Promoting the Value of Discursive Psychology for the Field of Human Resource Development: A Pedagogical Guide for Qualitative Researchers

To date, the study of how health and wellbeing are actualized in organizations and how an organization’s interactional practices shape an environment that is potentially harmful to health is understudied. Much of the research around this topic has centered on personality and individual differences, as well as health and safety or ergonomics. Little understanding exists of how interactional practices might serve to prioritize health and wellbeing. In this paper, we introduce discursive psychology (DP) – a qualitative approach to studying talk and text that focuses on examining what is accomplished through people’s interactional practices. We provide an overview of DP and discuss its underlying assumptions, analytic process, and quality measures. To illustrate the application of DP to HRD, we include data extracts that highlight the impact of question design. To conclude, we point to how DP might afford HRD scholars opportunities to generate new theoretical understandings about organizational practices.


Introduction
In the contemporary era, one of the most pressing concerns for organizations is the health and wellbeing of employees (Guest, 2017;Kowalski & Loretto, 2017), an issue that has been recently magnified with the COVID-19 pandemic. The recent BUPA Global Executive Wellbeing Index showed that organizations based in the UK are planning to increase their spending on wellbeing and mental health by 18%, and similarly the London School of Economics' wellbeing manifesto indicates that wellbeing is now a top priority for organizations and governments. In response to this global health challenge, organizations across the world have had to navigate unprecedented situations and find new solutions to issues impacting the wellbeing of their workforce and operations (Carnevale & Hatak, 2020). Despite the new challenges created by COVID-19, issues related to health and wellbeing have long been concerns for the field of Human Resource Development (HRD) (Danna & Griffin, 1999). Yet, to date, the study of how health and wellbeing are actualized in organizations and how an organization's interactional practices shape an environment that is potentially harmful to health, and, at its worst, toxic, have been rarely considered. In fact, much of the research around this has centered on personality and individual differences, as well as health and safety or ergonomics (e.g., Daniels et al., 2021;Daniels et al., 2022). Thus, there exists relatively little understanding of the place and role of interactional practices that could prioritize health and wellbeing and the ways in which work environments come to be and persist.
In this paper, we present a particular qualitative methodology that focuses on the close examination of interactional practices or language-in-use. 1 Specifically, we introduce discursive psychology (DP)an approach to studying talk and text that focuses on examining what is accomplished through people's interactional practices (Edwards & Potter, 1992;Wiggins & Potter, 2020). This is especially pertinent given that HRD itself has been conceptualized as a social and discursive construction (Sambrook, 2001).
In the 1980s and 1990s, DP was developed by UK scholars, Derek Edwards and Jonathan Potter. Drawing upon social constructionism (Berger & Luckmann, 1966;Burr, 2015), it orients to psychological matters, such as memory or cognition (among many others), as being made visible in and through discourse (Edwards, 1997). Those practicing DP methodology study psychological issues from the perspective of the social actors engaged in an interaction, exploring how psychological constructs (e.g., emotion, agency, identity, etc.) are managed and navigated by those involved (Te Molder, 2015). Thus, from this perspective talk or language-in-use is not considered neutral or as a way of accessing what is inside someone's mind; rather, talk is viewed as performative of social action (Edwards & Potter, 1992). DP leans on other language-based approaches, such as ethnomethodology (Garfinkel, 1967), conversation analysis (Sacks, 1992), and discourse analysis (Potter & Wetherell, 1987), with these perspectives shaping its core characteristics. Indeed, some scholars have even characterized DP has a type of discourse analysis (Lester, O'Reilly, Kiyimba, & Wong, 2018).
To date, DP is relatively underused within HRD, and thus in this paper we present a discussion of DP as a generative methodology for theory development in HRD. Given our substantive work has centered on the study of health and wellbeing, and how disability and ableism are made visible in conversational practices O'Reilly & Lester, 2017), we locate our discussion of DP in this context. Yet, in doing so, we do not intend to suggest that health and wellbeing are the only or even primary topics that one might explore when using DP; as indeed there is a wellestablished, interdisciplinary body of scholarship that points to a range of relevant topics to organizations and HRD (e.g., Potter & Hepburn, 2010). Rather, we locate our discussion in relation to health and wellbeing given its relevance to the historical present, as well as our own scholarly histories.
To begin, we provide an overview of the various approaches to studying languagein-use, and outline DP and the different strands of the methodology. We also overview the theoretical foundation of the approach and point to what types of data are most often used. Prior to sharing two illustrative data extracts, we discuss how one might carry out a DP-informed analysis. These data extracts are not intended to illustrate health and wellbeing within organizational structures and settings; rather, we use our own DP work from child mental health clinics as way to translate the value of DP for organizational researchers. We conclude the paper by pointing to the ways in which DP might afford HRD scholars opportunities to generate theory grounded in everyday organizational practices-whether they are focused on examining health and wellbeing in organizations or other topics of interests.

Introducing Language-Based Approaches
There are a range of different methodological approaches that fall under the rubric of language-based discursive approaches, including discourse analysis, ethnomethodological methods, and conversation analysis, among others. DP is situated within this rubric and has been greatly influenced by many of these approaches, particularly ethnomethodological methods and conversation analysis. Many of these approaches hold great relevance and promise for HRD researchers. This is especially true of ethnomethodological studies and conversation analytic work, which has long considered 'talk at work' (Drew & Heritage, 1992). The interest in 'workplace conversations' is generally traced back to the publication of Order in Court-a text focused on the practice of courtroom interactions (Atkinson & Drew, 1979). Since then, scholars have used these language-based approaches to explore many areas of organizational business, including police work (Stokoe et al., 2018), therapy Silverman, 1996), courtroom interactions, (Atkinson & Drew, 1979), and call centers (Hepburn & Wiggins, 2005), among others. Despite the extensive exploration of institutional business across many different organizational settings, these developments have not been fully embraced by organizational, HRD or business scholars, and thus the value of using language-based approaches has not yet been fully realized within these disciplines (Llewellyn & Hindmarsh, 2011).
We argue that the benefits of using language-based approaches for HRD research is considerable and use this discussion as a platform for generating a methodological dialogue centred on DP within the HRD community. The language-based approaches have created a distinctive way to explore human behavior in organizational settings that is different from what has been typically offered in organizational studies, or in psychology and sociology (Potter & Hepburn, 2011). According to Potter and Hepburn, this distinctiveness derives from the analytic focus of looking at social interaction in the workplace, as well as the focus on the organizational members as part of the interactions and how they are conducting institutional business. It is therefore this attention to discourse or interactional practices that becomes a central tenet of investigation; that is, a core focus for the HRD researcher as they seek to explore the wider issue of health and wellbeing in organizational settings. Such foci for HRD researchers are central to improvements in the field to move away from the domination of performance orientation and production efficiency, to recognizing the role of discourse in corporate social responsibility and helping individuals transform their wellbeing and aspirations in the workplace (Sambrook, 2004).
To explore this further, it is helpful to understand where DP sits in relation to the wider umbrella of discourse analysis, as DP grew out of the discourse analytic family of approaches and has been heavily influenced by conversation analysis and its related practices. We therefore provide next a brief overview of discourse analysis, and then conversation analysis for context.

Discourse Analysis
Discourse analysis is one of the widely recognized qualitative approaches focusing on the study of language and language use (Wetherell, 2001). Discourse analysis is a broad term that represents related but distinctive approaches to studying language, and different ways of doing discourse analysis, represent variations in theoretical, methodological, and analytic assumptions. For example, there are discourse approaches that have a critical lens that takes a macro, socio-political stance, such as critical discourse analysis (van Dijk, 1993) and Foucauldian discourse analysis (Arribas-Ayloon & Walkerdine, 2008), with others taking a more micro-focus, such as DP (Edwards & Potter, 1992). In other words, DP does not presume social inequalities or power differentials, but instead explores how speakers interact and how they elect to make inequality, power, or asymmetry relevant in their social interaction . Thus, DP, like conversation analysis, explores a level of granularity that provides a way to attend to individual lexical items, intonation, overlapping talk, and features of embodied action that the speakers themselves attend to (Potter & Hepburn, 2011). This level of granularity is important for exploring institutional talk as this is exactly what is needed for a sophisticated grasp of institutional order. For example, in our own research of mental health assessments, we were able to illustrate the sequential processes of progressivity that accomplish the institutional task of assessing a child's proposed difficulty (O'Reilly et al., 2015). Founded on the principles laid out by Byrne and Long (1976) that the interactional and sequential order of institutional business provides information about the contextual features and processes of physician appointments, our work allows for a close analysis of how the institutional goals specific to mental health assessments were made visible in the talk between the clinicians and clients themselves. For instance, the institutionality of the talk was evidenced in the way that the clinicians posed questions and marked the interactions as being about eliciting an understanding of the child's condition.
Although there are differences across discourse analytic approaches, they are united in their focus on the study of talk and text as used in social practice (Potter, 1997). The binding feature, then, is that discourse analysis views language not only as descriptive and representative, but as performative. In other words, it is assumed that people do things with and through language. They invite, they complain, they assess, they question, and so forth. Consider this hypothetical example: Speaker One: Would you be able to come to that committee meeting on Friday? Speaker Two: Sorry I can't, I am attending a seminar event that afternoon. Speaker One: No problem, how about Tuesday? Speaker Two: Yes, okay, I will put it in my diary, please send the specifics on email.
Here we can see that speaker one is heard by speaker two as issuing an invitation. Importantly, the social action of the first speaker is evidenced in how it is treated by the second interlocutor, as it is through the interaction that the performative nature of the talk is made evident. Thus, in this case the language of the two speakers 'perform' organisational life, the issuing of an invitation, the rejection of that invitation complete with the account for the rejection, and the new invitation. In this example, we can also see the relevance of conversation analysis, and its focuses on turn taking. DP has been heavily influenced by the practices associated with conversation analysis, and many scholars using DP also draw upon the conversation analysis methodological literature to facilitate their analytic process. Having a reasonable understanding of conversation analysis is therefore important when conducting DP research-which we consider in abbreviated fashion next.

Conversation Analysis
Conversation analysis was pioneered by Harvey Sacks in the 1970s and is a methodology that focuses on the study of social interaction with particular attention being given to the details of 'talk-in-interaction' as well as the sequential order of talk (Sacks, 1992). The aim of conversation analysis is to provide an understanding of how social members of an interaction make sense of and respond to one another in talk, and to examine the ways in which people perform social actions with language in that context (Hutchby & Wooffitt, 2008). Indeed, it is through the design of conversational turns that the visibility of the 'happenings' of social life are revealed, and an individual's conversational turn sets up the normative expectation of what comes next in the talk (Sacks, 1992). Thus, for conversation analysts there is a focus on sequence organization, whereby in social interaction the talk should be responsive to the conversational turn immediately prior to it (Heritage, 2011), and with turn design, in terms of how a speaker constructs their turn within the interaction and how that can be understood as performing a social action (Drew, 2013).
Understanding conversation analysis as aligned with DP in terms of its relevance to HRD and organizational studies, it is helpful to appreciate that there are two foci for analysts. Much of the early work in conversation analysis studied mundane or ordinary conversations and later became more applied, leading to a distinction between what is commonly referred to as 'pure conversation analysis' (or traditional conversation analysis), and applied conversation analysis. Indeed, in his early pioneering work on conversation analysis, Sacks (1992) was working with institutional data from a Suicide Prevention Center. In his analytic dialogue, his early noticings focused on how callers spoke in ways that avoided revealing their identities to the call taker. His development of the approach then considered how talk-in-interaction is consequential for organizational work, while attending primarily to the sequential order of talk.
Notably, the distinction between these two foci is blurred due to some overlap between them, particularly as much mundane or everyday talk can occur in organizational settings . Nonetheless, institutional interactions have features that distinguish them from everyday talk. In addition, institutional and professional identities become relevant to the work-based activities with which people are engaged (Drew & Heritage, 1992) and this is made evident within interactional practices. Thus, applied conversation analysis has a particular focus on institutional talk and institutional business, and while there are different types of applied conversation analysis, much of this work has the goal of informing practice  and is particularly relevant to HRD scholars.

An Overview of Discursive Psychology
As noted at the outset of this paper, DP considers psychological matters from the perspectives of interlocutors in a social or institutional setting (Te Molder, 2015). In other words, DP is concerned with the way in which mental states are formulated and how psychological categories are utilized in practice (Potter & Hepburn, 2011). The foundational text, Discursive psychology (Edwards & Potter, 1992), introduced this approach as one that challenges traditional, cognitivist ways of making sense of psychological phenomena, and radically differs from the conventional notion that mental states are the cause of what is being said (Edwards, 1997). This means that instead of being interested in people's intentions, perspeptions, and motives per se, DP scholars explore how those mental states are treated as a practical concern for members of the interaction (Te Molder, 2015). DP has three main aims, and these were outlined by Wetherell (2007). First, the main focus of a DP study is the examination of psychological topics with a clear focus on language, which includes social categories, attribution, emotions, representations, memory, gender, and identity. So, instead of viewing professional identities, such as a 'manager', 'shop floor worker', 'chief operating officer', etc. as fixed role identities, DP analysts treat them as flexible constructs created by individuals and organizations and used to perform social actions within institutional settings. Second, DP seeks to promote attending to different ways of theorizing and studying psychological constructs as made evident through talk. Third, DP aims to advance qualitative methods not just in psychology, but across fields where language can be a central focus. It does this by moving away from focusing on what people think to what people do. In other words, DP is innovative in how it investigates research problems by being focused on the exploration of language as it occurs in organizational practices and by reconfiguring our understanding of mental states and cognition.
The Important Role of Theory in DP Scholars who use DP generally begin by looking at how versions of reality are constructed with an interest in how these constructions are positioned as truthful (Potter, 1996). The notion that mental states and psychological phenomena are constructed through social interaction reflects the broad social constructionism theory that underpins DP. Notably, social constructionism is an assembly of diverse approaches (Burr, 2015) and is not a single epistemological position, but a rubric of different theoretical foundations (Gubrium & Holstein, 2008). This theoretical position is one that advocates that human experiences are not fixed or predetermined but are mediated historically, culturally, and linguistically (Burr, 2015). So, from a DP perspective, what speakers say is treated as a version of events constructed in their narrative in ways that position their identity in a specific way . In alignment with social constructionism, Edwards and Potter (1992) argued that it is appropriate and necessary to examine discourse in its own right and not necessarily use discourse as a vehicle to access underlying mental processes. In this way, DP is conceived of as an approach and not just a method in the sense of being a series of analytic steps (Potter, 2012).

Strands of Discursive Psychology
There are different forms or strands of DP, which have been developed over time. Generally, it is understood that there are three main strands of DP, and, while they all share the core characteristics of the approach, there are some distinctions between them (Potter, 2012).
The first strand of DP was the earliest form that grew from the work of Potter and Wetherell (1987) in the field of social psychology. Within this strand of DP, there was a focus on identifying interpretive repertoires; that is, the building blocks that speakers utilize when they construct a version of the world which are often signaled by figures of speech or tropes (Wetherell & Potter, 1988). Research employing this strand of DP tends to take a macro-social constructionist position, as it takes up a more critical perspective by attending to larger sociopolitical concerns (see, Reynolds, 2013, for an example). For analysts using this strand of DP, there is a broader interest in the study of cultural identity and power, and here DP can offer useful theoretical and methodological insights.
The second strand of DP grew popular in the 1990s and reflected a shift toward studying everyday conversations and natural institutional settings. It was this strand of DP that focused researchers on psychological matters as "…objects in and for interactions" (Potter, 2005, p. 789). With a stronger orientation to micro-social constructionism, this strand of DP aims to respecify psychological constructs, such as cognition and memory, as produced in and through interaction (Edwards & Potter, 1992). This orientation stands in contrast to viewing such constructs as static and existing apart from the very discourses that make them possible. For example, drawing upon DP, Lester and Gabriel (2014) illustrated in their analysis of educational psychology textbooks how 'intelligence' is a discursive entity; that is, a construct that can be respecified as produced in and through discourse. Notably, it was with the development of this second strand that DP became recognized as a distinctive form of discourse analysis.
The third strand of DP developed from the mid-1990s onwards as DP became more strongly influenced by conversation analysis and ethnomethodology. It was as DP evolved that the leading DP scholars became increasingly interested in the sequential organization of talk and attending closely to the detail of talk (see, Edwards & Potter, 2017, for an example). It was this engagement with conversation analysis that shifted the way researchers using DP collected data and emphasized attending more specifically to the micro-features of interaction (e.g., pauses, embodied conduct, prosody, etc.) (Potter, 2012).

Research Questions DP Can Address for HRD Scholars
Within DP, it is not common to develop a research question at the start of a project, as there is a broader interest in studying how social actions are achieved in and through social interaction (Potter, 2012). Indeed, this stands in contrast to hypotheticdeductivism wherein a researcher's study is guided by a specific hypothesis (Potter, 1996), or even the common suggestion that a research question should (or even must) guide the entirety of one's study. Rather, DP values a data-driven approach where a researcher develops a research question in relation to the collected data, noting that it is difficult to know what might be a relevant question until one collects the data . This orientation to the development of research questions does not mean that a DP study lacks focus or purpose; rather, research questions that are developed prior to the start of a study are oriented to as guiding rather than prescriptive or unchangeable. It has been suggested that guiding questions serve to allow for a more inductive process -one that results in research questions being closely connected to the social actions that are relevant within a given context.
Broadly, DP researchers seek to examine what takes place (or is achieved) within certain kinds of social interactions. Common to all DP study is an emphasis on taking a bottom-up, inductive approach, wherein the data drives the primary focus. This may result in ideas and insights arising that were not understood as relevant prior to beginning the study. As such, with an emphasis on maintaining an open, exploratory stance, researchers using DP often develop very open, initial research questions. For example, they may ask: • How do managers interpret and construct bullying at work?
• What is the nature of stereotyping when it comes to disclosing mental health concerns in the workplace?
• How does the language used in return-to-work interviews shape employee wellbeing?
While indeed an overarching research question may shape the direction of the project, it is common that gaining familiarity with the data will result in the research question(s) being further specified and clarified.

Types of Data Used in DP Research
DP researchers are interested in what is actually happening within a given context, rather than people's reports of their actions, perspectives, or experiences. They are generally influenced by Sacks's (1992) claim that: If we are to understand and analyze participants' own concepts and accounts, then we have to find and analyze them not in response to our research questions, but in the places they ordinarily and functionally occur…in the activities in which they're employed. (p. 27) Thus, within DP research, it is common for naturally occurring data to be collected (Kiyimba et al., 2019). While defining what counts or not as naturally occurring data is debatable, it is generally understood to be data that would exist even if it was not collected for research purposes (Potter, 2002). For instance, if a DP researcher was interested in studying how employee wellbeing is talked about and positioned in weekly staff meetings, they would collect data of the actual meetings rather than asking people to talk about what happened in the meeting or how they think the organization views wellbeing. The meeting, of course, would take place regardless of whether the researcher recorded it-making it naturally occurring data. Importantly, while there has been a growing preference to collect naturally occurring data (Potter & Hepburn, 2005), some DP studies do indeed include the collection of interviews and others forms of researcher-generated data (e.g., Lester & Paulus, 2012;Wetherell & Edley, 2014). As TEN Have (2002) argued, the degree to which data might be characterized as natural is "always relative to the analytic context in which it functions as data, so it is never an inherent property of the materials" TEN Have, 2002 (p. 529). Thus, arguably, data can never be totally, and absolutely natural, as the positionality of the researcher, the use of recording devices, and the analytic process will always subtly influence and shape the process, and thus the notion of 'naturally occurring' rather than 'natural' is favored to reflect that.
There are underlying reasons that DP researchers give preference to naturally occurring data. Perhaps the most notable is DP's post-cognitive position (Potter, 2002); that is, a position that points to the limitations of taking up mentalist perspectives of knowing and being. Rather, it is assumed that social science researchers can access how people do knowing and being by analyzing their talk in the moment, rather than asking them to report on how they think. In this way, many DP researchers argue that researcher-generated methods of data collection, such as interviews or focus groups, are not a mechanism for gaining access to what people 'really' think or feel (Potter & Hepburn, 2005), as it is not possible to really access what is inside someone's mind. Feelings, attitudes, memories, beliefs, etc., are instead not viewed as static entities but rather rhetorically and interactionally accomplished in and through talk (Edwards & Potter, 1992). Additionally, it has been argued that one of the key benefits of collecting naturally occurring data is that the analysis of conversations as they are unfolding allows for a researcher to work against decontextualizing their data (Sacks, 1987). In other words, the analyst remains close to the 'real' and naturally occurring conversation as the interlocutors within it contextualize and construct meaning, exploring the uptake and understanding conveyed by those in that conversation. Thus, rather than only the analyst's interpretation of the talk being foregrounded, the understanding of each turn is shown through the second speaker's response to that turn and retaining attention to the context of the talk as it occurs naturally in situ. Given this, it is most common for DP researchers using to plan to collect audio or video-recordings of social interactions as they occur within the context of their particular interest.

Process of Carrying Out a DP Study
Like other approaches to qualitative data analysis, a DP analysis is conceptualized as inductive and iterative. This kind of analysis requires that an analyst re-engage with the data in a layered and cumulative way, building toward a cohesive understanding of the primary patterns. While DP does not involve following a stepwise linear process, Potter (2012) described seven overlapping stages for carrying out a DP study which are intrinsically tied to the markers of quality for the approach. These stages involve data collection, data management, analyzing the data, and establishing validity.
First, DP researchers begin by obtaining data access and participant consent. Having access to quality, naturally occurring data can be quite challenging, depending on the context in which a researcher hopes to carry out their study. Acquiring audio or video recorded data, nonetheless, is essential to setting the stage for a quality analysis.
Second, when planning for and collecting data, a DP researcher must carefully consider their sampling practice and sample size. When determining the appropriate amount of data to collect, it is essential to recognize that the sample size for this kind of research does not equate to the number of people or participants. Rather, the sample size is the conversation (or discourse) itselfas this is the object of study. Thus, building a large interactional data set (e.g., 40 hours of talk) focused on a particular phenomenon of interest can be quite useful; however, this is not always necessary given even small amounts of interactional data (e.g., 1 hour of talk) can offer important and robust insights. In published DP literature, there are a range of sample sizes, with some including six hours (e.g., , and others including well over 30 hours (O'Reilly et al., 2020) or even more than 100 hours (Lester, 2014). There are no hard and fast rules about the size of the dataset, but rather a researcher must consider what they hope to make sense of and how the size of the dataset can support them in this process. In addition, the kind of data collected might be Table 1. Jeffersonian Symbols (Jefferson, 2004).

Symbol Meaning
(.) A period inside parentheses represents a pause that is hearable but not measurable (otherwise referred to as a micropause) (0.2) A number inside parentheses shows the length of a given pause [ ] Square brackets represent speech that is overlapping > < Text encased with 'greater than' and 'less than' symbols represents speech that is hearable as being faster than the surrounding speech < > Text encased with 'less than' and 'greater than' symbols represents speech that is hearable as slower than the surrounding speech ( ) A space between parentheses represents spoken words that are not clear and not possible to transcribe ((Raises their left hand))

Double parentheses with a description included within it offers contextual information Under
An underlined work or a portion of a word represents a rise in the volume or emphasis ↑ A rise in intonation is represented with an upward arrow ↓ A drop in intonation is represented with a downward arrow → A side arrows represent a sentence of interest that is of interest to the analyst CAPITALS When something is said loudly or shouted, capital letters are used Hum(h)our A bracketed 'h' represented laughter in talk = Equal signs represent speech that is latched ::: Colons represented speech that is elongated or stretched conversational (e.g., Lester et al., in press), text-based (Lester & Paulus, 2011), and/or include images (McCullough & Lester, 2021). Third, once the data are collected, the researcher might consider how they will manage the data. This often entails securely storing recordings, transcripts, images, analytic notes, etc. For many researchers in the contemporary era, the storage process is often systematized with the use of a qualitative data analysis package (Paulus & Lester, 2022).
Fourth, for many DP analysts, after data are collected, close analysis begins with directly engaging with audio or video-files, potentially using a qualitative data analysis software package to support direct labeling and memoing of the recorded files. This first level of engagement allows for a researcher to familiarize themselves with the data, often resulting in maintaining detailed notes that center around general questions about the language-in-use, such as: 1) what is being accomplished in and through the language use? and 2) how is the language structured to make this possible? Some analysts move next to the transcription stage, which in DP is understood to be an interpretive and situated process (Ochs, 1979). Generally, detailed transcripts include not just what is said but also how, as illustrated in the extracts we present below and in Table 1.
Because DP scholars give attention to the micro-details of language-in-use, and assume that these details are doing something, the transcription system aims to capture such details. This required detail is present in the Jefferson system of transcription developed for use in conversation analysis, and DP research typically borrows this system (Jefferson, 2004). It is easy to think of transcribing data as a mundane or simple technical process; and yet it is neither straightforward, nor neutral . Discursive researchers acknowledge that transcription is part of the analytic process and thus recognize that decisions they make about transcription are tied to the practical and epistemological concerns of the methodology (Hepburn & Bolden, 2017). The aim then of the Jefferson transcription system is to represent as much detail of the vocal, verbal, and multimodal aspects of the interaction as possible (Jefferson, 2004). This is critical, as transcripts are an important source of evidence for the reader to assess the quality of the analytic claims (Roberts & Robinson, 2004). In addition, if there is an interest in embodied actions (e.g., heads nodding in agreement), a multi-layered embodied transcript may be produced. Mondada's transcription system, for instance, might be layered on to a Jefferson transcript to represent embodied actions (see, Mondada, 2007, for an example). Some DP scholars have highlighted the value of producing transcripts in a synchronized way using qualitative data analysis software; that, is linking the audio or video recording to the transcript via timestamps to allow for an analyst to both listen/view their data and read the transcript (Paulus & Lester, 2022).
Fifth, throughout the research process, a DP researcher is continually re-evaluating and re-developing their research questions. As we noted above, research questions in a DP study are emergent, meaning they are continually refined as researchers comes to know the dataset.
Sixth, while woven throughout the stages we have already described, the analysis of the data is prioritized in a DP study. Within DP, analysis involves a systematic examination of the entire dataset to build a corpus of examples that can be analyzed carefully. The examples generally relate to the identified phenomenon of focus (e.g., how clients build a case that they are a supportive partner). DP analyses involve examining the entire dataset with the goal of building a corpus of data extracts around an identified phenomenon of interest (e.g., how do organizational leaders position disabled employees in the company-wide meetings). While particular to conversation analysis, we have found Drew's (2015) eight steps of analysis to be applicable and useful for conceptualizing a DP analysis. They include: examining the interaction for social activity; attending to turn-taking sequences; examining the details of the interaction (e.g., pronoun usage); exploring how other speakers (or recipients) respond; identifying language features that are shared; collecting cases of the phenomenon to build a corpus; accounting for the key conversational pattern; and writing up the findings in a detailed manner .
Seventh, establishing the validity or quality of a DP study is positioned as central. Some critics of discursive research may "confuse a focus on natural language data with some kind of interpretative free-for-all" (Edwards, 2012, p. 431). Contrary to this presumption, researchers have articulated a set of quality criteria to consider when evaluating a DP study. More particularly, there are two primary publications that are commonly cited when evaluating the quality of a DP study. The first publication was written by Antaki et al. (2003) and proposed six indicators for evaluating the quality of discourse research, including DP research. Indicator one, 'under-analysis through summary', speaks to the idea of simply summarizing themes that capture what participants talked about in an interaction without analyzing how the discourse was used. Indicator two, "under-analysis through taking sides', points to the idea of a researcher taking up a position that aligns or disaligns with a given speaker. Indicator three, 'under-analysis through over-quotation or isolated quotation,' occurs when an analyst fails to offer an interpretation of the data; rather, the analysis simply provides quotations of the data and allows it to stand alone. Indicator four is referred to as 'circular discovery of 1) discourses and 2) mental constructs' and highlights the idea of orienting to utterances as being reflective of inner thoughts or feelings, rather than recognizing that mental constructs are achieved in and through interactional practices. Indicator five, 'under-analysis through false survey', occurs when an analyst orients to their findings as being reflective of all people within a given category. Indicator six, 'under analysis through spotting,' occurs when an analyst simply spots conversational features (e.g., pronouns) rather than attending to what a given feature accomplishes or achieves in the interaction. Building upon these ideas,  more recent publication offered a contemporary discussion of quality in DP and argued that "quality is embedded in the practices of doing DP" (p. 410). They provided a comprehensive framework that outlined three key domains for assuring quality: 1) quality in data collection and management, 2) quality in the process of analysis, and 3) quality as related to validity and reliability. Within each domain, quality markers are outlined and include considerations related to the nature of the data, sample size, and the process by which analysis is conducted.

Case Example: The Role of Question Design in Institutional Talk
There are many ways in which DP research can contribute to the field of HRD. For example, DP can help illuminate how health and wellbeing are managed within the workplace, how organizations contribute to a disabling environment, and how health is managed when multiple organizations are involved. Indeed, the very way that organizations and their members talk about health and wellbeing offers a pathway forward for 1) talking differently, and 2) enacting structural changes based on understanding current interactional practice and their affects.
To provide a sense of the value of DP for HRD practice, we turn our attention to one of the fundamental aspects of institutional business; that is, the design of questions. Organizational work revolves around question-and-answer sequences, whether in a meeting, a recruitment interview, a misconduct hearing, a job appraisal, or anything else that occurs within an institutional framing. Questions-and-answers are a fundamental concern for conversation analysis as a standard adjacency pair (Sacks, 1992), and aligned with this, DP has an interest in the social functions that these pairs serve within different contexts. Further, to better understand organizational discourses, it has been argued that one must appreciate question-and-answer sequences (Ehrlich & Freed, 2010), because of their foundational work in such settings (James et al., 2010). The way a question is designed will shape and guide the way it is answered because of the preference structure that is inherent to the design. An invitation question, for example, is valanced toward a positive response, and a negative response warrants an account. Some question designs are treated as requiring an account and thus pose a challenge to the recipient in their framing. For instance, in our own work in mental health settings with children, we demonstrated that 'why' prefaced questions were treated as accountable and rarely elicited much in response, but when they did it was an explanation for their behavior or that of others (Kiyimba et al., 2017). See, for example, a data extract below and Table 1 where the Jeffersonian symbols are included (taken from Kiyimba et al., 2017, p. 235).
Here in the context of a mental health assessment, the Community Psychiatric Nurse (CPN) seeks an explanation of the disclosure from the adolescent client (Adol). The young person discloses an intrusive thought that he feared his girlfriend would 'come in and kill my family', and the wh-prefaced question from the CPN asks not why he thinks that, but rather why she would do that. In so doing, the CPN makes the girlfriend accountable for that potential action and seeks the reasons for her possible motivations for the presented behavior.
What is interesting for a HRD researcher here is the function of the why-prefaced question design, as these wh-questions are regularly used in workplace settings. Kiyimba et al. (2017) noted that what was especially notable about this why question was that it sought the motivations of another person and not the interlocutor in the interaction and this is more problematic and difficult to respond to. In the workplace, this also holds, as one work colleague being asked to comment on why another work colleague did or did not do something, or what motivated them to behave in a particular way, is in the domain of the other person and not the current speaker. Thus, any attempt to provide an account is arguably just speculation. Likewise, a similar challenge can be seen in other types of wh-prefaced circular questions whereby a speaker is asked to provide an account for the motivations of another.
We provide another example below taken from Lester et al. (in press) from the same data corpus. Here the child fails to provide a reason why his parents have bought him to the appointment claiming not to know why they have done so.
In this example, the child is being asked to speak to the motivations or reasons of another; that is, the parents, and the question is unsuccessful in the sense that the child makes a claim to insufficient knowledge to provide an answer. From an organizational perspective, when questions are asked of other persons in any institutional business, for a response to be forthcoming, the question needs to be in the recipient's epistemic domain to accomplish a full and elaborate answer; when it is not, there are risks that the question may fail. Thus, while the questioner may position the recipient as having the capacity or knowledge to provide an answer, when the nature of that is not within the domain of the recipient, it may result in a lack of response. Consider, for example, a line manager asking one of the administers why the senior sales executive is wearing red shoes to a meeting. Asking for the motivation of someone else's fashion choice is inherently difficult to answer unless that information has been directly disclosed. Here, in seeking the parental account for the visit from the child rather than the child's account, the knowledge is not in the child's domain; rather, it is knowledge known by the parents, and an elaborated answer is not provided showing that this information has not been disclosed to the child. This is the kind of understanding that DP research can reveal, allowing researchers to unpack how these kinds of questions, for instance, operate across settings. Indeed, this child mental health example is just one illustration of this social practice.

Conclusions: Discursive Psychology's Relevance to Human Resource Development
While DP is currently underutilized for organizational and HRD research, we suggest that it is a beneficial and valuable methodological approach for several reasons. First, although there are exceptions, organizational studies and HRD focused research has tended to rely on empirical materials that are distanced from the 'real-time' flow or 'live' conduct within a given institutional setting (Llewellyn & Hindmarsh, 2011). Yet, within organizations, members of the interaction are building their own worlds while simultaneously acting within them, constituting and re-working the institutional procedures, and displaying professional role identities (Potter & Hepburn, 2011). The institutionality of the talk is not merely determined by the setting alone; instead, it is related to how the members orient to that institutional setting or professional identities as relevant to the work activities they are engaged in (Drew & Heritage, 1992). DP is a particularly valuable methodology for researchers who seek to explore how these organizational practices unfold in situ; that is, for those who desire to examine, for instance, how health and wellbeing come to be in real-time. Within organizations this may look like analyzing leader and follower exchanges (Clifton, 2019), examining organizational bullying (Johnson, 2015), or considering the informal exchanges between colleagues (Mengis & Eppler, 2008). Given DP's focus on naturally occurring data, a researcher can study the actual practice thats shapes an organization and those that work within it. Rather, than asking people to report on their experiences, perceptions, etc., a DP study allows for a close analysis of how the day-to-day life of an organization unfolds. Organizations have a wealth of this naturally occurring data to be explored from analysis of talk in formal meetings to consideration of the language used in policies and organizational branding to the informal conversations that happen in the canteen. All this talk a shapes the culture and practice of the organization and currently little is known about exactly how this is happening.
Second, organizations would not exist without communication; the language-inuse, for instance, used within a team, between a manager and subordinate, in marketing and recruitment materials, etc., is all central to how organizational practices come to be. Yet, if we do not study these interactional practices, we will never fully understand how relationships and everyday social practices in organizations unfold in situ and shape the organization locally (at an individual) and broader level. This qualitative approach offers HRD scholars an opportunity to take up an innovative orientation to understanding the key mechanisms at play. Indeed, this paper is designed in some ways as an invitation to consider what this different approach to the study of organizational practices may reveal and how this new perspective allows for the generation of new theories that point to what is actually happening on the ground; that is, within the everyday and institutional interactions. This kind of understanding may very well also lead to understanding challenges to promoting health and wellbeing within an organization in nuanced ways, potentially leading to some novel solutions.
Finally, DP can help bridge the gap between science and practice and provide a translational opportunity for HRD scholars to identify recommendations for improving organizational cultures. The micro-attention to language and social actions can support researchers in exploring when and how various institutional tasks work effectively and where there is disruption to the flow of interaction. Indeed, DP research can spotlight specific interactional practices and subsequently be used as a training tool to highlight best practices (Stokoe, 2014); that is, practices that promote the health and wellbeing of an organization and its members. This potential to facilitate change arises from DP's emphasis on not just what members of an institution do, but most importantly, how their social practices unfold.
Note 1. Language-in-use refers to how language use is meaningful within sequences of interactions with others and through its use is performative .

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.