Secondary analysis is a research methodology in which preexisting data are used to investigate new questions or to verify the findings of previous work. It can be applied to both quantitative and qualitative data but is more established in relation to the former. Interest in the secondary analysis of qualitative data has grown since the mid-1990s, although the idea of reusing qualitative data has been met with a mixed reception by the social research community. This entry describes how the secondary analysis of qualitative data has been variously defined, promoted, practiced, and debated in the United Kingdom, Europe, North America, and Australia. It begins with an overview of the methodology, and how it differs from documentary analysis and other qualitative approaches. This is followed by sections on how qualitative data have been reused in practice, different sources of data, types of qualitative secondary analysis, and examples of previous studies from the author’s own work are described. In the remaining sections, the main epistemological, ethical, and methodological issues discussed in the debates about the methodology are explained. The entry concludes with a summary of the key perspectives and identifies some of the opportunities and challenges ahead in this field.
Both quantitative and qualitative data can be subjected to secondary analysis, although there are some important differences in the ways in which the methodology has been conceptualized and utilized in relation to these different types of data. In addition, whereas the secondary analysis of quantitative data is a widely accepted research practice, there has been a great deal of debate about the whys and wherefores of reusing qualitative data, which is examined in this entry.
In contrast to quantitative secondary analysis, which encompasses the reuse of numeric data from research projects and other sources (e.g., hospital audits, insurance databases, census records), qualitative secondary analysis has been defined more narrowly by Janet Heaton (2004), the author of this entry, as referring to the reuse of data from primary research studies (e.g., transcripts of interviews and focus groups, responses to open-ended questions in questionnaires, field notes, research diaries). This distinguishes it from documentary analysis (Plummer, 1983), which involves the examination of other types of qualitative materials and cultural artifacts that were created in the course of everyday life and “found” by researchers (e.g., newspaper entries, personal diaries, photographs), as opposed to “data” that were created for research in the first place.
Qualitative secondary analysis is also distinct from other specialist methodologies, such as conversation analysis and social media analysis, which again refer to particular types of naturalistic qualitative data that were generated in everyday routine interactions and not made for a research project. The methodology is also not to be confused with systematic reviews, syntheses, and meta-analyses of qualitative research, which generally involve going back over the reported findings of previous studies but not back to the original data sets, to review the state of knowledge on a given topic.
These methodological boundaries are not, however, always quite so clear-cut. For example, photographs may also be taken and used as primary data in a research project, transcripts of real-life conversations may be exchanged and reused by conversation analysts, and longitudinal data may be collected in research for both bespoke primary purposes and as a general resource for future studies of change. In these cases, the reuse of such materials may be broadly conceptualized as secondary projects, involving documentary analysis or conversation analysis or longitudinal analysis as the particular analytical approach. More generally, many of the analytical techniques and methods used in qualitative secondary analysis (e.g., thematic analysis) are the same as those used in primary studies, although some may be modified in this context (e.g., sampling strategies), and some may be unique to it (e.g., pooling data from existing multiple data sets). The boundaries between qualitative secondary analysis and these approaches are also liable to change as technical innovations facilitate new ways of working. For example, as qualitative data are increasingly appended to online publications in supplementary files, the possibilities for meta-analyses based on these available data (to synthesize the evidence) and for secondary analyses based on samples of these data (to investigate new questions or verify previous work) are becoming real, blurring the lines between these approaches.
Whether it is appropriate to compare the secondary analysis of quantitative and qualitative data has been queried by Niamh Moore (2007). She has argued that such comparisons have been used as a strategy to support the development of qualitative secondary analysis, when comparisons with other qualitative approaches might be more relevant. Similarly, whether it is appropriate to describe qualitative data as “preexisting” in secondary analysis and to invoke the use/reuse distinction, has been debated. For example, Moore (2007) argues that qualitative data are created and coproduced in (primary and secondary) research practice. She suggests that the reuse of qualitative data is best understood as a process of “recontextualising” and “reconstructing” these data (para. 2.3). However, while Martyn Hammersley (2010) accepts that this is the case and that the primary/secondary and use/reuse distinctions are problematic, he still believes that these terms are useful.
Thus, while the term preexisting is widely used to differentiate secondary from primary qualitative data, this does not necessarily mean that the contexts in which these data were collected and analyzed are not attended to in practice. As Heaton (2004) has argued, data can be collected and explicitly (re)interpreted in multiple (temporal) contexts, although the extent to which this is done varies. Indeed, in her examination of how qualitative data have been reused in practice, she suggested that secondary analysis could be used even more reflexively and creatively than it has been to explore:
the complex relationships between the analyst and the data—the contexts of data interpretation—as well as using the data to explore the perspectives of others—the interpretation of data in context. […] Thus, in post-modern qualitative research, primary researchers would collect and record stories from informants which they or archivists would then preserve for use in possible secondary studies. In using and re-using these data, primary and secondary researchers would work within, and move between, the various contexts in which the stories were told, recorded, rendered for analysis, interpreted and re-interpreted. A series of stories, each a bricolage, based on these primary and secondary data, would be produced and added to the resources available for qualitative researchers to work with. (Heaton, 2004, pp. 123, 124; original emphasis)
This entry explains in more detail the ways in which the secondary analysis of qualitative data has been defined, promoted, practiced, and debated by the international research community. The entry begins with an overview of the sources of qualitative data that have been used in practice, especially in health-related research where qualitative secondary analysis has been particularly embraced. Then, different types of qualitative secondary analysis are outlined. This is followed by a detailed description of three examples of secondary studies from Heaton’s research on health-related topics. These studies are used to illustrate the subsequent examination of the main epistemological, ethical, and methodological issues associated with the practice of reusing qualitative data. The entry concludes with a summary of the key perspectives and some reflections on the opportunities and challenges ahead in this field.
Sources and Usage of Qualitative Research Data
Interest in the possibilities of secondary analysis of qualitative data has grown since the mid-1990s, supported by various developments in research infrastructure, policy, and practice. An increasing number of qualitative secondary studies have been published in North America, Europe, and Australia, drawing on data from three key sources: data archives, informally shared data, and self-collected data.
Many countries now have national data archives and other institutional repositories that hold data sets from the social sciences and humanities. Previously, these tended to focus on the collection of statistical data sets. Where archives were interested in qualitative material, they focused mainly on oral histories and cultural artifacts (e.g., The Institut für Geschichte und Biographie in Germany and the Mass Observation Archive in the United Kingdom), or they accepted qualitative material that was coded for statistical analysis (e.g., the Human Relations Area Files project in the United States). However, an increasing number of archives now accept qualitative data sets from research studies and make them available for secondary analysis.
In Europe, the Consortium of European Social Sciences Data Archives works with associated data services and repositories to promote access to social science research data. Several archives in the United Kingdom and Europe hold qualitative as well as quantitative data sets including the UK Data Archive (UKDA), the Finnish Social Science Data Archive (FSD), the Danish Data Archive, the Czech Social Science Data Archive, the Norwegian Centre for Research Data, the Swedish National Data Service, the Irish Social Science Data Archive, and the beQuali Archive, part of the DIME-SHS program of work coordinated by Sciences Po in France. Elsewhere, in the United States, the Inter-University Consortium for Political and Social Research at the University of Michigan promotes data archiving and its collections include some qualitative data sets, the Henry A. Murray Research Archive in the Institute for Quantitative Social Science at Harvard University also holds longitudinal data sets which contain some qualitative data, and the Qualitative Data Repository at Syracuse University holds data from qualitative and mixed-methods research projects. In Australia, the Australian Data Archive at the Australian National University is made up of several subarchives, including a qualitative section.
Advances in qualitative data archiving have been led by the United Kingdom, where formal sharing of all types of qualitative data across the social sciences has been promoted by a major funder of social research, the Economic and Social Research Council (ESRC). The ESRC established the world’s first specialist facility—the Qualitative Data Archiving Resource Centre (Qualidata)—at the University of Essex in 1994. The center was set up to stimulate and facilitate the archiving of qualitative data sets in existing repositories across the United Kingdom and has since been incorporated within the wider UKDA. Its work has helped to increase the number of qualitative data sets available for secondary analysis and also their accessibility by cataloguing them and by digitizing a selection. Further advances in computing are likely to improve opportunities for preserving qualitative data and to facilitate easier access to data sets online in the future.
Qualitative data archiving and reuse has also been promoted in the United Kingdom through a number of related ESRC policy and funding initiatives. For example, since 1995, ESRC policy has required as a condition of its awards that researchers make available qualitative data sets arising from their work either through the UK Data Service or an appropriate digital repository. Also, when applying for funding from the ESRC, researchers have to demonstrate that the proposed primary research could not be carried out using any existing archived data set. The ESRC has also funded major programs of work and large projects to develop resources and capacity for qualitative secondary analysis. These include a Qualitative Archiving and Data Sharing Scheme, which comprised five projects (2005–2006); the Changing Lives and Times: Relationships and Identities Through the Lifecourse project (2007–2011), which involved the creation of a specialist archive for qualitative longitudinal studies (“Timescapes”) and work to develop ways of analyzing these data; and a Secondary Data Analysis Initiative program, which supported research based on archived quantitative and qualitative data and sought to develop capacity for working with these data (2012–2019).
At the same time, other institutional archives have emerged in the United Kingdom and elsewhere, linked to particular research programs. For example, since 2000, qualitative data from a series of national primary studies carried out by the Health Experiences Research Group (HERG) in the Nuffield Department of Primary Care Health Sciences at the University of Oxford have been retained by them, with participants’ permission, for further use in research and teaching. In these studies, people’s experiences of particular medical conditions and health care in the United Kingdom are captured in semistructured interviews that are digitally recorded and transcribed. Findings from the primary analysis of these data are disseminated via a website (www.healthtalk.org), so that other patients and their families, as well as the public and health-care professionals, can learn from the participants’ first-hand accounts. The findings are presented in the form of thematic summaries of key topics, supported by video, audio, and/or written extracts from the interviews. Copies of the anonymized interview transcripts are also available in full for use in secondary studies by approved users, and subject to a fee and a license agreement.
Over the years, these developments have led to an increase in the number of qualitative data sets archived in the United Kingdom and other countries. A review of the data sets held by some of the data services across Europe found that the UK Data Service had nearly 1,000 qualitative and mixed-methods collections accessible via its catalog, and the FSD had 177 qualitative data sets; other archives had fewer collections including some with under 10 data sets (Bishop & Kuula-Luumi, 2017). The HERG Archive at the University of Oxford has been reported to contain over 75 collections on different conditions and health-related topics (Ziebland & Hunt, 2014).
While official policies in the United Kingdom have promoted the archiving and formal sharing of qualitative data sets, internationally a number of researchers have carried out qualitative secondary studies using data from two other sources. One is through informal data sharing, whereby analysts obtain their data directly from other researchers. The primary researchers who collected the data may or may not be involved in the secondary analysis as a collaborator or in an advisory capacity. Single or multiple data sets may be shared and reused in full or in part, depending on the aims and scope of the secondary research. The other source is analysts’ own data collections. In secondary studies based on these data, sole researchers or whole teams may reuse data from their previous research to investigate new questions or to follow up emergent themes from the primary work.
Little is known about the extent to which these three sources of data are being reused in practice. An early review of qualitative secondary studies published in the international health and social care literature, carried out when the archiving of qualitative data sets was in its infancy, identified 65 studies that involved the reuse of existing data sets (Heaton, 2004). The majority were by researchers based in the United States and Canada (n = 51), while the remainder were from the United Kingdom (n = 12), Sweden (n = 1), and pan-Canada/Sweden (n = 1). Of the 65 studies, 9 were by researchers who had no involvement in the primary studies, of which only two were based on data that had been formally archived. The majority of studies (n = 36) were based on data that the researchers had collected themselves. The other studies (n = 20) were carried out by teams using a mix of data assembled from their own work and other sources.
As investment in the archiving of qualitative data sets has grown and the number of collections available in archives in the United Kingdom and elsewhere has increased, so too have calls for more information on how much this costs—and on how many, and which, of these data sets have been reused, by whom, and with what outputs (Parry & Mauthner, 2005). In the aforementioned review of data sets held by archives across Europe, some information on the usage of the data sets in two European archives was provided, along with a note of the difficulties with the metrics used for this purpose (Bishop & Kuula-Luumi, 2017). It showed that, between 2002 and 2016, the UK Data Service had information on the usage of 444 qualitative and mixed-methods data sets. Of these, 177 (40%) had not been downloaded and the remaining 267 (60%) had been downloaded at least once. The latter included 108 collections that had been downloaded 1–10 times, 60 that had been downloaded 11–20 times, and 15 that had been downloaded more than 100 times. Most of the downloads were by postgraduate students (42%); the rest were by staff in higher education institutions (27%), undergraduates (25%), and other users (6%). The data were being downloaded for the purposes of research (15%), teaching (13%), and learning (64%), the last including students’ assignments and users’ exploratory work. The associated analysis of usage of the qualitative data sets held by the Finnish Data Service found that there were 323 downloads of qualitative data sets in 2015. These were downloaded for undergraduate study (41%), master’s work (28%), teaching (20%), and research (11%).
Some examples of qualitative secondary studies based on the reuse of archived data were also described in the review, together with observations on the difficulties of identifying such studies through literature searches (Bishop & Kuula-Luumi, 2017). As Heaton (2004) noted, even after obtaining and reading candidate publications in full, it can be difficult to decipher the authors’ relationship to the data and other parameters of the work because of incomplete reporting of qualitative secondary studies. Notwithstanding these limitations, in the next section, different types of qualitative secondary analysis are described along with some examples of studies that have been identified from Heaton’s periodic searches of the literature.
Types of Qualitative Secondary Analysis
In her seminal work elaborating the possibilities of qualitative secondary analysis, Sally Thorne (1994, 1998) envisaged different ways in which the methodology might be used. She suggested that, through a process of “analytical expansion,” researchers could make further use of their own data to address “new or extended” questions (p. 548). Through “retrospective interpretation,” they could examine new questions that were raised but not addressed in the context of the primary study. Through “armchair induction,” they could use inductive methods of textual analysis to develop theories. Through “amplified sampling,” they could compare multiple distinct and theoretically representative data sets. And through “cross-validation,” they could “confirm or discount new findings and suggest patterns beyond the scope of the sample in which the researcher personally has been immersed” (p. 267).
A subsequent review of the health and social care literature showed that, just as Thorne anticipated, qualitative data were being reused in this field in various ways, some of which were similar to what she had envisaged (Heaton, 2004). Five types of qualitative secondary analysis were identified based on this review of emerging practice and are described in the following subsections. The first three types vary according to the degree to which the aims of the primary and secondary work converge or diverge; the last two types further distinguish studies according to the combinations of data that are utilized.
In this type of secondary analysis, the focus of the secondary study transcends that of the primary work. New empirical, methodological, or theoretical questions are explored that are distinct from the aims of the original research. For example, some analysts have examined the use of metaphors in participants’ accounts of medical encounters using existing data (e.g., Jairath, 1999; Jenny & Logan, 1996). Others have reused data to compare different methods of textual analysis (Atkinson, 1992) and to examine the value of different approaches to biographical analysis (Jones & Rupp, 2000).
This approach involves the in-depth investigation of an issue, or aspect of the data, that was not addressed, or was only partly covered, in the original research. The focus may be on a particular issue or theme that emerged from the primary work and/or a subset of the data. Unlike supra analysis, the subject of this type of secondary analysis is more closely related to that of the primary work. As a result, in some cases, it may be difficult to distinguish where primary research stops and secondary analysis starts, particularly when the supplementary analysis is carried out by the same researchers who carried out the primary work. This type of secondary analysis was found to be the most common in Heaton’s (2004) review, and it has since been adopted in several studies (e.g., Benbow, Forchuk, & Ray, 2011; Greenhalgh, Stones, & Swinglehurst, 2014; Jowsey, Pearce-Brown, Douglas, & Yen, 2014).
Whereas the first two types of secondary analysis involve the investigation of new questions or emergent issues, the purpose of reanalysis is to verify and corroborate the findings of previous work. Only one example approximating this type of secondary analysis was identified in the review. In it, a form of methodological triangulation was used to re-examine data originally collected by the first author on women’s experiences of losing and gaining weight after dieting (Popkess-Vawter, Brandau, & Straub, 1998). Whereas the primary analysis was based on reversal theory, the secondary analysis was a content analysis performed by two independent coders “with no consideration” for this theory; the secondary analysis was carried out to provide “a validity check for the primary coding and an accuracy check for complete interpretation” (Popkess-Vawter, Brandau, & Straub, 1998, p. 71).
Secondary studies vary not only in terms of the extent to which their aims diverge from, or converge with, the primary studies from which they are derived, but also according to the number and type of primary studies involved. In amplified analysis, two or more qualitative data sets are utilized. These data may be aggregated to form a larger data set or used to compare different populations. For example, Carl May and colleagues (2004) reused data from a series of studies carried out between 1995 and 2001 to examine how family doctors conceptualized chronic illness and its management in their consultations with patients. An increasing number of this type of secondary analysis has been published since Heaton’s (2004) review, using multiple data sets. Examples include secondary studies based on two data sets (e.g., Long, Sque, & Addington-Hall, 2000), three data sets (e.g., Evans et al., 2012; Murray et al., 2010), four data sets (e.g., Cheraghi-Sohi et al., 2013; Riegal & Dickson, 2016), and six data sets (e.g., Davidson, Edwards, Jamieson, & Weller, 2019). In one study, 13 data sets were reused in one phase of a study (Squires et al., 2015), which also shares characteristics of assorted analysis, described in the next subsection.
In assorted analysis, secondary analysis of qualitative data is combined with additional primary research and/or documentary analysis of relevant materials. For example, in the aforementioned study by Janet Squires and colleagues (2015), a secondary analysis of 13 data sets from four countries was conducted in the second phase of a multiphased investigation of the effects of context on the implementation of interventions. Interviews with 312 health-care professionals were re-examined to learn about the barriers and enablers to the application of research evidence in practice. In the first and third phases, a concept analysis of context and additional semistructured interviews with professionals were conducted to further investigate the ways in which context influences the use of evidence in clinical practice. The results from these three phases were synthesized, and the content validity of the resulting context framework was tested in a final stage of the work.
In this section, three examples of secondary studies drawn from Heaton’s work, based on the different sources of qualitative data identified earlier, are described in depth.
Secondary Study 1: Young Adults With Chronic Illness Project (Using Archived Data)
This project, titled “Mastering chronic illness while growing up: the experiences of young adults and the advice they give to their contemporaries,” was funded by the ESRC (2010–2012). The data used in this study were drawn from three studies that were independently carried out by the HERG at the University of Oxford between 2006 and 2008 and subsequently archived by them for possible use in future research and teaching, with the participants’ consent. As noted earlier, the primary studies were designed to provide public information on what it is like to learn that one has a particular medical condition, what managing it involves, and how it affects people’s lives. A number of topics were covered in the interviews including several that were common across the three studies. For example, the participants were usually asked what advice they would give to other young adults in a similar situation and likewise to health-care professionals who work with people with their condition.
The secondary study came about after the author reviewed the online extracts from two of the HERG projects and obtained a small sample of the full interview transcripts to help define the aims of a proposed secondary study. Two of the primary researchers who had carried out most of the interviews were also consulted to further check that the idea for the secondary study was relevant and viable using the proposed data sets. This process led to a revised set of aims and objectives being defined in relation to three of the data sets in the HERG Archive (Heaton, 2014). The overall aim of the resulting secondary study, which is an example of both supra analysis and amplified analysis outlined earlier, was to examine young adults’ sense of mastery of their chronic illness and how this related to their lived experience of managing the condition while they were growing up. It investigated what it meant to them to successfully manage a long-term condition while growing up, whether and how they achieved this sense of mastery, and how they in turn were helping others by sharing their experiential knowledge online. The study was led by the author of this entry and carried out in consultation with the primary researchers. A group of young adults who were familiar with the HERG program of work were also consulted about the analysis and interim findings in the course of the work.
The three data sets comprised interviews with young people and young adults aged 15–29 who had type 1 diabetes, epilepsy, and a mix of other long-term conditions. Each study had aimed for a maximum variation sample and included people from across the United Kingdom. The type 1 diabetes data set contained interviews with over 35 individuals aged 15–29, diagnosed between the age of 1 and 24 years. The epilepsy data set contained interviews with over 35 individuals aged 16–28, diagnosed between the ages of 3 and 22 years. The long-term conditions data set contained interviews with over 30 people aged 16–29 who had one or more chronic conditions, including juvenile arthritis, epilepsy, muscular dystrophy, asthma, diabetes, cystic fibrosis, and sickle cell disease. Some of the individuals were diagnosed at birth, some as late as 19 years. Overall, 103 of the 105 young adults in the three studies were diagnosed or had symptoms before the age of 20 years and hence had experience of living with a long-term condition for some or all of their teenage years. Two participants were excluded from the secondary analysis because they developed their condition later in life and hence had no experience of growing up with the condition as a young person.
A combination of analytical approaches was used in the secondary study. Thematic analysis was used to identify similarities and differences in participants’ sense of mastery across the sample. This type of analysis, working across cases, helped to map the range of views held by the participants on what it meant to them to control and master their condition, and whether they felt they had ever achieved this sense of mastery. Specialist software was used to help facilitate this part of the analysis, along with charts showing person-by-person views on particular topics of interest. Narrative analysis was also used to examine the ways in which the young adults described the development of their sense of mastery, through a more in-depth analysis of a selection of individual interviews representing different types of experiences. This form of analysis allowed for the relationship between individual participant’s sense of mastery and his or her personal lived experience of a particular condition, as constructed and represented through the narrative, to be examined in detail.
Together, the combined approaches enabled variation in young adults’ ideas about control and mastery to be observed within and across the three original studies. They also provided an insight into how, through the medium of the interviews, the participants shared what they had learned from their experiences for the benefit of others with the same condition. As the research progressed, some of the concepts and tools of complexity theory were also introduced and used to develop the analysis and interpret the findings. The analysis showed how the young adults captured and conveyed in their accounts the mundane complexity of living with and managing a chronic illness, challenging the traditional medical view of the linear relationship between adherence to a therapeutic regime and the achievement of control and autonomy (Heaton, Räisänen, & Salinas, 2016).
This research identified a need for health-care policy and practice to move away from a linear approach centered on promoting control through better adherence, toward a multipronged approach which supports young people and young adults not only to achieve control of their condition but also to retain their autonomy; helps them to build their capacity to adapt to the manifold factors that influence their control and/or autonomy while growing up; and uses testimonies of their experiences of living with a chronic illness as a means of helping others understand and deal with the complexity, uncertainty, and personal nature of the experience.
Secondary Study 2: Dementia Project (Based on Informal Data Sharing)
This ongoing study is based on anonymized qualitative data that were informally shared with the author by the primary research team, following discussions with the lead researcher about the original work and after ethical approval was obtained for the proposed secondary work. The primary study was titled the Memory Impairment and Dementia Awareness Study and funded by the ESRC (2006–2010). In it, participants’ awareness of and coping with memory changes following a diagnosis of dementia were examined using longitudinal and mixed methods. The study was set in north Wales, in the United Kingdom, where a sample of 101 people with early-stage dementia and a relative or friend who knew them well were recruited. As part of the study, these participants were interviewed separately, at two time points, around 12 months apart, generating a large qualitative data set.
The secondary study is a supplementary analysis of the experiences of a subset of the people with dementia who were living alone and their nonresident relatives or friends who were supporting them. There were 24 dyads in this situation. All but four of these dyads were interviewed twice, giving a total of 88 interviews. The sample of people with dementia includes 21 women and three men with dementia, aged 68–91 (average 81); they had Mini-Mental State Examination scores ranging from 19 to 29. The sample of nonresident relatives or friends includes 17 women and seven men, aged 33–79 (average 55), of whom 22 were relatives (11 daughters, 6 sons, 1 sister, 1 brother, and 3 nieces), and two were female friends. All the people with dementia were living alone at the time of the first interview, except for one person who was only living alone at the second interview, after his disabled daughter had left home. At the second round of interviews, two of the participants with dementia had moved into care homes.
In the early stages of examining how the people with dementia were managing to live at home independently, it was noted that they often expressed the desire to continue living at home for as long as possible. Both the people with dementia and their nonresident relatives or friends had been asked for their views on the future and whether they ever discussed it together. The participants also talked about these and related topics at other points in the interviews; hence, there was good coverage of these matters in the data set. Using specialist software, a coding framework was developed to index the participants’ views concerning their future outlook and the ways in which they approached the future. The coded material was retrieved and manually summarized and arranged in grids designed to facilitate analysis of the themes by dyads and by the participants’ sociodemographic characteristics. The ongoing analysis is examining how the people with dementia and their nonresident relatives or friends perceived the future, how they each approached it, and how their respective future outlooks relates to their present well-being.
Secondary Study 3: Hospital Discharge Project (Using Self-Collected Data)
This project was done on the back of a study titled “Carers perspectives on hospital discharge procedures for young adults with physical and complex disabilities,” which was funded by the National Health Service (NHS) National Research and Development Programme for People with Physical and Complex Disabilities, from 1995 to 1997. The purpose of the primary study was to examine informal carers’ perspectives on hospital discharge procedures and the adequacy of continuing care arrangements, for adults aged 18–65 with physical and complex disabilities. It followed the introduction of the Carers (Recognition and Services) Act 1995 in England, which was intended to make health and social care services take into account the needs of informal carers when planning hospital discharge and making arrangements for continuing care in the community for patients.
The primary study was conducted by Heaton, early in her career, with two colleagues at the University of York. Carers of patients with physical and complex disabilities were recruited through various NHS hospital settings in the north of England. The sites included Special Care Units (Younger Disabled Units, a spinal injury unit, and a burn unit) and acute wards. Carers’ and, where possible, patients’ views on hospital discharge and continuing care arrangements were obtained through semistructured interviews conducted up to 3 months after discharge. Managers and professionals from health-care and social care services were also interviewed to explore their involvement in planning hospital discharge and continuing care arrangements. A total of 58 interviews were conducted with 22 carers, 13 of the 21 adults with physical and complex disabilities, 15 managers, and 14 professionals.
A secondary analysis of the data from the study was conducted as part of the author’s part-time doctoral research work. It needed to be distinct from the primary study in order to meet the requirements of the degree. Initially, the aim of the secondary study was to examine carers’ and patients’ experiences, and developments in policy and practice concerning carers and hospital discharge procedures, from a Foucauldian perspective. Analysis of policy documents and official guidelines began while the primary study was underway. Once the interviews had been completed and the data set compiled, carers’ experiences of hospital discharge and continuing care arrangements were examined in the new theoretical context. This aspect of the work was carried on after the primary study had been completed.
As the secondary analysis progressed, it became focused on the temporal organization of hospital discharge from the Special Care Units where the majority of patients had stayed. The primary study had identified issues around the timing of hospital discharge and delays in the implementation of community care packages, but these were examined alongside other matters relevant to the main aims of the research and they were not examined in any depth or in any theoretical context. The secondary doctoral work provided an opportunity to examine the problems that carers and patients reported with various time gaps or disjunctures in the discharge process in more depth, drawing on concepts and theories from the sociology of time; hence, it may be described as an example of supra analysis outlined earlier in this entry.
The results of the secondary analysis showed how the temporal organization of discharge from Special Care Units varied from standard hospital discharge procedures outlined in policy and guidelines in that it was often staggered over a period of time (Heaton, 2001). It also varied depending on whether patients had been admitted to Special Care Units following an emergency hospital admission or for reassessment following a decline in their condition. In the former emergency cases, the timing of discharge was negotiated throughout their stay, whereas, in the latter elective cases, the length of stay was usually shorter and predefined from admission, and these patients also generally went home at weekends.
Analysis of the carers’ and patients’ experiences revealed how, even in Special Care Units where often lengthy stays meant that there was time to plan and phase discharge, the process did not always go smoothly. Some of the problems reported were indicative of underlying tensions between the rhythms and routines of hospital care, community care, and domestic family life. For example, delays in the implementation of arrangements that had been agreed in discharge planning, including provision of equipment and adaptations, demonstrated a lack of working to common time frames operated by hospitals and social services, housing and benefit agencies. In addition, the timetabling of service provision for families was not always synchronized with carers’ needs. While the provision of paid carers after discharge enabled some informal carers to continue working, the scheduling of these hours and the unreliability of paid carers created difficulties for the families. In order for discharge to work more smoothly, it was suggested that there was a need for hospital discharge procedures and community care arrangements to be better coordinated around, and more responsive to, the mundane rhythms and routines of everyday family life.
Issues in Qualitative Secondary Analysis
As noted earlier, the promotion of qualitative secondary analysis has been met with a mixed reception by the research community. Since the mid-1990s, there has been a growing debate over the epistemological, ethical, and methodological aspects of archiving and reusing qualitative data, which are described in the following subsections.
Much of the debate around the reuse of qualitative data has been about whether secondary analysis of this type of data is compatible with the basic tenets of qualitative inquiry. A key concern is whether qualitative data collected to address a primary research question can be reused to address other research questions (Heaton, 2004; Hinds, Vogel, & Clarke-Steffen, 1997; Thorne, 1994, 1998). This problem of data “fit” is seen as a particular problem in qualitative research, where the collection of data might be refined during a study in an iterative process that is responsive to emerging findings, resulting in a data set that is tailored to the primary research questions. Use of open-ended topic guides in interviews can also result in a rich but relatively unstructured data set, where a range of topics are covered in varying degrees of depth, depending on the direction of the interviews, which may not provide sufficient material on the secondary topics of interest. However, others have argued that secondary analysis allows for unexpected topics that emerge from primary research to be followed up and that these are worthy topics of investigation precisely because they have emerged spontaneously (Corti & Thompson, 2004).
Another concern is whether researchers can effectively reuse qualitative data that other researchers have collected (Berg, 2008; Corti & Thompson, 2004; Hammersley, 1997; Heaton, 2004; Mauthner, Parry, & Backett-Milburn, 1998; Thorne, 1994). When the secondary analyst was not involved in collecting the data, he or she does not have the benefit of personal knowledge and experience of being involved in the fieldwork that produced the data. As a result, the secondary analyst lacks the primary researcher’s detailed knowledge and understanding of the context in which the data were collected; he or she also has a relatively distant relationship to the data, which may be compounded by the data set having being anonymized and stripped of other identifying features (although this last point also applies to the primary researchers who conducted the interviews, who also often work with the data in its anonymized state). For some, this means that the secondary analysis of qualitative data in general is not tenable (Mauthner et al., 1998) or that it is not appropriate in relation to some types of qualitative studies where context is particularly important, such as data from participatory research and interpretive phenomenological research, as well as some types of qualitative data, such as field notes (Chauvette, Schick-Makaroff, & Molzahn, 2019).
It has been noted, however, that this problem of not having “been there” is not particular to secondary analysis, as many qualitative studies are carried out by teams of primary researchers whose members are variously involved in the fieldwork (Heaton, 2004). Others have also argued that there are techniques for capturing and sharing contextual information, such as primary researchers providing metadata for their data set; archivists documenting missing data; analysts consulting the researchers who collected the data where this is feasible; and analysts using different processes of reflexivity and critical distance in their work (Corti & Thompson, 2004; Fielding, 2004; Haynes & Jones, 2012). Yet others have suggested, more radically, that the distance of the secondary analyst can be an advantage (Mason, 2007) and that all analysis is relational, given that the production and interpretation of data are always framed by and conducted in relation to the contexts of the investigations (Heaton, 2004; Moore, 2007).
Yet another concern relates to one of the suggested uses of secondary analysis—reanalysis in order to confirm or discount previous research findings. Whether the results of qualitative research can or should be verified in the same ways as studies using statistical methods has been queried by some (e.g., Hammersley, 1997). Here, tensions in the epistemological foundations of quantitative and qualitative research, and across different traditions of qualitative inquiry, are most apparent. Methods of verification and reproduction of findings derived from positivist-based approaches, which underpin the data-sharing imperative in quantitative research, are anathema to many (but not all) qualitative researchers, for whom alternative strategies have been developed to help establish the trustworthiness and authenticity of their work. However, there is some limited support for the idea of preserving data for replication in both quantitative and qualitative research (e.g., Schneider, 2004), although in practice such studies are rare.
There has been a lot of debate about the ethics of archiving and sharing qualitative data sets. Studies of researchers’ attitudes to qualitative data archiving and sharing have found some reluctance to deposit this type of data in archives, linked to their concerns about breaching trust and undermining their relationship with participants, which they perceive to be privileged and special (Broom, Cheshire, & Emmison, 2009; Yardley, Watts, Pearson, & Richardson, 2014). There has been less research on participants’ attitudes to the archiving and sharing of qualitative data. However, one study on data sharing in research in general found that there was conditional public support for it—but also low levels of awareness about existing practices and uses of data (Aitken, de St. Jorre, Pagliari, Jepson, & Cunningham-Burley, 2016).
Whereas in the past it was left to researchers and archivists to decide whether or not to archive qualitative data and, if so, how to protect participants’ identities and sensitive information, increasingly this decision has been shifted to participants themselves. It is now more common for informed consent to be sought from participants in advance, at the point of data collection in the primary study. However, given that information on exactly how the data will be reused, by whom, and for what purposes, will not be available at this point, only a generic form of consent can be obtained at this time (although researchers can give a general idea and some examples of the sorts of uses they anticipate might be possible and permissible). The alternative is to seek consent retrospectively, as and when particular secondary studies are planned. However, this is problematic because it requires that participants’ names and contact details are retained, kept up to date, and can be shared and used for this purpose. Recontacting participants also presents researchers with logistical and ethical difficulties when people have changed address or may have died.
When consent for reuse has not been obtained at the point of data collection, whether analysts decide to seek fresh consent for a secondary study may partly depend on who collected the data and on the type of qualitative secondary analysis planned. For example, when a supplementary analysis is carried out by the same researchers who collected the primary data, and when the aims of the secondary and primary research are relatively congruent, the analysts may adjudge that there is no need to seek further ethical approval (e.g., Bristowe, Selman, Higginson, & Murtagh, 2018). In some cases, data have also been anonymized in order to facilitate secondary analysis when there was no consent for reuse, and it was not feasible to recontact the participants (e.g., Thomson, Bzdel, Golden-Biddle, Reay, & Estabrooks, 2005). It can, however, be difficult to remove all the information that might prevent participants’ identities being inferred when working with small and unique populations, without stripping the data of contextual information and rendering it meaningless for secondary analysis (Parry & Mauthner, 2004).
In addition to obtaining informed consent and anonymizing data, there are other ways in which the interests of participants can be further safeguarded in qualitative secondary studies. One option is for primary researchers to impose conditions of access to the data when they deposit it with an archive. This can limit access to approved users, or for approved uses, or for a period of time; users can also be asked to sign license agreements confirming that they will respect the confidentiality of participants and the stated terms of usage. Another option is for proposed secondary studies to be submitted to research ethics committees for independent review of whether they are consistent with the terms that participants consented to in the primary study and for assessment of the potential benefits and risks of any harm resulting from the work (e.g., Long-Sutehall, Sque, & Addington-Hall, 2010). A further safeguard is for analysts to involve stakeholders in the design and/or conduct of secondary studies to help ensure that they are acceptable to the wider community (Thorne, 1998), although this has seldom been done (for an example of patient and public involvement in a secondary study, see Exemplars, Secondary Study 1).
Methodological work relating to qualitative secondary analysis has tended to focus on the technicalities of preserving qualitative data sets for archiving, rather than the process of reusing the resulting data. Thus, archivists in the United Kingdom and other countries have provided guidelines for researchers on how to obtain informed consent, assign copyright, anonymize data, and create metadata to facilitate qualitative data archiving. A minimum set of guidelines for contextual information required to facilitate the secondary analysis of interview data has also been proposed from a researcher’s perspective (Berg, 2008). An international review of guidelines on the deposition and reuse of qualitative data in archives found that 38% of social science repositories had written guidelines (Antes, Walsh, Strait, Hudson-Vitale, & DuBois, 2018). The review also found variation in the 12 sets of guidelines that were located and that archivists sometimes imposed additional requirements on a case-by-case basis. It concluded that, while some variation and flexibility in guidelines was appropriate, the findings raised questions about best practices in qualitative data archiving and reuse. The same may be said about informal data sharing and the reuse of analysts’ own data, for which there are no guidelines for researchers about the best ways of preserving, sharing, and reusing their data outside of the regulated environment of a dedicated data archive, and where practices are likely to be even more variable.
There are relatively few guidelines for researchers on how to conduct, report, and review the performance of such studies. Criteria for assessing the quality and completeness of primary data sets, and for determining the fit of secondary research questions, have been proposed by Pamela S. Hinds, Ralph J. Vogel, and Laura Clarke-Steffen (1997). Some of the information that might be relevant to include in reports of qualitative secondary studies have also been identified, including about the sources of the data sets used, how the aims of the secondary study related to that of the primary work, whether ethical approval was obtained for the secondary work, and how the secondary researchers’ previous (lack of) involvement in the primary work affected the analysis (Heaton, 2004). In general, however, there still remains a lack of guidelines and codes of conduct for researchers and others to follow on how to design, carry out, report, and appraise the quality of qualitative secondary studies.
The main resources available for researchers to learn more about the methodology are the increasing number of worked examples and commentaries that have been published exploring different aspects of the process of doing qualitative secondary analysis in depth. Examples of these works include a description of the genesis of a secondary study (Heaton, 2014); a detailed account of the process of “sorting” data from two data sets (Long-Sutehall et al., 2010); a worked example of a qualitative secondary analysis and synthesis (Turner, Percival, Kessler, & Donovan, 2018); accounts of different approaches to and processes involved in the secondary analysis of qualitative data, including unsuccessful attempts (Hinds, Vogel, & Clarke-Steffen, 1997); and illustrations of particular approaches to analyzing qualitative longitudinal data (e.g., Davidson et al., 2019; Irwin, 2013; Tarrent, 2016).
Secondary analysis of qualitative data has a complex history. Since the mid-1990s, the retention and reuse of qualitative data from primary research studies has been officially promoted in the United Kingdom and Europe and facilitated by data archives. At the same time, qualitative researchers in these and other countries have pioneered the use of qualitative data obtained through informal data sharing and from their own collections. While official policies have focused on developing the methodology in relation to the formally archived data sets, the underground movement of qualitative secondary analysts has developed the approach more organically, grounded in the practice of qualitative research.
Different patterns of usage have emerged, depending on whose qualitative data have been reused, by whom, for what purposes, and in what combinations, with each permutation raising unique epistemological, ethical, and methodological issues. New patterns of usage continue to emerge, for example, around the growing interest in big data and the potential for pooling multiple archived data sets for secondary analyses based on these new “assemblages” of qualitative data (Davidson et al., 2019). Each turn opens up scope for fresh debates over whether it is appropriate and acceptable for qualitative data to be deconstructed, fragmented, pooled, and reconstituted in this way, and what can be gained by reworking these data in different ways.
With these various developments, the debates surrounding the archiving and secondary analysis of qualitative analysis have crystallized into three broad positions. For some researchers, the notion of reusing qualitative data for anything other than historical research is anathema (e.g., Mauthner et al., 1998) or inappropriate in relation to certain types of qualitative studies or data (e.g., Chauvette et al., 2019). For others, archiving qualitative data is a way of extending the reach of social research, enabling researchers to make more use of existing valuable qualitative data sets, and providing the limitations of the secondary analyst’s distance from the data can be minimized (e.g., Hammersley, 1997). For yet others, the secondary analysis of qualitative data collected by others or by oneself in the past provides an opportunity to carry out social research where the relational nature of data—and the role of distance itself, including its affordances and limitations—is recognized as being integral to qualitative data analysis, where the multiple and mutable contexts of data collection and interpretation can potentially be writ large (e.g., Heaton, 2004; Mason, 2007; Moore, 2007).
Finally, while both official and underground support for qualitative secondary analysis continues to grow, there remains some uncertainty over how it should be promoted in the future (Heaton, 2008). For example, should policy continue to be centered on formal data archiving and sharing or should a mixed economy be officially supported, when informal data sharing and the reuse of self-collected data are also facilitated? Should all qualitative data sets be archived or a selection and, if the latter, based on what criteria? And while there has been some research on researchers’ attitudes to the sharing and reuse of qualitative data, very little is known about the public understanding of and attitudes to the retention and reuse of qualitative data from research, or about how the public and other stakeholders can be involved in the conduct of qualitative secondary studies. There remains a need for debate about the rights and responsibilities of participants, researchers, funders, ethics committees and review boards, social science organizations, publishers, and other stakeholders involved in primary and secondary research, and how their respective interests can be balanced in social research policy and practice.
Corti, L. (Ed.). (2006). Editorial [Special issue]. Methodological Innovations Online, 1. Retrieved from https://journals.sagepub.com/toc/mioa/1/2
Corti, L., & Bishop, L. (Eds.). (2016). Digital representations: Opportunities for re-using and publishing digital qualitative data [Special issue]. SAGE Open. Retrieved from https://journals.sagepub.com/page/sgo/collections/special-issues/digital-representations
Mruck, K. (Ed.). (2005). Secondary analysis of qualitative data [Special issue]. Forum: Qualitative Social Research, 6. Retrieved from http://www.qualitative-research.net/index.php/fqs/issue/view/13
Various. (2007). Special section: Reusing qualitative data (peer reviewed). Sociological Research Online, 12. Retrieved from http://www.socresonline.org.uk/12/3/contents.html
2016). Public responses to the sharing and linkage of health data for research purposes: A systematic review and thematic synthesis of qualitative studies. BMC Medical Ethics, 17, 73. Retrieved from https://doi.org/10.1186/s12910-016-0153-x, , , , & (
2018). Examining data repository guidelines for qualitative data sharing. Ethical Issues in Social Sciences and Health Systems Research, 13, 61–73. doi:10.1177/1556264617744121, , , , & (
1992). The ethnography of a medical setting: Reading, writing and rhetoric. Qualitative Health Research, 2, 451–474. doi:10.1177/104973239200200406 (
2011). Mothers with mental illness experiencing homelessness: A critical analysis. Journal of Psychiatric & Mental Health Nursing, 18, 687–695. doi:10.1111/j.1365-2850.2011.01720.x, , & (
2008). Reanalyzing qualitative interviews from different angles: The risk of decontextualization and other problems of sharing qualitative data. Historical Social Research, 33, 179–192. (
2017). Revisiting qualitative data reuse: A decade on. SAGE Open. Retrieved from https://doi.org.10.1177/2158244016685136, & (
2018). Invisible and intangible illness: A qualitative interview study of patients’ experiences and understandings of conservatively managed end-stage kidney disease. Annals of Palliative Medicine. Retrieved from http://dx.doi.org/10.21037/apm.2018.12.06, , , & (
2009). Qualitative researchers’ understandings of their practice and the implications for data archiving and sharing. Sociology, 43, 1163–1180. doi:10.1177/0038038509345704, , & (
2019). Open data in qualitative research. International Journal of Qualitative Methods, 18. Retrieved from https://doi.org/10.1177/1609406918823863, , & (
2013). Patient priorities in osteoarthritis and comorbid conditions: A secondary analysis of qualitative data. Arthritis Care & Research, 65, 920–927. doi:10.1002/acr.21897, , , , , , … (
2004). Secondary analysis of archived data. In C.Seale, G.Gobo, J. F.Gubrium, & D.Silverman (Eds.), Qualitative research practice (pp. 327–343). London, England: SAGE., & (
2019). Big data, qualitative style: A breadth-and-depth method for working with large amounts of secondary qualitative data. Quality & Quantity, 53, 363–376. doi:10.1007/s11135-018-0757-y, , , & (
Eurolmpact. (2012). Older patients’ attitudes towards and experiences of patient-physician end-of-life communication: A secondary analysis of interviews from British, Dutch and Belgian patients. BMC Palliative Care, 11, 24. Retrieved from https://doi.org/10.1186/1472-684X-11-24, , , , , , , &
2004). Getting the most from archived qualitative data: Epistemological, practical and professional obstacles. International Journal of Social Research Methodology, 7, 97–104. doi:10.1080/13645570310001640699 (
2014). Choose and book: A sociological analysis of “resistance” to an expert system. Social Science & Medicine, 104, 210–219. doi:10.1016/j.socscimed.2013.12.014, , & (
1997). Qualitative data archiving: Some reflections on its prospects and problems. Sociology, 31, 131–142. doi:10.1177/0038038597031001010 (
2010). Can we re-use qualitative data via secondary analysis? Notes on some terminological and substantive issues. Sociological Research Online, 15. Retrieved from https://doi:10.5153/sro.2076 (
2012). A tale of two analyses: The use of archived qualitative data. Sociological Research Online, 17. Retrieved from https://doi.org/10.5153%2Fsro.2523, & (
2001). Hospital discharge and the temporal regulation of bodies. Time & Society, 10, 93–111. doi:10.1177/0961463X01010001006 (
2004). Reworking qualitative data. London, England: SAGE. (
2008). Secondary analysis of qualitative data. In P.Alasuutari, L.Bickman, & J.Brannen (Eds.), The handbook of social research methods (pp. 506–519). London, England: SAGE. (
2014). Growing up with chronic illness: A secondary analysis of archived qualitative data. SAGE Research Methods Cases [online series]. Retrieved from http://methods.sagepub.com/case/growing-up-chronic-illness-secondary-analysis-of-archived-qualitative-data (
2016). Rule your condition, don’t let it rule you: Young adults’ sense of mastery in their accounts of growing up with a chronic illness. Sociology of Health & Illness, 38, 3–20. doi:10.1111/1467-9566.12298, , & (
1997). The possibilities and pitfalls of doing a secondary analysis of a qualitative data set. Qualitative Health Research, 7, 408–424. doi:10.1177/104973239700700306, , & (
2013). Qualitative secondary data analysis: Ethics, epistemology and context. Progress in Development Studies, 13, 295–306. doi:10.1177/1464993413490479 (
1999). Myocardial infarction patients’ use of metaphors to share meaning and communicate underlying frames of experience. Journal of Advanced Nursing, 29, 283–289. doi:10.1046/j.1365-2648.1999.00838.x (
1996). Caring and comfort metaphors used by patients in critical care. Image: Journal of Nursing Scholarship, 28, 349–352. doi:10.1111/j.1547-5069.1996.tb00386.x, & (
2000). Understanding the carers’ world: A biographical-interpretive case study. In P.Chamberlayne, J.Bornat, & T.Wengraf (Eds.), The turn to biographical methods in social science: Comparative issues and examples (pp. 276–289). London, England: Routledge., & (
2014). What motivates Australian health service users with chronic illness to engage in self-management behaviour?Health Expectations, 17, 267–277. doi:10.1111/j.1369-7625.2011.00744.x, , , & (
2000). Conflict rationalisation: How family members cope with a diagnosis of brain stem death. Social Science & Medicine, 67, 253–261. doi:10.1016/j.socscimed.2008.03.039, , & (
2010). Secondary analysis of qualitative data: A valuable method for exploring sensitive issues with an elusive population?Journal of Research in Nursing, 16, 335–344. doi:10.1177/1744987110381553, , & (
2007). “Re-using” qualitative data: On the merits of an investigative epistemology. Sociological Research Online, 12, 1–4. Retrieved from https://doi.org/10.5153/sro.1507 (
1998). The data are out there, or are they? Implications for archiving and revisiting qualitative data. Sociology, 32, 733–745. doi:10.1177/0038038598032004006, , & (
2004). Framing the doctor-patient relationship in chronic illness: A comparative study of general practitioners’ accounts. Sociology of Health & Illness, 26, 135–158., , , , , , … (
2007). (Re)Using qualitative data?Sociological Research Online, 12, 1–13. Retrieved from https://doi.org/10.5153/sro.1496 (
2010). Archetypal trajectories of social, psychological, and spiritual wellbeing and distress in family care givers of patients with lung cancer: Secondary analysis of serial qualitative interviews. British Medical Journal, 304, c2581. Retrieved from http://dx.doi.org/10.1136/bmj.c2581, , , , , & (
2004). Whose data are they anyway? Practical, legal and ethical issues in archiving qualitative research data. Sociology, 38, 139–152. doi:10.1177/0038038504039366, & (
2005). Back to basics: Who re-uses qualitative data and why?Sociology, 39, 337–342. doi:10.1177/0038038505050543, & (
1983). Documents of life: An introduction to the problems and literature of a humanistic method. London, England: George Allen & Unwin. (
1998). Triggers of overeating and related intervention strategies for women who weight cycle. Applied Nursing Research, 11, 69–76., , & (
2016). A qualitative secondary data analysis of intentional and unintentional medication nonadherence in adults with chronic heart failure. Heart & Lung, 45, 468–474. doi:10.1016/j.hrtlng.2016.08.003, & (
2004). Building a scientific community: The need for replication. Teachers College Record, 106, 1471–1483. (
2015). Identifying the domains of context important to implementation science: A study protocol. Implementation Science, 10, 135. Retrieved from http://dx.doi.org/10.1186/s13012-015-0325-y, , , , , , … (
2016). Getting out of the swamp? Methodological reflections on using qualitative secondary analysis to develop research design. International Journal of Social Research Methodology. Retrieved from https://doi.org/10.1080/13645579.2016.1257678 (
2005). Central questions of anonymization: A case study of secondary use of qualitative data. Forum: Qualitative Social Research, 6, Art. 29. Retrieved from http://dx.doi.org/10.17169/fqs-6.1.511, , , , & (
1994). Secondary analysis in qualitative research: Issues and implications. In J. M.Morse (Ed.), Critical issues in qualitative research methods (pp. 263–279). London, England: SAGE. (
1998). Ethical and representational issues in qualitative secondary analysis. Qualitative Health Research, 8, 547–555. Retrieved from https://doi.org/10.1177/104973239800800408 (
2018). Synthesizing qualitative data sets to improve the design of trails and complex health interventions: A worked example. Qualitative Health Research. Retrieved from https://doi.org/10.1177/1049732318803894, , , & (
2014). Ethical issues in the reuse of qualitative data: Perspectives from literature, practice, and participants. Qualitative Health Research, 24, 102–113. Retrieved from https://doi.org/10.1177/1049732313518373, , , & (
2014). Using secondary analysis of qualitative data of patient experiences of health care to inform health services research and policy. Journal of Health Services Research & Policy, 19, 177–182., & (