Analysing complexity: developing a modified phenomenological hermeneutical method of data analysis for multiple contexts

ABSTRACT Qualitative data analysis has been criticised for a lack of credibility over recent years when vagueness has been afforded to the reporting of how findings are attained. In response, there has been a growing body of literature emphasising a need to detail methods of qualitative data analysis. This paper adds to this body of knowledge by presenting a reflexive narrative review of the development of a modified version of the ‘phenomenological hermeneutical method for interpreting interview texts’. This modification permits greater transparency when dealing with data that relates to multiple contexts. A visual model to represent this modified approach is also presented. Additionally, an exploration of the underpinning theoretical basis to this data analysis method, and modification, is provided. Accordingly, this paper demonstrates this modified qualitative data analysis method ensures that subjective relativistic origins and contexts of interpretations are identifiable, whilst permitting full objectification of the text.


Background
Qualitative data analysis has been criticised for a lack of credibility over recent years when vagueness has been afforded to the reporting of how findings are attained. For example, Bazeley (2009) argues that 'thick description' alone is not sufficient to constitute full analysis. To counter these arguments there has been a growing body of literature emphasising a need to detail, more comprehensively, methods of qualitative data analysis (e.g., Flood, 2010;Smith & Firth, 2011). This paper adds to this expanding body of knowledge. It presents a reflexive review of the development of my modified phenomenological hermeneutical method of qualitative data analysis; an approach to analysis which permits transparency when dealing with data relating to multiple contexts. This paper also presents the model I developed that seeks to provide a concise visual representation of this modified method.
The modified data analysis method to be presented was devised to permit analysis of qualitative data gathered in my hermeneutic phenomenological research study that investigated undergraduate student nurse experiences of learning during study abroad journeys (Morgan, 2018(Morgan, , 2019. Participants in the study (N = 20) had undertaken a nursing placement in one of three study abroad types: • EU students who came to the UK for a three month higher education exchange, also comprising a clinical placement (n = 6); • UK students who undertook a four week clinical placement to countries outside of the EU (non-EU) comprising Africa and Asia (n = 6).
Data was gathered from two digitally recorded semi-structured interviews per participant: a postreturn interview, conducted as soon as possible after the end of the study abroad experience; and a follow up interview, conducted three months after return. Follow up interviews were conducted on campus or via Skype (in the case of EU students who had returned to their home countries). These interviews were designed to elicit the 'lived experience' of students with a focus upon learning: processes, influences, strategies and impact. Interviews were transcribed into written text format for analysis. Whilst the focus of my research was upon the pedagogy of student nurse study abroad, I anticipate this modified analysis method holds utility for hermeneutic phenomenological research which is located within other disciplines and subjects. To facilitate the reader to consider this transferability, this article will present a reflexive review of the development and application of my modified data analysis method. I have named this method a 'Modified Phenomenological Hermeneutical Method Of Data Analysis For Multiple Contexts'. I adopt a narrative approach for this reflexive review, and integrate examples from my analysis to illuminate its emergence and application. At each stage of the analysis process, I complement the narrative by presenting each component of my visual model. Finally, I draw these components together to present the fully assembled model.
The extracts of analysis, included in this paper, relate to participant experiences of students who undertook the non-EU study abroad placement type (Africa and Asia). As this was a small cohort, to protect participant identity, the specific country visited will not be identified, and pseudonyms will be applied. In particular, I will focus upon Alex, a UK student who visited a rural African area. Alex's study abroad experience was arranged (with university support) through a private UK company who organise healthcare placements across the world. During the four week placement, Alex lived in a secure, Company student house, with other international health students. Staff at the house included a local cook and a local manager. Alex's placement experience comprised an allocation to the local hospital, in which she rotated between clinical areas. She was assigned to local Registered Nurses, whom she 'shadowed'. In addition, Alex also took the opportunity to volunteer at a local orphanage. The company also arranged social and cultural activities with fellow students.

Reflexive review and modification of a phenomenological hermeneutical method of analysis
When commencing my research I selected Lindseth and Norberg's (2004, p. 145) 'phenomenological hermeneutical method for interpreting interview texts' as my analysis method. I felt this approach was closely aligned to the ontological and epistemological origins of my research and it offered a transparent process with practical application. As discussed, my research followed a hermeneutic phenomenological philosophy which emphasises the acquisition and interpretation of situated (in space and time) subjective experience (gathered though interviews) as key to disclosing phenomena, with the aim to provide an enhanced understanding of being-in-the-world (Heidegger, 1962). Lindseth and Norberg's method is specifically designed for analysis of such phenomenological interview data that has been transcribed into written format. Importantly, the key principles of phenomenology, such as revealing subjective 'lived experience', are evident throughout Lindseth and Norberg's data analysis process, and these philosophical underpinnings are clearly translated to offer a transparent and practical approach to analysis. As part of this transparency, I felt Lindseth and Norberg's process permitted an understanding of phenomena to emerge whilst also protecting against any limiting effects of researcher presuppositions (or implicit assumptions about the world). For example, researcher presuppositions could limit the emergence of meaning and the extent of understanding (this will be further elaborated upon later in this paper). Hence, as I had pre-existing experience of the topic area under investigation, I was concerned to ensure understanding emerged inductively from participant experience, and was not limited by presupposition. Lindseth and Norberg's (2004) method mirrors Ricoeur's (1971Ricoeur's ( , 1976 Theory of Interpretation which proposes that interpretation of the text moves through a process from guess (about a phenomenon) to validation (confirming understanding of a phenomenon). Lindseth and Norberg's method reflects this process of movement and comprises three key stages: naïve reading (from which the guess emerges), structural analysis (an explanation stage, where understanding is generated from systematic analysis of the text) and comprehensive understanding (the 'interpreted whole' (p. 150), when a deep, and validated, understanding of the text and the phenomenon has emerged). Lindseth and Norberg's method therefore provides a systematic method of data analysis that ensures understanding has emerged from subjective 'lived experience', and which has not been limited by researcher presupposition. This method permits a clear record, and audit trail, that evidences development of such emergent findings and understanding of phenomena. These stages of analysis are outlined, in depth, below.
Stage one: naïve reading Lindseth and Norberg (2004) describe that the text must be read many times in an open-minded manner so that a first 'naïve understanding' is arrived at. At this stage, the researcher will consider neither the research question nor their prior understanding to reduce influences of presupposition. Lindseth and Norberg (2004, p. 148) use the term 'bracketing' in order to be open to experience. The term 'bracketing' is most usually interpreted to be a suspension of prior understandings. However, hermeneutic phenomenological philosophy suggests it is not possible to suspend or 'bracket' prior experience or understanding (Heidegger, 1962). I therefore alternatively employed reflexivity which ensures the influence the researcher exerts upon the research is identified and managed (Speziale & Carpenter, 2007). For example, to reflexively monitor and prevent the influence of presupposition, I kept a researcher journal to illuminate and reflect upon my thoughts when undertaking naïve reading (and throughout the research process).
This first stage facilitates the 'guess' process and a subjective and surface (or superficial) understanding of the whole experience is attained (Ricoeur, 1971(Ricoeur, , 1976.
The following extract, from analysis of Alex's return interview, highlights this stage. Alex's return interview was conducted within 2 weeks of return from Africa. Interview duration was 1 hour 30 minutes and when transcribed, this equated to a text of 9,221 words. Naïve reading of this text permitted the following naïve understanding to emerge. This naïve understanding is presented in the note format in which it was originally generated to preserve authenticity of the analysis in action: 'Alex was very motivated to make a difference. The clinical mentor was very supportive and helped her to learn, and the hospital was focused on teaching the students. But the learning experience was not just formal lectures and explanations, it was the exposure to the reality of lack of resources and medical conditions, of poverty, that brought most insight. Alex described her learning as a mindset change, an understanding that things just mean different things to different cultures, and due to their circumstances, e.g., meanings of life and death. And she also noted the centrality of spirituality within the culture, which was not evident in the UK. Alex learnt through reflection with self and with others, through talking, and support of the other students in her house. These students formed a community very quickly.
They were not absorbed into the local community as the security restrictions prevented this, the gap company identified that they could be a target for pickpockets for example, -the Other, but they were made to feel welcome e.g., Hair-braiding by locals. There were identified as being from the gap company as all the locals knew who they were and they wore t-shirts when out -it wasn't a 'touristy area but the real Africa'.
Alex felt she could make a small difference. She describes culture shock when first arriving -chickens and cows etc. but noted how this became the norm very quickly.
She notes that the experience forced her to learn and also forced her to apply her knowledge, and this amazed her as she didn't realise she had learnt so much in her UK studies.
She noted that the language could be a barrier to communicating but later in the interview she also noted how this was a learning facilitator as it helped her to learn to communicate without language.
She was apprehensive about returning to UK placements as she felt that this experience may make it difficult for her to work with the 'worried well'; and identified that 'in the grand scheme of things' their worries are nothing.
The meaning of being a learner during study abroad for Alex was about being supported by community -of nurses, of gap company, of other learners and the local community. At the same time the totality of being in the experience meant she was forced to learn, forced out of her comfort zone, forced to face reality, forced to take on responsibility. The experience was about being reflexive and reflective throughout. It is a spiritual experience and enabled understanding of culture that went beyond the basics i.e. it was a philosophical learning experience -an understanding of how cultures view things differently -an ontological difference? Alex was motivated to take on these challenges as her aim was to make a difference.' The model component representing stage one of analysis -'stage one: naive understanding' is presented in figure 1.

Stage two: structural analysis
Stage two is the explanation stage, it is the mid-stage between surface and depth understanding and applies the principle of distanciation (Ricoeur, 1976). Distanciation is a process that places the text distant from its temporal origins. It escapes the original intentions of the author (Ricoeur, 1973); (who in this case, I have interpreted to be my study participants as they provided the content of the text during interview). Distanciation permits horizons to be enlarged and is considered productive when it enables meaning to be generated beyond that which the originators of the text may have seen from their horizon (Ricoeur, 1976). The researcher is demonstrated here to be an active interpreter of the text, assigning meanings to the given content. Distanciation, applied by Lindseth and Norberg (2004) during structural analysis, commences with the identification of 'meaning units' from the text. A meaning unit captures a single meaning and it may vary in length from a few words to a paragraph. Meaning units therefore isolate and remove text from context. The following are examples of identified meaning units that I generated from the text of Alex: Meaning unit 1: 'I think being put in those circumstances it forces you to learn' Meaning unit 2: 'The knowledge I gained in the UK, it kind of reinforced on international placement. And I was like "well would they not do this?". And it was all little bits that I hadn't realised I'd learnt, but it gave the opportunity to force me to apply it and think through things logically, using that background experience to think through why they were doing things or what could be done. It was really good and I amazed myself sometimes it was "oh I did know that!" and I didn't realise how much I did know about that . . . ' The researcher then condenses these meaning units into shorter items, or condensed units, which retain the original meaning, and these are then reflected upon and compared and contrasted with one another. Examples of condensed units I developed from Alex's meaning units highlight this process: Condensed unit 1: 'Circumstances forced to learn' Condensed unit 2: 'Forced to think and apply knowledge gained in UK' Condensed unit 3: 'Reinforced learning in UK' Themes and sub themes are then assembled from these condensed units. For example, sub themes that I assembled from Alex's condensed units include: Sub theme: 'Being forced to learn' Sub theme: 'Manifesting latent knowledge' To prevent researcher presupposition from limiting understanding, these themes and sub themes are further reflected upon to identify whether they validate, or confirm, the original naïve understanding. If the themes do not validate this understanding then the cycle of analysis commences again, re-commencing at stage one. Lindseth and Norberg (2004) suggest that several rounds of analysis may be required to uncover various meanings that may exist about the phenomenon. Therefore, this process continues until no new meanings emerge and the naïve understanding has been confirmed. It is important to note that the term structural analysis is interpreted here to mean analysis of the structures (or themes) of the phenomenon (Van Manen, 1997) rather than its application in pure hermeneutics, when it may imply that it is the structure of the text itself that is the subject of analysis.

Reflexive development of the structural analysis stage
Analysis of each individual participant's data progressed well, however, whilst Lindseth and Norberg (2004) offer insight into how to analyse an individual transcript, I felt it was not clear at which point the researcher begins to compare and contrast each individual interview text analysis (and therefore subjective reality) with others to form an overall, or comprehensive, understanding of the phenomenon. Further, as my research comprised three different contexts in the form of study abroad types (described above), I was also concerned to ensure that each individual 'lived experience', within each of these three different study abroad types, were explored so that an understanding of learning in each type could be revealed prior to developing a comprehensive understanding of the phenomenon more generically. Whilst Lindseth and Norberg suggest reflecting on the emergent themes in relation to the context of the research (such as different study abroad types) at a final 'comprehensive understanding' stage, I felt it was important to preserve experiences specific to each study abroad type throughout each stage of data analysis in order that were not 'lost' during this process.
To capture such contextual elements, Bazeley (2009) also recommends describing, comparing and relating the characteristics or situation of the participants during analysis. I therefore incorporated these aspects into Lindseth and Norberg's approach so that the varied study abroad types could be explored individually before then combining with the other types. In order to achieve this, I further differentiated at the stage of structural analysis. Accordingly, I introduced two sub-stages, these are: 'individual structural analysis' (as described above); and an additional 'combined structural analysis'. This development represents a refinement in relation to moving from the individual participant experience (which I have termed 'the individual horizonal perspective' or 'individual horizon') to combined experiences of the phenomenon (respectively termed 'the combined horizonal perspective' or 'combined horizon'). The model component representing the individual and combined horizonal perspectives is presented in figure 2.
During combined structural analysis, participant analyses were grouped by context (the study abroad types). Then once in their grouping, a 'combined context-grouping naïve understanding' was produced. This comprises a synthesis of each individual participant's understanding.
The next extract is from the 'combined context-grouping naïve understanding' of the non-EU study abroad type, which includes the experiences of Alex. This example discusses my naïve understanding of student arrival experiences: ' . . . Arrival for Jack was described as surreal and of being gobsmacked when travelling from the airport to the house. Jill noted that everything was so different on arrival, ranging from the physical environment to the driving. She also noted that there was no concept of time. Alex notes culture shock when first arrivingchickens and cows etc. but noted how this became the norm very quickly. Andy was absorbing things when she first arrived and it took a day to sink in. Lucy's initial arrival experience was one of shock when travelling from the airport and seeing local life and the student noted how the shock became less as she got used to the experience. Marty felt culture shock when arriving for placement as everything was so different. She saw differences in life, culture and healthcare . . . ' Following generation of a 'combined context-grouping naïve understanding', individual sub themes that had emerged from each group member's participant interviews were compared and contrasted, and 'context-grouping sub themes' emerged. As highlighted below, this extract is of UK students who went to non-EU countries: Extract of context-grouping sub theme: 'arrival shock and disjuncture' Jack . L16-23. ' . . . it was surreal, we had a drive from the airport to the house, me and X didn't really speak to each other we just looked out of the window and were gob-smacked about the place, it was just, it was what I was expecting but it wasn't at the same time. it was so surreal.' Jill .  experience when you watch it on the telly or anything like that. And I knew that I wasn't going to be naïve, as I obviously didn't know what it was going to be like, but I did try to visually prepare myself. The chickens were a shock and the cattle just waltzing past the front doors.' Andy . L8-11. ' . . . It didn't sink in until a day afterwards. It was about a 30-40 minute drive, we got picked up. I was absorbing everything from the start.' Lucy . L14-15. 'You saw all the children working and houses on the way from the airport, and it was "oh my god".' Marty . L74-76 ' . . . It was really busy. There was an armed guard at the gates, erm, so yeah, I suppose it was just a bit of a culture shock, like it was just completely different to anything I'd seen before . . . ' When all three study abroad types had been analysed and the 'context-grouping naïve understanding' confirmed for each (by reflecting against the emergent sub themes), the three context-grouping understandings (representing each study abroad type) and all sub themes were then compared and contrasted and developed as phenomenon sub themes and themes. For example, the sub theme of 'arrival shock and disjuncture', was evident in all study abroad types (but, as I had analysed each research context separately, I was able to identify that this manifestation varied depending on the extent of difference between usual and the study abroad reality) so this became a sub theme which fed into the broader theme 'experiencing and making sense of the different reality'.
To represent this modified second stage of analysis, I have developed model component -'stage two: modified structural analysis' (figure 3).
As indicated, by analysing study abroad types separately during structural analysis, it was possible to identify the similarities and differences for each type of study abroad experience, this proved invaluable during the final stage of analysis, which is the comprehensive understanding stage.

Stage three: comprehensive understanding
Stage three followed Lindseth and Norberg (2004) method to attain a comprehensive understanding. Themes were further considered in relation to study abroad type and research question. All text was re-read, and, as advocated by Lindseth and Norberg, other texts were also considered. For my research, to broaden the horizon of understanding and prevent researcher presuppositions from limiting the findings, literature that comprised theory pertinent to expand understanding of the emergent themes, was considered. For example, theories of liminality, communities of practice, transformative learning and threshold concepts resonated with the research findings. As an educationalist I had a surface awareness of these theories, therefore in depth review of this body of literature was undertaken to confirm whether these resonations could be validated and utilised. This process of engaging with literature generated a deeper understanding of the text and the phenomenon. From this widened horizon, sub themes and themes were further refined. As Lindseth and Norberg caution that it is vital not to force literature upon data, it was necessary at this stage of analysis to ensure that the literature was congruent with the text. Literature was therefore used to 'illuminate the interview text and interview text illuminate the chosen literature' (Lindseth & Norberg, 2004, p. 151).
This stage of analysis is an appropriation, or a 'making of ones own' the text in order that possible ways of being-in-the-world and meanings may emerge (Ricoeur, 1976). This form of appropriation is considered by Klemm (1983, p. 143) to reflect 'the world projected by the text' and represents 'distanciation linked to the full objectification of the text' (Ricoeur, 1976, p. 74). As identified by Ricoeur (1976)  Contexts' therefore enables an interpretation that permits understanding to develop from a surface (a superficial understanding) to a deep understanding (an understanding emergent from analytical processes, in this case analysis of participant experience and supported by theory), whilst considering the multiple contexts in which the phenomenon exists. This process of appropriation consequently resulted in development of a comprehensive understanding of the phenomenon that also integrates pertinent theory.
Stage three of analysis is represented in model component -'stage three: comprehensive understanding' (figure 4).
Prior to completion of the comprehensive understanding, it is important to also note that, as is evident above, movement from sense to reference is not linear. The movement between the parts and the whole, to develop and then validate, or confirm, interpretations, occurs throughout the data analysis process, this is frequently referred to as the hermeneutic circle (Gadamer, 2004  movement is represented by the inclusion of arrows throughout each stage of my developed model. Ricoeur (1976) further considers this hermeneutic process to comprise a hermeneutic arc, as the, previously discussed, explanation stage forms a bridge between surface and depth semantics. Whilst Klemm (1983) additionally suggests it is more accurate to consider a spiral process whereby there is an ever-deepening comprehension when striving to attain authenticity through appropriation. For my analysis, I synthesised these suggested hermeneutic reflective approaches and represent my application in the model component highlighted in Figure 5.
In line with this approach, to facilitate completion of the process of data analysis, I additionally undertook a final round of confirmation by comparing the developed comprehensive understanding against the previous texts. Finally, I followed Lindseth and Norberg's (2004)   recommendation that the results of analysis should be conveyed to the audience in an everyday language. Such conveyance ensures that the interpretation offered remains true to the phenomenological principle of gaining an insight into the world as it is lived by the participants. This reporting also permits opportunities for appropriation by future readers (Ricoeur, 1976), who may choose to apply the findings, as they feel relevant, to their own practice contexts. This, as suggested by Lindseth and Norberg (2004) is the point at which interpretation may be most productively applied to life.
An extract of the final confirmed comprehensive understanding for the phenomenon of learning during study abroad is offered below. This understanding retains an everyday reporting style whilst necessarily incorporating the expanded theoretical horizon into this narrative.
'Student nurse learning during study abroad is experienced within the context of difference and liminality as students anticipate and then undertake a journey into, through and return from a different reality . . . students experience disjuncture when exposed to difference in the study abroad reality as students' frames of reference do not explain the reality as it manifests. Greater disjuncture and also shock are experienced when the degree of difference between realities is significant . . . Being a student nurse learner during study abroad therefore means to be a liminal entity, actively making sense of difference, disorienting dilemmas and troublesome experiences in the liminal space of the different reality. Learning is influenced by others and the outcome of the experience of learning during study abroad is change and transformation.'

Further theoretical considerations
At an epistemological level, my 'Modified Phenomenological Hermeneutical Method Of Data Analysis For Multiple Contexts', has enabled a symbiosis of the potentially conflicting philosophical elements of Heidegger's and Ricoeur's theories. Specifically, as discussed, Heidegger gives emphasis to subjectivity and temporality, in order to generate understanding of phenomena as they are experienced in relation to points in time (Heidegger, 1962). Whilst Ricoeur advocates for objectification of the text, through distanciation and appropriation (Ricoeur, 1976). My modified approach to analysis preserves both Heidegger's and Ricoeur's principles. Firstly, my approach captures and reflects upon the subjective experience of the study participant, in context (for example, study abroad type). Secondly, application of both distanciation and appropriation, enables the researcher to interpret subjective experiences presented across these multiple contexts and from a widened horizon. My modified approach therefore permits understanding to progress beyond the individual context of the horizon in which the experience is situated. This enables the structures, or themes, that comprise the phenomenon, irrespective of contextual specifics, to emerge. Finally, engaging hermeneutic reflections, by validating each stage of understanding with previous stages, returns interpretation of the text to its subjective origins. In this way my modified process of data analysis addresses any concerns that objectification of the text may

Move on to combined structural analysis
Researcher's Interpretation (spiralling within and between stages, with ever developing understanding) Figure 6. Visual model of the 'modified phenomenological hermeneutical method of data analysis for multiple contexts' (using the example of study abroad research which comprised x 3 contexts (study abroad types) and x 2 interviews per participant).
lose sight of subjective realities when multiple contexts (such as study abroad placement types) are being researched.

Presenting the visual model representative of a 'modified phenomenological hermeneutical method of data analysis for multiple contexts'
As identified, to illuminate the modified data analysis process employed, I have developed a unique visual model ( Figure 6). This model represents my interpretation of Lindseth and Norberg (2004) approach and includes the reflexive modifications I made to the data analysis process to preserve experiences in context. It also demonstrates application of the iterative approach provided by the hermeneutic circle (Gadamer, 2004). This interpretation comprises, within its overall framework, the hermeneutic arc of Ricoeur (1976) and, when each stage is considered together, it represents a hermeneutic spiral (Klemm, 1983). As highlighted, this process permits movement from guess to validation, so moving from a surface to a deep or comprehensive understanding of the phenomenon, whilst preserving each unique experience in context. This method therefore resolves philosophical tensions between subjectivity and objectivity.

Conclusion
This paper has provided a narrative review of the reflexive development and application of my modified approach to data analysis when there are multiple contexts. The modifications I have made to Lindseth and Norberg (2004) 'phenomenological hermeneutical method for interpreting interview texts' are specifically the inclusion, at stage two, of two sub stages: 'individual structural analysis' and 'combined structural analysis', and also the development of a unique visual model of the modified data analysis process.
Returning to the background rationale offered at the beginning of this paper, my 'Modified Phenomenological Hermeneutical Method Of Data Analysis For Multiple Contexts' addresses concerns raised about qualitative data analysis' lack of credibility and vagueness of reporting of how findings are attained. My modified approach ensures the subjective origins of an interpretation continue to be identifiable, and the individual voice and context of each participant is preserved throughout the data analysis process. This approach to analysis permits transparency of movement from individual subjective realities to combined analysis of multiple realities, in different contexts, whilst providing a clear audit trail. This modified data analysis method is therefore important as it has demonstrated trustworthiness and offers confidence that emergent analysis is reflective of 'lived experiences' of being-in-the-world.
Finally, whilst the focus of my research is upon the pedagogy of student nurse study abroad, as the analysis method is underpinned by hermeneutic phenomenological philosophy, this modified analysis method is transferable to research conducted in other disciplines and subjects. My 'Modified Phenomenological Hermeneutical Method Of Data Analysis For Multiple Contexts' thus holds utility for application to a wide range of hermeneutic phenomenological research which comprises multiple contexts.

Disclosure statement
No potential conflict of interest was reported by the author.

Notes on contributor
Debra A. Morgan is a Senior Lecturer in Nursing, and also a Director of Transnational Education, at Northumbria University. Since joining the University, I have been extensively involved in development of student mobility programmes and international partnerships. This work includes providing both internal and external consultancy.
I am passionate about the provision of evidence-based educational practice, and, to this effect, I am research active, currently focusing on transcultural education. I value engagement with a global community of practice and regularly attend, and present at, national and international events. I also co-lead the Advance HE Connect 'Authentic Learning for Global Citizenship' Network.