that occur when the methodological fit is low (i.e. the consistency between the research question, prior work, research and contribution to the literature). For qualitative studies, this could involve reinventing the wheel, but without producing new knowledge, or so-called ‘fishing expeditions’, quantitative studies might use constructs and lack reliability and validity, and hybrid research might be characterised by an uneven status of evidence (Edmondson and McManus, 2007: 1170). As Cascio (2012: 2543) points out, methodological challenges in IHRM research strategies are ‘neither inevitable nor inscrutable’, but as we see especially in qualitative (international) research (Birkinshaw et al., 2011), they need careful attention, a clear position indicating opportunities and limitations of the chosen approach when researchers aspire to make a valuable contribution to advancing our knowledge in IHRM. As for all HRM research, rigour is an important issue for quality publications (Wright and Ulrich, 2017). We hope that the notions, typologies and contributions included in this Special Issue will help advance research in IHRM in that direction. Drawing up a research agenda on research paradigms in IHRM and their potential application is beyond the scope of this Special Issue introduction, as the field of IHRM is so broad. However, the recent literature reviews cited here, as well as the analyses provided in the contributions to this introduction, should represent a great start for anyone taking up quality research in IHRM.
Footnotes
Acknowledgements
The authors of this introduction, and the guest editors of this Special Issue on Research Paradigms in International Human Resource Management, would like to thank Professor Rolf Brühl, ESCP Business School Berlin, Germany, and Professor Renate Ortlieb, University of Graz, Austria, for their very valuable inputs to this introduction. Furthermore, we would like to thank the reviewers of this Special Issue, who have contributed to ensuring the overall development and quality of the six contributions: Torsten Biemann, Rolf Brühl, David Caprar, Vincent Cassar, Heidi Collins, Marjaana Gunkel, Anne-Wil Harzing, Jasmin Mahadevan, Wolfgang Mayrhofer, Yvonne McNulty, Daniela Noethen, Victor Otra, Henriett Primecz, Sylwia Przytula, Markus Pudelko, Raquel Redondo Palomo, Ramon Rico, Joanna Samul and Barbara Sieben, Last but not least, we would like to thank the team of the Chair of Human Resource Management and Intercultural Leadership at ESCP Business School Berlin, especially Michael Volk, Katharina Salmen, Lynn Schäfer, Tobias Schumacher and Agnès Cretté for ensuring the format quality of this Special Issue.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.
