Abstract
We review ontological, epistemological, and methodological concerns in writing up research, distilled from selected inductive studies published in leading academic journals. From this analysis of practices emerges the following categorization, (a) rhetoric, (b) craftsmanship, (c) authenticity, (d) reflexivity, and (e) imagination, which informs the writing up of appealing and convincing qualitative research. We give examples and propose actionable writing heuristics. We offer reflections and recommendations on how qualitative research writing could be improved and its diffusion accelerated.
Half-convinced writers trying to half-convince readers of their [the writers’] half convictions would not on the face of it seem an especially favorable situation for the production of works of very much power.
Once a chronic victim of institutionalized discrimination (Symon, Buehring, Johnson, & Cassell, 2008), the “poor cousin” of the “red-headed stepchild” of inquiry (see Smith, Madden, & Plowman, 2014) is finally taking her rightful place at the adults’ table, and we are fortunate today to be working in a climate of relative methodological plurality. Many forms of inquiring, knowing, and sharing cohabit alongside each other, and increasingly qualified reviewers (Easterby-Smith et al., 2008) are receptive to “honest subjectivity” and creative craftsmanship (Cunliffe, 2011). Moreover, scholars can today access a myriad of helpful articles, blogs, books, and studies with recommendations on implicit and explicit criteria used by editors and reviewers of qualitative manuscripts (e.g., Czarniawska, 2008; Geertz, 1988; Gioia, Corley, & Hamilton, 2012; Locke, Golden-Biddle, & Feldman, 2008; McGaughey, 2004; Patton, 1990; Pope, Mays, & Popay, 2007; Savall, Zardet, Bonnet, & Peron, 2008; Symon, Cassell, & Dickson, 2000; Webb, 2003). Such criteria, namely (a) contribution to theory, (b) novelty, (c) transparency, (d) well-articulated methods, and (e) good writing, are mostly shared by both authors and reviewers (Pratt, 2008, pp. 486-488). 3
Some challenges and tensions have been pointed out, however, such as (a) how to develop new theory while staying firmly embedded in extant theory, (b) the need for detail and transparency on methods, journey, and original data while adhering to standard journal formats, (c) how to provide enough data to permit readers to draw their own conclusions and yet provide interpretation, and (d) the reviewers’ evaluation of the story (lessons) authors try to convey (Pratt, 2008). Recommendations include stylish writing, precise craft and creativity, and the navigation of the tension between communicating and impressing (Sword, 2012). Rich case stories (Burgelman, 1996) and/or detailed storytelling (Boje, 2002; Czarniawska, 2004; Czarniawska & Joerges, 1996; Rosile, Boje, Carlon, Downs, & Saylors, 2013; Weick, 1977, 1995) can help scholars make cognitive and affective sense, and create a compelling context. Daft (1983) proposes to embrace nonlinearity and dwell in poetry, emotion, room for error, and surprise. Cloutier (2016) sees academic writing as a social activity performed at the intersection of several practices: thinking, talking, drawing, and reading. Bansal and Corley (2012) ask for no less than a “unique and inspiring story” (p. 511), and Cunliffe (2010) claims that textual validity is also about “imagination as a means of interrogating the relationship between individuals and the world” (p. 231).
On Writing
Writing is perceived as central to qualitative research: how the story is experienced, (de)constructed, and proposed—and how it is in turn received and interpreted by the reader. While traditional quantitative (confirmatory) methods use confidence intervals, significance levels, Cronbach alphas, bootstrapping, and other psychometric reliability scores as “built-in safeguards for warranting claims” (Barley, 2006, p. 19), qualitative work often consists of “mere” text. So, what makes good scientific writing? This is what this study sets out to investigate. Golden-Biddle and Locke (1993), in their inquiry on what makes ethnographic work appealing, purport that a good criterion to judge texts is the ability to “convince.” They break down convincing along three dimensions that are authenticity, plausibility, and criticality. Research should convince its audience that it is substantive, trustworthy, and purposeful, that it will contribute to the advancement of a field of knowledge, but also that the author is capable of conducting a qualitative inquiry (Lindsay, 2011; Marshall & Rossman, 1999; Odena, 2013).
Moreover, in a narration from an interpretative perspective of science, writers and readers interact and interpret the proposed accounts. Such interpretations create a “fundamental asymmetry” (Golden-Biddle & Locke, 1993, p. 596) between reader and text and thus a degree of independence from the writer. Writing is a way of knowing, a “method of inquiry” (Richardson & St. Pierre, 2005). Texts are invitations to readers to participate actively in this process of knowing, to combine propositions and interpretations at hand with separate assertions from their own world (Black, Freeman, & Johnson-Laird, 1986) and thus come to learn and to grow by questioning many taken-for-granted ideas and beliefs about that world. Writing is therefore not just “mopping up” (Richardson, 1994, p. 516) once the research is done. Writing is researching. Such writing practice can be learned from theorists of fields that have a head start over management science in terms of the construction and interpretation of texts that deal with contextually grounded social experiences from the perspective of the actors and their sense making: literary criticism (e.g., Booth, 1961), philosophy (Melles, 2008; Rorty, 1982), anthropology (Geertz, 1973), linguistics (Iser, 1978), and psychology (Black et al., 1986). Convincing emerges not only from what message a text conveys, but also from how it conveys the message, that is, how it is crafted.
In this study, we examine some works of management and organizational scholars who have published their exploratory research in fine journals. Specifically, we explore the following questions: How did they convince? How did they deal with the tensions between expanding iterative methods, daring creative tools, and yet conform to “genre constraints” (Barley, 2006, p. 19)? How did they share their rich micro moments of insights, playfulness, and logics of action (Holt & den Hond, 2013), while convincing in terms of scientific rigor? How are their methods, thought processes, labels, and assumptions made transparent and plausible to the reader? What can we learn from them? We strive to make a contribution by illuminating emerging criteria, conventions, templates, and patterns of qualitative inquiry, and by selecting and examining specific studies from which we believe many scholars can learn. We purport that qualitative research can be narrated with creativity and imagination, in line with its epistemological stance, and still be considered meaningful and academically rigorous by reviewers and scholars.
To substantiate this claim, we first present a synthesis of seminal qualitative scholarly work published during recent years, and highlight principal stratagems that authors deploy to convince, in terms of both substance and scientific rigor. Grounded in this synthesis, we frame a categorization scheme of convincing writing, that is, parameters on which readers can engage with a given research text. For each of these, we present examples and a set of concrete, actionable writing heuristics and strategies that scholars might find helpful when they work toward bringing compelling qualitative studies into the zeitgeist of management and organizational science.
In Search of Meaning, Patterns, and Inspiration
Our long-standing passion for qualitative research, our own practical dabbling, some helpful reviewers, and finally three independent literature reviews on qualitative research in the 1990s and 2000s have yielded our sample of inductive studies that made it—between 2009 and mid-2015—into the following academic outlets: Academy of Management Journal, Administrative Science Quarterly, Organization Science, Journal of Management Studies, Organization Studies, and Strategic Management Journal. Our sampling rationales were to examine generalist management journals that accept a variety of research approaches. These journals are considered “leading” both by the academic community and in terms of the Financial Times ranking of business schools. 4
How did these studies convince? We set out to understand how these authors articulated their practices, observations, and insights (Gephart, 2004) in a manner that reassured reviewers of both the scienticity and the noteworthiness of their contributions. We scanned through their studies for logics, patterns, perhaps standards, strong stories, and creative, generative, and candid gestures. This was challenging because what we were looking for could often not be captured in simple keywords. Ultimately, what inspires, motivates, and triggers insight is in the eyes of the beholder, in this case ours. We each independently pulled out compelling narratives, research designs, and visualizations. We organized our findings in different manners, and compared them. We shared, negotiated, and eventually converged on a “sample within the sample” of 29 studies, which for a variety of reasons touched us, caught our attention, and could help answer our research questions (Table 1).
Selective Sample of Qualitative Studies in Leading Journals (2009-2015): Remarkable Characteristics and Actionable Heuristics.
The studies at hand represent a broad spectrum of ontological, epistemological, and methodological approaches, ranging from ethnographic narratives to grounded, coded aggregations and pragmatic case study approaches; from personalized to very neutral, “realist” writing styles; from conservative linear designs to works with large degrees of structural and methodological independence. However, they display one common characteristic: a masterly way in which they negotiate the many fields of tension and the dilemmas of qualitative research (see Pratt, 2008). We believe most authors, rather than seeing dilemmas or dichotomies (which would imply either/or choices), perceived opportunities and saw dimensions, continua of possibility on which one could discourse, negotiate, and navigate, all to progress and to find an optimal balance between different epistemic, processual, narrative, and/or technical determinants. We reorganized our sample several times, striving to find a suitable aggregation logic to the emerging writing (and reviewing) conventions. We took individual notes and memos (some coded, some narrative, some schemata) and shared them; we asked some of our doctoral students to do the same. We then collectively de-/re-/co-constructed them and eventually identified five continua that had likely informed the substantive contributions of the papers at hand—and their acceptance:
A confident, clear, and candid rhetoric. Different narrative styles (from highly personalized to visual, to realist) that elegantly combine both journey and resulting theory; tangible, plausible bridges between first-order voices from the field and rigorous second-order conceptualization; enthralling and genuine, albeit “conceptually-mediated” (Van Maanen, 1988) accounts of the social experience at hand. A solid and transparent methodological craftsmanship. Adhesion to canons (e.g., in the basic flow from introduction to problem statement, methods, data, discussion, and conclusion) on the one hand, and deviation from them on the other hand; liberties taken with methods and tools, namely in many creative ways that data and its interpretation is iterated, framed, and displayed. A compelling, lively authenticity and energy. Energy in narrating the meanders of (particularized moments in) the everyday realities encountered in the field and conveying both the rich cognitive and emotional process of immersion and the synthetically substantiated theories—and elegantly oscillates between them. A strong reflexivity. Reflexivity in terms of the writing as a cognitive process designed to reflect and relate to the readers’ worlds, yet asymmetrical enough to attract the audience’s full attention and scrutiny, and even go as far as pushing it to self-reflect and reexamine taken-for-granted assumptions and beliefs. A touch of imagination, some brave abductive leaps. A trust in one’s mind, instincts, and intuition to trigger generative, insightful moments; the sagaciousness to understand and value such moments not merely as introspectively satisfactory, but as data in their own right, and, conjectured from them, novel sketches of seeing and organizing the world.
The common denominator in our sample is thus an ability to communicate beyond a fundamentalist either–or logic in the evoked qualitative research writing. To perceive and express these fields of tension not as constraints, but as spaces of possibility on which to oscillate and find a purposeful hermeneutic balance between objectivity and subjectivity; conformity to formal research standards and structures and creativity faithfulness to the real, iterative process; process and theory; personalized narration and more neutral, realist posture, and so forth. These five writing practices—and the interplay inside and across them—were the essence from which the authors built their substantive contribution—and gained the engagement, and ultimately the conviction of reviewers and readers.
We now briefly comment on our sample as a whole. We then address the aforementioned categorization scheme of convincing writing, sharing findings for each, and follow up with a set of actionable heuristics in dedicated tables. These practices, while presented separately, are not mutually exclusive. They have a distinctive conceptual character, but they are also overlapping and interrelated, and authors may use similar writing tactics to gain in authenticity, reflexivity, or novelty. For example, the transparent and precise way some scholars deal with their translation of field (first-order) voices into theoretical (second-order) language, is a matter of rhetoric, but also of authenticity, in that the reader will or will not engage with proposed aggregation and translation and consider them plausible, and finally of methodological craftsmanship. A lively account of a field situation helps to convince in terms of authenticity, of “having been there,” but of course that same disarming “conceptually-mediated” (Van Maanen, 1988) liveliness, is a testimony of and call for reflexivity—and imagination. What incites reflexivity is in turn only a small step away from imagining new, other possible selves, futures, and realities. Our understanding for now is that the practices might be interconnected by epistemological and methodological craftsmanship. We frame these and their principal relationships in Figure 1.
Five characteristics of convincing writing and their interplay.
Convincing Writing: How Those Who Made It, Did It
Approaches in our sample vary greatly to include many forms of narratives, process models, iterations (between concepts, quotes, and definitions), temporal perspectives, aggregation levels, decision-making patterns, further research instigators, pathway analyses, taxonomies, theorizing and framing, contrasting logics, historic contexts, network analyses, a wide array of graphic displays, mixing of tools and methods, extreme cases, multilevel analyses, diverse forms of triangulations, data tables, textual theory building, and well-documented coding and data structuring procedures. Many authors adopted a classical linear design, along a description of field, research interest, and theoretical foundations, then explicit or implicit references of methods, data collection practice and analytical steps, and some form of theory, illustrated through framing figures and tables. Constructivist articles are distinctly longer: They contain extensive explanations (legitimizations) of philosophical underpinning and process and propose many original quotes when linking raw data to constructs, concepts, and contexts. Many studies do not expatiate on their epistemological or theoretical underpinning at all. Only few authors explicitly position themselves as performing their analysis from a constructivist stance. We now present the five categories of persuasive writing practices, each with some examples, and substantiate some of the observed tactics and heuristics in an actionable way. We noticed that many of the papers in our sample (Table 1) excelled in several of these five categories.
Rhetoric
All published studies convince through their rhetoric. Rhetoric is the expertise of discourse, the capability to inform, persuade, and engage one’s audience. Whenever scholars frame ideas for a dedicated readership, they engage in rhetoric. Techniques to use and understand rhetoric are manifold and include speech analysis, persuasive appeals (e.g., pathos), puns, canons, tropes, figures of speech, and medium (McLuhan, 1964). One key tension seems to be between the rigorous “normalized” way of writing (e.g., second-order or aggregated themes), and a conscious narrative personal identity that may persuade the reader that there is a real foundation for what was written, as opposed to simple manipulation or paralogizing of the data or not mentioning deviant cases or crucial information.
In our sample, heuristic choices of convincing through rhetoric include the following:
An explicit awareness of the centrality of rhetoric. This includes a conscious and explicated writing effort in terms of storytelling, vocabulary, syntax, style, rhythm, energy, attitude, and so on. Also, titles and subtitles go well beyond the mere identification of subject and content: Every word is handpicked to convey—as a tag line in an advertisement—the phenomena under study in their full richness, complexity, and cultural identity. Dealing with length emerges as a key competency. The authors have managed to be lively, while synthesizing the data to minimize the repetitions and redundancies that can come with iteration. A coherent posture and engagement. This coherence emerges in terms of a chosen epistemological stance, role of the researcher, and/or corresponding voice, style, vocabulary, visualization choices, and methodological and theoretical claims. Paradigmatic awareness also concerns all interactions with the actors in the field and understanding the paradigms embedded in the empirical conversations and texts (Rostron, 2014). Both Paquin and Howard-Grenville (2013), in their network orchestration study, and C. Wright, Nyberg, and Grant (2012), on identity genres, convince (despite their picturesque, emotional titles) with a radically conceptual style, using traditional language, such as “we analyze,” “we focus on,” precise methodological indications “embedded case design,” “theoretical sampling,” and a factual presentation of findings, albeit peppered with vivid and precise field quotes. In contrast, Costas and Grey (2014), on the temporality of power, use a more conversational narrative, like “what if, instead, they are imagining…” and “our ideas are inspired by….” All studies deploy strong quantification heuristics, such as “amounting to over 1,000 pages of transcript,” “37 days in the field,” interviews “between 50-120 minutes,” and “transcribed within two weeks,” to convey rigor and process control. Specific techniques and heuristics that bridge between conflicting and opposing goals. These techniques include novelty and familiarity, journey and theorizing, qualitative phrasing to the fullest and concessions to perceived-to-be traditional, nonexploratory, reviewers. For example, both Michel (2011, organizational change) and W. K. Smith (2014, dynamic decision making) combine a refreshingly noncomplex, bold, and energetic first-person narrative on the one hand, with many pages of highly detailed, matter-of-fact, clinically argued explanations on the methodology adopted on the other.
5
A presentation of theory in a coherent, plausible, and practical manner. This presentation stimulates readers to build and/or test it further. Theories can be narratives or framed and visualized. In either case, the reader engages and adds to the theorizing process (Berger & Luckmann, 1966; Geertz, 1973) and through this engagement decides whether to trust it, even though “how, precisely,…and in what fashion are far from clear” (Van Maanen, Sørensen, & Mitchell, 2007, p. 1153). But, we conjecture with some confidence that writers benefit from simulating this virtual conversation, or co-construction with the reader, in that they anticipate it. Noninvolved peers (or other stakeholders, e.g., practitioners) could read their texts and discuss them. Many authors report on such tactics under the overall terms of triangulation and member checking. Narratives have strong sense-making power and convey authenticity; visually framed theoretical contributions (integrated frameworks, process models, figures, maps, etc.) act as boundary objects (Thompson, 2005) that fit into the epistemic parameters of a larger audience. Many studies have paid great attention to the visual presentation of their theories in an actionable, usable way. The medium is also the message (McLuhan, 1964). Some opportunism is not only allowed but also imperative if one has something of relevance to say.
The authors’ vocabulary implies an exploratory approach (since the sample was chosen on these grounds), but diverse rhetorical concessions to the mainstream order of writing are illuminated: reassuring adjectives such as “tangible,” “robust,” “grounded” to demonstrate rigor; “surprising,” “remarkable” to show noteworthy deviation from extant theory; presence in the field “extended to,” “more than” n months, to express the intensity of the field immersion; and quantifying terms like “nearly all,” “mostly,” “significantly,” “exhaustive,” “correlate” to address the statistical mind (see also Brannen, Piekkari, & Tietze, 2014; Golden-Biddle & Locke, 1993). Up to eight pages of clinically detailed explanation and quantification of method (number of hours, transcript pages) and abundant use of known terms such as “theoretical saturation,” “open/axial coding,” “triangulation” (Star & Griesemer, 1989) were used to bridge between their abductive approach and a context perceived as essentially positivist. Authors did include the self in their interpretative accounts, but none went as far as the strongly personal styles and nonlinearity of journeys in earlier seminal studies, such as Dutton and Dukerich’s (1991) account of the homeless or Van Maanen’s (1988) tales of the field. Further common grounds are a compelling introductory section and a capacity to sustain a narrative tension as they build up a plot with their data, derive from it an explanation, and conclude with a powerful theoretical narrative (e.g., see how Ariño & Ring’s, 2010, storytelling intertwines data, process contingencies, and theory narratives in an enthralling and precise manner).
There are key moments when the scholar needs to establish authority for the reviewer, and later the reader. This authority is not established through measurable validity and reliability parameters, but by sharing one’s analysis of the many research and rhetorical decisions all along the process (e.g., Jonsen et al., 2013), starting with the discovery of the research story. This is a both a skill, and a very personal process and validation of choices where peers can bring clarity to the writing and monitor bias. Narrative researchers must make such precise choices. They need to determine whether storytelling is an appropriate ontology, to find out what they seek to know, and which, among available generic storytelling paradigms, narrativist, living story, materialist, interpretivist, abstractionist, and practice (Rosile et al., 2013, p. 560), best fits the research interest and context. Narratives in our sample were essentially of two, mostly intertwined paradigms: practice and interpretivist.
For practical referencing, we have summarized the most powerful rhetorical heuristics of convincing in a table (Table 2).
Convincing with Rhetoric.
Craftsmanship
By epistemological and methodological craftsmanship, we mean first and foremost the conceptual and technical expertise of the methods and tools applied, and their interplay, as they are presented to the reader. Trust in methodological soundness and rigor is typically intimated through compliance with defined validity and reliability criteria and a design along structures such as introduction, theoretical underpinning, methods, data aggregation, theoretical discussion, and so on. In qualitative research, more subtly and for lack of objectively measurable criteria, the work needs to convince more holistically, as a balanced ensemble of method, epistemological stance, research intent, journey sharing, rhythm, and language, but also empathy, honesty, criticality, and independence. One key tension seems to be between reassuring the reader that the process is properly executed (Knorr-Cetina, 1981) and the methodology mastered and the relatively small text space “allowed” to explain this convincingly. In other words, how much space should be dedicated to describing the methodology and process? This varies from journal to journal, and it is perhaps driven more by journal norms than author choices. See also the previous section on rhetoric, items 3 and 4.
The studies that we examined could not be more different, and yet they all have this tight fit and inner coherence between culture, structure, process, and substance. Elements of convincing through craftsmanship include the following:
A masterly epistemological and methodological savoir faire. Paroutis and Heracleous’s (2013) study on discourse in our sample displays such savoir faire. Following a self-designed four-stage approach that draws from exploratory research, phenomenology, and hermeneutics, and intertwining it creatively with a savvy mix of tools in a double-loop parallel analysis that is eventually connected for patterns to emerge, they manage to knit a cloth of precision, beauty, and substance. In a sense, they demonstrate the need to master methodologies to juggle them, to deviate from them, with such confidence, and end up with something the reader experiences as “right,” as convincing. The rhetoric is finely chiseled, and the tables and figures are in harmony with the text. Everything is precise and detailed, and yet extremely parsimonious. A strong independence and intellectual probity. Ariño and Ring’s (2010) study on fairness in alliance formation is a noteworthy example of a candid narration of how field expectations (they thought they would be observing a successful alliance formation in real time) are proven wrong (the alliance was eventually aborted), how researchers adapt and objectives shift (trust in alliances became an issue), and how data access and collection intentions are renegotiated (lively accounts of tactics and ruses to get and keep field access). Michel’s (2014) monumental 10-year ethnography among investment bankers is written in a much more controlled, result-oriented manner. Rather than sharing contingencies, she explains her research field and analytic strategies in a radically matter-of-fact, parsimonious manner. A logic of theory presentation with the reader in mind (customer orientation). Writing is also being smart about wanting to be read: appealing storytelling with a touch of emotion, impeccable conceptualization, rhetorically built and argued, engaging the reader’s reflection, actionably packaged in smart tables and/or helpful figures and/or graphs, integrated models, and hypotheses, effectively connected back with original quotes and with the literature. Such concrete visualizations of process steps and theory stimulate readers to build and/or test it further. Clark, Gioia, Ketchen, and Thomas (2010) provide two tangible models that can be built on by other researchers, a process model on identity change (p. 426) and a framework destined for testing propositions on transitional identity (p. 433). Moschieri (2011) also delivers a palpable process model (p. 387) and a set of propositions ready for testing. Michel’s (2011) well-designed integrated framework of organizational control and body action roles (p. 352) is easily adoptable for testing, as is A. L. Wright and Zammuto’s (2013) model on institutional change processes in cricket.
Data coding is not a necessary part of the procedure in exploratory research. Good narratives are powerful sense makers. But coding, though it neither interprets nor builds theory, seems popular, as indicated by our sample: Scholars appreciate the efficiency of coding when faced with vast amounts of narrative data in which they hope to find what their respondents’ “real world” is like. The same goes for overview tables, frequently required by reviewers and thus probably a smart write-up strategy to enhance publication chances, despite occasional voices that regret such standardization and call for more pluralism (e.g., Bansal & Corley, 2011). Barley (2006) purports that methods are particularly interesting when they get “close enough to behavior to show how people wittingly and unwittingly build and maintain their social worlds” (p. 17). Such closeness, and a smart writing strategy to handle the myriad of data (over 20 pages of notes per day), can be observed in Levina and Orlikowski’s (2009) study of shifting power relations in a fast-growing strategic alliance context, where, on the basis of their highly detailed every day observations, they then performed a critical genre analysis to untangle and explain how the power relation among participants on the project were being reproduced or challenged. This analysis involved examining how control over genre enactment was exercised (e.g., who set the time and location of a meeting, who was responsible for the agenda, etc.), as well as who had competence in and participated in the various genre enactments, and how. For each of the genres identified in our genre analysis, we specified the status distinctions implicated in its enactment. (p. 678)
Convincing with Craftsmanship.
Authenticity
The dimension of authenticity is principally about the process, that is, the researcher’s journey. Convincing, here, comes from truthfulness to the field: trust in his or her engagement and empathy that he or she has truly “been there,” with an open mind-set, gotten his or her hands dirty, and hung around a long while, developed a deep and detailed understanding of the setting at hand, and so forth, so as to be trusted as a legitimate ambassador, translator, and interpreter of the setting under study. Also included is the notion of plausibility. Even though authenticity and plausibility are not the same, they are related. We found that when we as readers had established trust in the researcher’s competency and probity, we were willing to engage, to be receptive to propositions, even if these were a priori exotic or contestable (see also Fox-Wolfgramm, Boal, & Hunt, 1998, for an excellent example of this).
One main tension seems to be between how the study really developed and writing it in a way that captures the attention of the reader and is publishable. Richard P. Feynman (1965) phrased it bluntly in his 1965 Nobel Prize lecture: We have a habit of writing articles in scientific journals to make the work as finished as possible, to cover up all the tracks, not to worry about the blind alleys or describe how you had the wrong idea first, and so on. So there isn’t any place to publish, in a dignified manner, what you actually did in order to get to do the work.
Elements of convincing in our sample through authenticity include the following:
A lively, honest, detailed, and engaging narration of the field. This includes the actors, their realities, and the relations that the researchers were able to build with them to build trust and deeply engage the reader and an understanding of narrative types and levels: macro level, holistic; micro level in situ; individual practices, praxis, or practitioner narrative; and so on (Fenton & Langley, 2008). A clear (honest) explanation of the research design steps and procedures. All tables display forms of stepwise data aggregation and contain some deeper levels of meaning, links between properties, and interpretation. Categorization is used to describe phenomena and to distinguish from new data or to make inferences (Dey, 2007) and is usually the core level of finding and, therefore, central to presentation. Codes are fractured into a conceptual level by categorization (Glaser & Strauss, 1967), that is, texts are highly synthesized and conceptual here rather than stories from individuals or groups (Strauss & Corbin, 1998). Sometimes they are enriched with original quotes that help the reader understand categories, but add length. In sum, the generic prototype study here presents tables with data sources, and aggregate tables with first-order and second-order steps and some type of “aggregation dialectics,” preferably with representative proof quotes built in as the backbone of the theory—in the cases where theory is developed. Process, theorizing, and framing are visualized for more power (Mengis & Eppler, 2008). Category labeling is almost uniform: (a) organizing categories, examples, final coding categories; (b) first-order (informant) concepts, second-order themes, and aggregate analytical terms. These exact terms inform most inductive studies published in the past two decades.
6
Writing up research in terms of role model studies seems to be a paying strategy, also as a learning process for young scholars (Pratt, 2009; see also Worren, Moore, & Elliot, 2002). Using a recognized template increases the likelihood of publication—and of being read and cited: the most cited study of our sample, Andriopoulos and Lewis’s (2009) comparative case study on ambidexterity, is also, seemingly, the most objectivized and most rigorous.
7
Forms of validation/proof of scientific rigor in line with the chosen stance. These include respondent or member checking or other forms of triangulation techniques (Jonsen & Jehn, 2009). Some studies mimic traditional mainstream inquiry in that they objectivize their analysis process to the extreme. With this, they reduce the interpretive and narrative potential of the method chosen, in favor of some type of proof of validity and generalizability. Others, in line with a more transformational mind-set (Cho & Trent, 2006), allow for reiterations and expansions of themes and findings. It is difficult to know how far objectivizing procedures are caused either by the authors’ lack of familiarity with interpretive and narrative approaches and thereby lack of self-confidence to write it up inductively, or by their expectation of and/or actual reviewers’ comments asking for more validity of the deductive, traditional kind. Most papers use a linear, organized form to suggest a controlled research process and validate through craftsmanship, by quantifying and qualifying research effort and design (see rhetoric) and by citing examples of scholars who performed a similar approach. Huy (2011), in an energetic, first-person narrative on middle managers’ emotions, sustains the narrative theorizing through sheer volume (over 200 interviews), and triangulation through observation, focus groups, and internal documents, detailed in a table (Table 2, p. 1395). An actionable, credible, and pleasurable account of the field, as perceived and lived by the actors under study. The authors have gone there with humility, empathy, a subtle touch of humor, and an open mind. We also find evidence of this open mind-set—often in smart, precise tables—which displays receptivity and how our own a priori beliefs were left at home. For example, Reinecke and Ansari’s (2015) rich and lively dual temporal study convinces with concise tables (especially Table 3, p. 632, composed of respondent quotes on contrasting temporal structures). A capacity to actively engage readers in the narrative. This is important not only for them to discover the realities of the field at hand, but to reflect on propositions that are a priori beyond their sphere of the probable, to rethink and question taken-for-granted assumptions and ideas (also reflexivity). Van Dijk, Berends, Jelinek, Romme, and Weggeman (2011), for example, take institutional theory head-on in their study on radical innovation by proposing a series of novel and counterintuitive arguments on how embedded agents can overcome legitimacy crises to facilitate radical innovation: In other studies, the presence of competing demands from multiple institutional constituents has been presented as a dilemma to overcome…similarly, whereas Oliver (1991) argued that environmental uncertainty stimulates organizations to conform to established institutional logics, our findings suggest that ambiguity provides opportunities for transformation. (p. 1509)
Convincing with Authenticity.
The degree to which researchers apply some of the heuristics above depends on their choice of researcher role, voice, and style. A “confessional” style (Van Maanen) would go quite far in sharing personal learnings, introspections, mistakes, iterations, changes in research tactics, and so on, whereas scholars adopting a “realist” posture would camouflage or tone down such moments.
Reflexivity
Reflexivity concerns the research in all aspects, but principally the result, the substantive contribution. Does the final writing propose something distinctive, a singular contribution? Does it position this contribution offering space to reflect? Reflexivity—the act of critical (self-)reflection on biases and preferences—has been central to qualitative inquiry and impactful research (MacIntosh et al., 2017), and more so since the debate on the lack of predictability of the management world, and the fragility of the increasingly dynamic and transitory nature of organizations has emerged and taken shape with such terms as postmodern, linguistic, and narrative (Denzin, 1997; Lyotard, 1984). Some even argue that narrative, rather than a method of inquiry, is a process of understanding, of becoming; and that narrative is, therefore, reflexivity (Lyle, 2009). There is a consciousness of the coexistence of multiple, albeit connected realities and diverse cultural ways of being and that these realities and ways are constructed, deconstructed, and co-constructed by the actors involved. Reflexivity is strongly linked to authenticity and feeds on some form of articulation of belief, doubt, and experience and how these are resolved to arrive at stable propositions (Locke, Golden-Biddle, & Feldman, 2008). Convincing arises from an interplay of varying points of view—that of the tensions in the method, the actors in the field, the researcher, his or her colleagues, the reviewers, and ultimately the readers. The value of a particular inquiry arises not from its approximation of an ultimate truth, but from its usefulness as an instrument in the social debate leading toward such “truth” (Fendt & Sachs, 2008). Practice issues can no longer presume to be resolved by validated nomothetic knowledge, but must be researched, particularly in co-construction, often across communities (e.g., scholar-practitioner).
Reflexivity here builds on Schön’s (1982) reflective practice and extends it, in that beyond the reflection-in-action and the reflection-on action (post-thoughts) of both the respondents and the researchers themselves, it establishes a quasi-dialectic with the reader by creating a space for others to engage in critical thought by confronting them—through a powerful narration, a plausible aggregation of findings, and a cultural critique—with unusual, unfamiliar, perhaps contestable, and yet irrevocable propositions. In our sample, it was mainly narratives on issues connected with power that engaged the reader in this way: for example, Korczynski’s (2011) study on the dialectical sense of humor at the workplace, in which he frames the power and the (liberating, soothing, and/or humiliating) potential of humor. Canato, Ravasi, and Phillips (2013) begin their study with a respondent quote that pulls the reader right into the dilemma under study, the field reality, the stakes, and the reigning culture: Six Sigma had this terrifying thing of not wanting errors. But if you do innovation the way we do, pure risk is something you have to be able to admit and accept…. With all these gate reviews, Post-it would have never seen the light…and if Six Sigma would block a hypothetical new Post-it, then Six Sigma is not for us. (p. 1724) All data were transcribed, usually within the two weeks of data collection. In the first step the researcher read and re-read the data, whilst simultaneously reading studies on the various themes emerging from the data. The process of data analysis and emergence of themes was iterative. Initially…. The researcher then reread the data to specifically look for…. Following this, the researcher systematically analyzed her data with respect to…. (p. 920, italics added)
For practical referencing, we have summarized the principal heuristics of reflexivity in a table (Table 5).
Convincing with Reflexivity.
Imagination
By imagination, we mean the capability of authors to capture the very essence of a social reality and to hold it up before the reader, in the way of writing, like a match to light a candle. It requires a grain of folly, a vehement inquisitiveness, and sensitivity, spontaneity, inventiveness, intuition, insight, and passion. We look at the research interest. Is the phenomena grasped at a particular angle, does the field design lift the study out of the often-banal mainstream, does the researcher seek out “extreme” respondents, are connections surmised in unusual places, and is there any stepping out of the comfort zone, pragmatism, inference to the best explanation (Argote, 1999)? It also concerns a particular resourcefulness in the face of adversity, when things go wrong, when the field closes, when contingency disrupts the research context, and so on. Furthermore, we think of Weick’s theory construction through disciplined imagination (Weick, 1989), by paying particular attention to facts, values, goals, memories, or questions we might forget, fear, push aside, or neglect. Such behavior might necessitate the willingness to unlearn (Kleinsasser, 2000), the ability to shed familiar practices, methodological templates, and so forth and venture into the unknown. Weick (2002) speaks of “dropping one’s tools to regain lightness and agility” (p. S15), which relates back to good craftsmanship, as it requires great mastery of research methods to be able to deviate from them.
Did we find evidence of such behaviors and heuristics in the studies of our sample? Yes and no. Many studies had elements of creativity, mainly in terms of methodology (see Paroutis & Heracleous’s, 2013, innovative multimethod approach). We see resourceful, creative ways to visualize complex relationships (e.g., Paquin & Howard-Grenville, 2013, who combine process observation with empirical network analysis devices such as mapping and snapshots, and Harrison & Rouse, 2014, who have imagined conceptually and visually appealing illustrations for their models of elastic coordination in creative work), or smart designs (A. L. Wright & Zammuto’s, 2013, highly creative desk study), or unusual, inspiring ways to present findings in tables and figures (e.g., Lipparini, Lorenzoni, & Ferriani, 2014, with their unusual and exciting graphics to display knowledge flows between firms; Reinecke & Ansari, 2015, with their culture-grounded clock time/process time approach, and their astute theorizing process model of ambi-temporality; Clark et al.’s, 2010, transitional identity evolution graph). Others have shown great creativity in negotiating access to difficult fields (Harrison & Rouse, 2014, a prime example of creative use of an extreme case setting), or in engaging the respondents, that is, by making them draw or mime their realities, ambitions, fears and relationships, rather than interview them (E. Barrett & Bolt, 2007). But rather than verbalizing these situations, that is, recognizing them as data and/or as strategic moments in the theory-building process, authors have generally toned them down.
In contexts of uncertainty, knowledge and opportunities are not so much sought but constructed based on insights. Insights often lie in unexpected places, in intersections, in micro-caches, in the liminal, and/or in holistic perspectives. We claim that theory is generated at least partly in the unconscious (Crawford, Dickinson, & Leitmann, 2002), and emerges at certain moments of reflection, or of play, or of absent-minded musing (Peirce, 1992), or while washing the dishes. These moments, these hunches, have been referred to as conceptual leaps (Klag & Langley, 2013), or generative moments (Carlsen & Dutton, 2011), moments at the liminal between seeing/understanding a novel construct, and being able to put it into words. These are small frequent occurrences, not one singular magic insight (Van Maanen et al., 2007). They can be provoked. Heuristics can include empathy, a genuine interest in and curiosity for people and phenomena; a strong bias for action and experimentation (Burch & Goshal, 2004), that is, work with what you have and the readiness to accept disgrace, ridicule, embarrassment, confusion, and deceit (Simard & Laberge, 2015); multiplying encounters with diverse, unusual people in unusual places, and co-construction of ideas; embracing contingency as an opportunity rather than a problem (Harmeling, Sarasvathy, & Freeman, 2009); and hanging out, doing nothing, musing (Peirce, 1992). By creating, through such behaviors, a climate, a petri dish receptive to chance, happenstance, or serendipity, hunches can become process inherent. To invite serendipity, we need to leave some space, some improvisation, some oxygen in our research design and behavior. So, researchers are advised to practice such heuristics as generous hanging around in the field with no apparent purpose, being there before and after the moment when we are supposed to be there, and multiplying encounters with diverse actors, mainstream but also “extreme users” (Brown, 2009). And we need sagacity—a general openness and motivation (Weisenfeld, 2009), the capacity to multiply conceptual leaps and weave them together (Mintzberg, 2005), to discover from a very small number of observations and the opportunism to seize it when it comes our way.
Do we get published because we “go mainstream” or because we are creative? One key tension seems to be between how much imagination authors use throughout the study process and the extent to which this can be written up at all, and/or in a way that defies the norms. Although we found glimpses of high creativity in our (selective) sample, authors must wonder, like we did, why anybody would take the risk of “being too different.” When today’s canon is a clear explanation of the research process, reference to a handful of seminal methodology scholars, a somewhat repetitive language (labels), table structures and theory-building levels on tables, packaged into a classical, linear write-up. In other words, while some (few) lively storytelling narratives have broken ground, digging deeply into the true potential of inquisitive and exploratory analysis, many may not make it into leading management journals yet.
For practical referencing, we have summarized the heuristics of imagination in a table (Table 6).
Convincing with Imagination.
Discussion
This is a study on a sample of a large variety of qualitative approaches. We based our independent choices (Sandelowski, 1995; Schatzman & Strauss, 1973) on our experiences and our passions as qualitative scholars in a continuing strive to become subjective thinkers in a Kierkegaardian sense, and at the same time armed with contemporary pragmatism and craftsman’s integrity that makes us question and doubt our work and relentlessly reexamine our assumptions (Ryle, 1968). We have focused on the matter of writing qualitative research, as a central element of interpretation and meaning making. We have set out to distil what constitutes persuasive writing, and how it can appeal to and convince readers, including reviewers. We have displayed our condensate along five key writing practices, emerging from reading, debating, and negotiating the texts at hand, and we have proposed, for each, a set of actionable heuristics.
We purport that management researchers could and should matter not merely as researchers, but as human beings who set out to understand other human beings, and who through that probing come to discover new and better selves, if not different, and reexamine stances, beliefs, and ideas, and stimulate our readers to do the same and understand the worlds that we shape and that shapes us. But to make a difference, we must be read, just like composers must be heard and (visual) artists must be seen. And this requires persuasive, and often amazing, writing. We must also be capable of writing relevant knowledge (Courpasson, Arellano-Gault, Brown, & Lounsbury, 2008)—a subjective construct that starts with asking good, actionable questions on many urgent organizational and societal issues at hand (Jonsen et al., 2010). Writing is central to such research and intricate in its process. We examined how scholars who have passed the test of publication have negotiated the tensions along five distinct but related practices of convincing—see Figure 1—and proposed some actionable heuristics for each (Tables 2 –6). Rather than proposing a template, we suggest and maintain that each writer must find his or her personal style and footprint in line with his or her research interests, personality, stance, empirical field, theoretical ambition, and targeted audience. We offer this study in the hope that it can inspire scholars to venture more often and more deeply into the challenges and pleasures of qualitative writing and thus discover its infinite generative power.
Once written, qualitative inquiry may still be hard to publish, but once it is, it is intensely read, appreciated, and cited. 8 So where could qualitative inquiry be going? Early seminal studies, thanks to their detailed process descriptions, have begun to serve as role models well before they even entered the research manuals: Dutton and Dukerich (1991), Larson (1992), Van Maanen (1988), Charmaz (1983), Orlikowski (1993), Isabella (1990), Eisenhardt (1989), and later Corley and Gioia (2004), to name a few. Later, more explicitly methodological contributions proposed clear and rigorous conceptual underpinnings to the different emerging approaches, which in turn gave confidence to many novice scholars to take the path of qualitative inquiry, including a much-needed focus on studying processes and temporality (Hatch & Schultz, 2017; Langley, Smallman, Tsoukas, & Van de Ven, 2013; Srikanth, Harvey, & Peterson, 2016). These (comprehensive elements of) boilerplates bring craftsmanship and confidence to scholars, but also canons of rigor, substance, and scienticity to supervisors, editors, and reviewers.
This perhaps comes at a price insofar as the writing (up) of qualitative research is today expected to look like an “Eisenhardt template” or a “Gioia template” (term used by Langley & Abdallah, 2011); “the Gioia method for building grounded theory” (as recently phrased by Meister, Sinclair, & Jehn, 2017) or a “Langley template.” Qualitative designs are becoming kosher—on the condition that they display a recognizable structure that typically includes such terms and labels as “data,” “coding,” “analysis,” “triangulation,” “themes,” “aggregated dimensions,” and so forth.
It appears that we still tend to perceive scientific method as a tool in a quest for objectivity and, “by innuendo and osmosis” (Stout, 2001), associate objectivity with linear cognitive processes. Subjectivity—creativity—is by contrast undesirable. But research can—and must—be creative. If we as scholars want to take on the big questions, if we want to contribute, however modestly, toward addressing the challenging societal, cultural, economic, and technological impacts of an accelerated digital world and address some of the daunting dysfunctions and asymmetries at hand, we might need to think differently about the world, perhaps radically differently. When studying novel phenomena we often have neither much data, nor theory to test. We are thus doubly in the unknown and must imagine new ways of researching new ways of managing and organizing. Abductive reasoning, for example, is one form of research that can do that, and go beyond explanation (“what is”), toward imagining new possible futures (“what could be”). 9 See also Van Maanen et al. (2007).
Moving forward in the field, we could try to break down some of the silos of our disciplines and venture beyond our comfort zone (Cheng, Henisz, Roth, & Swaminathan, 2009; Lee & Brown, 2013). We have a responsibility to work hand in hand with scholars from other disciplines, including our own historic reference disciplines (Locke, 2015), practitioners, and a wide variety of other stakeholders, across functions, cultures, and continents (Heracleous, 2011; Pfeffer, 2007; see also Kuhn, 1978; Tashakkori & Teddlie, 2010). The big questions, those that matter, may require it. This is a challenge since social realities are, as stated, messy, stressful, idiosyncratic, irrational, context-bound, and contingent. Interpretive techniques of many kinds, some accounted for in this paper, can “provide a deeper and richer understanding of the issues under investigation” (Van Maanen, 1979, p. 520), and unpack the phenomena of interest (Birkinshaw, Brannen, & Tung, 2011), such as complex, fleeting, and contingent moments of managerial agency. Suddaby (2015) and Sword (2014) suggest that academics could improve their writing by taking inspiration and analogies (Oppenheimer, 1955) from other genres (see Lord, Dinh, & Hoffman, 2015, for a recent example). We advocate such a transdisciplinary mind-set, capturing not only what each science can teach us but also what sciences may see together when they collaborate (Khapova & Arthur, 2011). This could also include writing in different and novel ways, and also for practitioners’ outlets—excellent rhetorical practice and communicating with an audience typically larger than academia.
Regardless of the abundance of insightful recommendations available at our fingertips, we remain in a dilemmatic space between the rising demand for qualitative inquiry on the one hand, and a cultural, structural, and technical incapacity of our still dominantly positivist academic world to actually process such inquiry on the other. We believe that this dilemma is temporary. It characterizes all nascent innovations, institutional disruptions, and/or paradigm changes facing a “glass ceiling” (Cotter, Hermsen, Ovadia, & Vanneman, 2001). Innovators deploy rhetorical strategies to construct meanings and justify interpretations (Lawrence & Suddaby, 2006) until a critical mass, an institutional “tipping point” (Gladwell, 2000) is reached, at which point enough individuals in a system have adopted—or learned to deal with—a paradigm, so that its further rate of adoption becomes self-sustaining (Angehrn & Nabeth, 1997). Importantly, systemic shifting is costly (Easterby-Smith et al., 2008). Thus, until now we may have been somewhat comfortably caught in the echo chamber of our own institutionalized voice and norms for persuasive writing. Nevertheless, the tipping point is within reach.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.