Building the Road Ahead: Developing Theory in HRM Through Literature Reviews
这篇社论概述了文献综述如何通过系统综述、叙事综述和整合综述等方法，在人力资源管理研究中推动理论发展，并为在《人力资源管理杂志》发表综述提供指南。
Literature reviews play a key role in HRM research, allowing us to look back at the development of knowledge within a field of study, and to present an overview of the state of the art. In this way, reviews can classify, represent, problematize, configure, aggregate, integrate, interpret, and explain the literature (Kunisch et al. 2023). Literature reviews thus enable scholars to analyze and synthesize “an existing body of literature by identifying, challenging, and advancing the building blocks of a theory through an examination of a body (or several bodies) of prior work” (Post et al. 2020). In addition to looking back, literature reviews look forward, setting out new paths for the future development of research (Elsbach and van Knippenberg 2018). In this way, reviews help to develop theory by exposing emerging perspectives, analyzing assumptions, clarifying constructs, establishing boundary conditions, testing new theory, theorizing with systems theory or theorizing with mechanisms (Post et al. 2020). Theory-driven reviews therefore involve both integrative and generative processes (Post et al. 2020), where the former seek to analyze and synthesize prior research (Torraco 2016), whilst the latter seeks to generate new ideas and theories based on the review of this prior research (Gatrell and Breslin 2017). This dual role of both looking back and looking forward means that the review has to simultaneously align itself with prior work, whilst at the same time diverging from this (Breslin and Gatrell 2023). This generative dimension of reviewing is key to the development of theory, connecting research in original ways to present the literature in a new light (Breslin and Gatrell 2023; Post et al. 2020). A literature review is therefore more than a systematic descriptive summary of prior research (Cropanzano 2009), critically interrogating and reimagining the literature (Alvesson and Sandberg 2020). In this editorial, we present an overview of the different ways in which authors of literature reviews can contribute to theory in HRM. We evaluate a range of review methods and set out guidelines for publishing literature reviews in Human Resource Management Journal (HRMJ). There are several types of review method which can be used to develop theory. In general, reviews have different goals including describing, understanding and explaining the literature, which has an impact on the scope or breadth of the review (Rowe 2014). Many reviews are descriptive in nature, and do not seek to develop theory. However, reviews published in HRMJ should go beyond summarizing/synthesizing the literature, to put forward alternative theoretical directions or solutions. In the following section, each of these methods will be presented, and critically evaluated using examples from HRM to illustrate key points. Perhaps the most widely known review method is the systematic literature review, which emerged in the field of medicine as a method for synthesizing research in a systematic, transparent and reproducible manner (Cook, Mulrow, and Haynes 1997). In their seminal work, Tranfield, Denyer, and Smart (2003) drew on this research to present the systematic review to the field of management research. The systematic method was seen to counteract the implicit biases, lack of critical assessment, lack of rigor and “infrastructural arrangements” associated with narrative reviews of the literature (Jones and Gatrell 2014; Tranfield, Denyer, and Smart 2003). Through the systematic approach, management scholars could instead apply specific principles to counteract bias by making explicit the values and assumptions underpinning a review (Snyder 2019; Tranfield, Denyer, and Smart 2003). Reviews could then provide reliable evidence for policy and practice. Systematic literature reviews thus claim to adopt a replicable, scientific and transparent process, minimizing bias through an exhaustive search of the literature (Tranfield, Denyer, and Smart 2003). For example, published in the Journal of Management, Boon, Den Hartog, and Lepak (2019) carry out a systematic review to understand the synergistic effects of interactions within HR systems. Their review develops theory by highlighting areas of conceptual and empirical confusion in the composition and measurement of HR systems (Boon, Den Hartog, and Lepak 2019). The systematic review process involves a number of stages. First a review panel comprising a range of experts is established, who set out the purpose and scope of the review, and with this, a specific research question. Scoping studies may be conducted to help in this process, clarifying definitions and refining the purpose and scope. The product of this planning phase is the review protocol, which defined the study's search and selection strategy. Second, the review scholars begin their search of the literature, using keywords and search terms developed through the scoping study (Tranfield, Denyer, and Smart 2003). This search process should be free from preexisting beliefs, and thus open to so-far unknown research from other fields (Jones and Gatrell 2014). The outputs of this search process are then selected using inclusion and exclusion criteria, which has been formulated by the review panel. These criteria aim to include research which is relevant to the predefined research question (Hiebl 2023; Snyder 2019). Inclusion criteria may be based on other characteristics such as an assessment of article's quality using journal rankings as a proxy. Third, and having selected a sample of papers, the reviewers then independently read and assess the papers (Tranfield, Denyer, and Smart 2003). The purpose of this stage is to produce a synthesis of the literature, by summarizing and integrating findings (Tranfield, Denyer, and Smart 2003). This synthesis is an inductive and interpretive process, which includes both a descriptive and thematic analysis of the selected papers. Crucially, it is through this process that new generative insights and interpretations of the literature arise. Whilst the systematic approach claims to be replicable (Snyder 2019; Tranfield, Denyer, and Smart 2003), there are a number of problematic issues associated with the process. First, whilst the systematic approach is well suited to reviews in which the aim and scope can be set out in advance, this is often not possible. Given the planned nature of the systematic review, the method may be ill-suited to research questions that are less well-defined or when the review seeks to develop theory through a more explorative process (Fan et al. 2022). For example, when the aim and scope of the review is more broad and less well defined, the search and selection process can occur hand-in-hand with the formulation of the research question. As a result, the stages noted above can occur concurrently and through iterative steps (Locke, Feldman, and Golden-Biddle 2022). Tranfield, Denyer, and Smart (2003) note that the review protocol should be both transparent and flexible, as management reviews often involve a “process of exploration, discovery and development.” However, the huge investment of time and resources needed to complete an exhaustive systematic search and selection of the literature will discourage review scholars from repeating the process in an iterative fashion. Second, keyword searches often miss important research, and as a result snowballing through manual searches is needed to supplement the systematic approach (Greenhalgh and Peacock 2005). However, the reporting of snowballing techniques is often opaque (Hiebl 2023). Third, whilst searches normally include two or more databases (Hiebl 2023), these searches are time dependent, as the content of databases can change over time. Finally, inclusion criteria such as the assessment of relevance to the research question is a subjective process which lacks transparency and is difficult to replicate. Furthermore, it is assumed that inclusion criteria remain valid, even after the reviewer has delved into the literature and changed their understanding of that literature. Whilst snowballing and manual search are possible, these processes supplement at the fringes. Despite these issues, over the past decade practices of systematicity have become increasingly prevalent in all published literature reviews in management (Simsek, Fox, and Heavey 2023). As noted above, systematic approaches are not well suited to tackle the full spectrum of research questions, and so researchers have selectively borrowed practices from the systematic approach (Simsek, Fox, and Heavey 2023). From narrative to integrative review methods, systematic practices of search and selection are now commonplace. Simsek, Fox, and Heavey (2023) found that whilst the degree of systematicity varies across published papers and review types, a minimum level is expected in published papers. Interestingly, the more authors reported systematicity practices, then the greater the research impact of the paper (Simsek, Fox, and Heavey 2023). Whilst systematic reviews are planned in advance through defined research questions and review protocols, narrative reviews represent a process of exploration and discovery. These reviews use a narrative to present the literature in a new light, through a related concept, or from a different perspective, bringing to the fore the interpretations of the review scholar (Rowe 2014). For example, in Vardi and Collings' (2023) narrative review of talent management published in HRMJ, the authors analyze three streams of the talent literature (talent management, human capital, and organizational psychology) through the lens of paradox theory. They develop a framework of five tensions of talent: inclusive and exclusive, innate and acquired, transferable and context-dependent, a focus on subject and object, and input and outcomes (Vardi and Collings 2023), which the authors argue has important implications for the underlying assumptions behind how talent is understood and managed. Published narrative reviews rarely consider all aspects of the literature, and instead examine it through a particular lens defined by the article's objectives, as the authors tell a story (Torraco 2005). Much like the story tellers of old, narrative reviews engage and convince the reader through the rigor, strength and coherence of their arguments and critique. As a result, the review should not be unsurprising (Rowe 2014). Given the different orientation of narrative reviews, the stages of the review process are often less planned than a systematic review. First, the research question is less well defined than with a systematic approach. This question may represent broad aims for the review, which makes the scope of the review more difficult to define in advance. Second, the process of search is one of discovery, as the review scholar explores the literature using a range of keywords. As a result, search and selection criteria are not necessarily predetermined in advance (Fan et al. 2022). Third, papers are selected based on their broad relevance to the research question, but also based on the interest they spark in the mind of the reviewer. In this way, the discovery of papers may lead to a change in key words and subsequent searches. Fourth, the reviewer interprets each paper through an emerging lens which represents their unique understanding of the literature. By critically interrogating and reimagining the existing literature, the review scholar can “generate new and ‘better’ ways of thinking about specific phenomena” (Alvesson and Sandberg 2020, 1291). In this process, they make connections between concepts and themes within the domain and beyond. Narrative reviews compare and contrast research findings, re-interpret these through new perspectives, and thus generate new theoretical insights. This process involves both members of the review team and wider colleagues, as each discusses their interpretation of the readings. Like members of a book club, they recount the papers they have read and reflect on their impressions and understandings. By sharing narratives, a co-construction of new understandings emerges. Through this snowballing of selected papers, and iterative process of creation (Locke, Feldman, and Golden-Biddle 2022), the review scholar discovers the literature alongside an emerging re-interpretation of that literature. The “stages” of the review process are therefore intertwined. In the systematic approach, theory development is focused on the stage of reading and interpreting selected papers. This assumes that a relevant sample of papers has been (or even can be) selected in advance of the reading of those papers. The narrative method takes a different approach, and puts generative theory development at the center of every stage of the review process. The processes of narrative reviews have been described as informal (Hammersley 2001) and incremental (Jones and Gatrell 2014). Some have also argued that the method lacks thoroughness and systematicity, resulting in biased and subjective reviews of the literature (Fan et al. 2022; Snyder 2019; Tranfield, Denyer, and Smart 2003). For example, papers are selected based on the author's discretion, and thus the accuracy of the position taken and presented has been questioned (Greenhalgh 1997). In response to these criticisms, and as noted above, review scholars have increasingly opted for more “systematic” methods in their search and selection processes (Simsek, Fox, and Heavey 2023). To retain the element of discovery, these systematic searches have been supplemented with manual searches, with up to 70% of selected articles discovered through “snowballing” techniques (Greenhalgh and Peacock 2005). The integrative review shares many features with both the systematic and narrative review method (Fan et al. 2022). Torraco (2005) describes an integrative literature review as a review which “reviews, critiques, and synthesizes representative literature on a topic in an integrated way such that new frameworks and perspectives on the topic are generated.” These reviews are systematic in the sense that they set out to synthesis the “state of the science” in a rigorous and replicable manner (Elsbach and van Knippenberg 2020). They are also exhaustive in nature, systematically tracing much (or maybe even all) of the literature on a selected topic back to its roots (Callahan 2010). Integrative reviews also share the generative aspects of narratives reviews, as they review, critique, and synthesize the literature to generate new theoretical frameworks, insights and perspectives (Cronin and George 2023). For example in a review published in HRMJ, Huang, Wilkinson, and Barry (2023) carry out an integrative literature review of employee voice. Drawing on HR, organizational behavior and industrial/employment relations literatures, they develop a multi-level model of contextual factors the processes through which employee are into different of Wilkinson, and Barry 2023). The authors the of contextual to which for to be and and types of contextual Integrative reviews such as this, therefore of a body of evidence to new insights through (Elsbach and van Knippenberg However, there are aspects of the integrative approach that from both narrative and systematic First, the aim and scope of the integrative review is broad in nature, setting out to review across different and research and are thus key features of this review as they systematically the literature on a selected topic (Callahan 2010). These reviews also seek to both the state of the whilst setting out new directions for theory development (Elsbach and van Knippenberg 2018). of the approach argue that the and development of research in management a review, critique, and of and knowledge (Torraco 2005). Second, search and selection processes should be transparent and to such an that the findings can be by (Torraco 2005). Given the broad nature of this review key words should the different terms and concepts used by different inclusion criteria to that different perspectives are in the synthesis of the literature. The integrative review is therefore wider in scope than narrative reviews integrating different conceptual and research (Cronin and George 2023). Through this inclusion of different theoretical perspectives, the integrative literature reviews can help develop both a synthesis and of the literature (Torraco 2005). Finally, the reading and interpretation of the literature seeks to and across different (Cronin and George 2023). This process involves the of themes within each alongside the of themes across thus involves the out of themes from these (Cronin and George 2023). such themes involves to consider the themes (Cronin and George 2023). is through this process of critical assessment, and synthesis of the literature, that integrative reviews generate theoretical and integrative theoretical frameworks (Breslin and Gatrell 2023; and van Knippenberg Torraco 2005). However, it is argued that these insights should from the review and as a result, arguments or perspectives the selection and interpretation of the literature (Elsbach and van Knippenberg 2020). and George (2023) use the of and to the scope of narrative and integrative Whilst narrative reviews generate insights by on the in or integrative reviews compare different across different (Cronin and George 2023). The integrative review thus across of new connections to other related Despite the of integrative reviews, there are also First, and the breadth and scope of integrative reviews, they are in terms of time and and as a result, often involve the of a team of Second, such a review process such as conceptual (Snyder which may be beyond the of many As a result, more authors are to an integrative review (Elsbach and van Knippenberg 2018). Third, it is argued that integrative reviews are exhaustive and in their scope (Callahan 2010). one the literature like in an empirical it be argued that this can be a number of (Elsbach and van Knippenberg 2018). As a result, argue that to it can be (Elsbach and van Knippenberg 2020). A more development in literature reviews to the use of or analysis is used to analyze the between a sample of by of papers which have been in the same these are by other papers, then it assumes that these share a research or research In this way, it or of articles of and et al. 2017). A number of are to these including and to the literature in this way, can connections between themes and within a These can the development of the literature over research connections across As a result, they are used in the of literature. However, most published reviews using remain descriptive in nature, as the and connections between authors or seminal papers. In other these descriptive review do more than the outputs from or when by more review analysis can be important for theory development through literature reviews (Fan et al. 2022). papers a review protocol which is by the aims of the review and the specific research question. As with systematic this protocol seeks to make the search and selection process transparent and the sample of papers have been a of analysis can then be using a to a range of different and connections within the literature. is at this stage that most authors the review process, the findings of the analysis in their However, and in to generate theoretical insights from these findings, is Drawing on the in the review scholars can their search and into the literature to the nature of these In this way, authors can develop theory to explain the connections and found through the and this approach in their review of published in the Journal of Management They used the outputs from techniques to develop a integrative model of research and 2022). They three streams of research, which they then using theories of to the of They therefore on the analysis to examine the theoretical of and integrative ways to understand the through and development and 2022). In with these in review HRMJ seeks to literature review articles that make a to theory development in Human Resource Management and the management of at Literature reviews published in HRMJ should be and and the journal reviews which are in A range of review methods are as noted above, the paper makes a conceptual setting out or new research The key criteria used to evaluate literature review are as Literature review papers should be generative in nature and make a to theory, how scholars research in their is important therefore to how this knowledge within a In after reading the review, it should be how researchers the field This also involves review from other published and it within the literature. Literature review articles should be in a review of the literature, and review methods should be both rigorous and of scope to the Reviews can be focused on a specific or be in However, it is important that the review is and that methods are transparent and HRMJ papers using a range of review methods including systematic literature reviews, integrative reviews, critical narrative reviews and review types to beyond a descriptive review of the state of within a and make a to theory. For example, analysis papers to from the and connections through their analysis to advance new theoretical insights. To this papers are often with a systematic or critical literature review. should also provide of key stages of the literature review process. have selection and processes been keywords have been databases have been have papers been from the selected have selected papers been or These will of change on the literature review HRMJ also theory papers, and these should be from the literature reviews First review papers to be transparent with to the methods used to and the literature (Callahan 2010). of the review method this transparency an for the the focus of theory papers is in the of a specific theoretical (Elsbach and van Knippenberg 2020), and thus they on specific relevant literature (Callahan This selection of the literature is by the author's perspectives (Elsbach and van Knippenberg 2020). papers on the other to a selection of literature (Elsbach and van Knippenberg 2020), with a more inclusion of related research (Breslin and Gatrell 2023). Third, and as a result of the above, authors of theory papers may or evidence from the wider literature (Cropanzano scholars instead consider all the evidence through their search protocol, and be critical in their approach to a of prior (Jones and Gatrell 2014). Theory-driven literature reviews play a key role in building the and the research The of represents the nature of literature of the and different and they to new (Post et al. 2020). also the past with the They are to the through the that in the and change how they the as they through In this way, the but are also by Whilst reflect the paths we they are for to They are more than but and review a new for to as they through the research reflect the the reader takes as they the research through the review. also reflect the nature of the review process. the selection and interpretation of the literature is planned or it through a process of discovery, we read and how we this on we have read A research can be as a of knowledge each with its and (Cronin and George 2023). to these of researchers in new up and across There are many that the can Some will be Some will seek to many within an integrated of will specific in new following paths that are less and more These latter paths open up and of the in new ways (Breslin and Gatrell 2023; Post et al. 2020). HRMJ seeks reviews which set out new paths within the research in HRM and the management of at work, connecting of researchers in new are from the paths we as we to on the with These be using rigorous techniques and They to within the of both the and wider research Whilst within a beyond these to a wider range of to understand both looking forward and back, with the of has HRMJ we seek reviews that across within and beyond on new through the research