EDMG611Wk5

Nurse Researcher

May 2015 | Volume 22 | Number 5 © RCNi / NURSE RESEARCHER8

Correspondence to Catherine Houghton [email protected]

Catherine Houghton RGN, RCN, BN, MHSc, PhD is a lecturer at the School of Nursing and Midwifery, National University of Ireland, Galway, Republic of Ireland

Kathy Murphy BEd, MSc, PhD is professor of nursing at the National University of Ireland, Galway, Ireland

David Shaw PhD Csci is a lecturer at the Open University, Milton Keynes, UK

Dympna Casey RGN, MA, PhD is a senior lecturer at the National University of Ireland, Galway, Ireland

Peer review This article has been subject to double-blind peer review and checked using antiplagiarism software

Author guidelines journals.rcni.com/r/ nr-author-guidelines

Qualitative case study data analysis: an example from practice

Cite this article as: Houghton C, Murphy K, Shaw D, Casey D (2015) Qualitative case study data analysis: an example from practice. Nurse Researcher. 22, 5, 8-12.

Date of submission: February 2 2014. Date of acceptance: April 16 2014.

Introduction Qualitative case study methodology (QCSM) has become a comprehensive approach to describing and exploring complex issues relevant to nursing (Luck et al 2006, Anthony and Jack 2009). Its flexibility allows for creativity in its implementation, yet it is rigorous in gaining an in-depth understanding of the field of interest (Keyzer 2000, Pontin 2000). However, there is little practical guidance on how data analysis is conducted in QCSM.

This paper describes an approach to analysis in case-study research based on the strategies of

Miles and Huberman (1994), guided by the analytical principles outlined by Morse (1994).

Background QCSM is suitable for exploring experiences and situations when the researcher is interested in a phenomenon and the context in which it occurs (Yin 2003, Luck et al 2006, Salminen et al 2006). As a methodology, it can be used to examine or explore events, situations, programmes and activities (Berg 1995, Hancock and Algozzine 2006). There is ample literature available to guide researchers on making methodological decisions regarding

Abstract Aim To illustrate an approach to data analysis in qualitative case study methodology.

Background There is often little detail in case study research about how data were analysed. However, it is important that comprehensive analysis procedures are used because there are often large sets of data from multiple sources of evidence. Furthermore, the ability to describe in detail how the analysis was conducted ensures rigour in reporting qualitative research.

Data sources The research example used is a multiple case study that explored the role of the clinical skills laboratory in preparing students for the real world of practice. Data analysis was conducted using a framework guided by the four stages of analysis outlined by Morse (1994): comprehending, synthesising, theorising and recontextualising. The specific strategies for analysis in these stages centred on the work of Miles and Huberman (1994), which has been successfully used in case study research. The data were managed using NVivo software.

Review methods Literature examining qualitative data analysis was reviewed and strategies illustrated by the case study example provided.

Discussion Each stage of the analysis framework is described with illustration from the research example for the purpose of highlighting the benefits of a systematic approach to handling large data sets from multiple sources.

Conclusion By providing an example of how each stage of the analysis was conducted, it is hoped that researchers will be able to consider the benefits of such an approach to their own case study analysis.

Implications for research/practice This paper illustrates specific strategies that can be employed when conducting data analysis in case study research and other qualitative research designs.

Keywords Case study data analysis, case study research methodology, clinical skills research, qualitative case study methodology, qualitative data analysis, qualitative research

Data analysis

© RCNi / NURSE RESEARCHER May 2015 | Volume 22 | Number 5 9

classification and identification of the case, and the collection of multiple sources of evidence (Stake 1995, Yin 2003, Casey and Houghton 2010).

However, methodological guidance regarding data analysis techniques and processes is limited (Tellis 1997, Yin 2003), with few accounts of the inherently intuitive analytical procedures used provided by researchers (Tesch 1990). An integrative review of QCSM by Anthony and Jack (2009) acknowledged data analysis but only broadly, outlining that the principles of content analysis were applied.

Other qualitative methodologies, such as phenomenology, ethnography and grounded theory, have principles from their philosophical underpinnings that guide analysis and provide researchers with justification for the analytical decisions made. Researchers conducting QCSM need to provide the same justifications for decisions and provide rigorous reports from their findings, often with complex sets of data from multiple sources.

Furthermore, in multiple case study research, there are two types of analysis: ‘within-case’ and ‘cross-case’ (Eisenhardt 1989, Miles and Huberman 1994, Creswell 1998, Stake 2006). Within-case analysis provides a detailed description of each case and themes within it (Creswell 1998); cross-case analysis is carried out to analyse themes across the cases to identify similarities and differences (Eisenhardt 1989).

In qualitative data analysis, themes are developed that capture and unify the nature of the phenomenon (DeSantis and Ugarriza 2000). However, it is not enough to explore themes as separate entities – their inter-relationships must be clarified or the analysis will be incomplete (Ayres et al 2003).

This paper gives a detailed account of the analytical steps and process used to analyse the data in a multiple case study exploring the role of Ireland’s Clinical Skills Laboratory (CSL) in preparing nursing students for the real world of practice. This analysis, which is based on the strategies of Miles and Huberman (1994), and guided by the analytical principles of Morse (1994), provides an analytical framework that may be adapted and applied by future researchers using QCSM.

The CSL is a learning laboratory, often designed to replicate a real clinical environment, where students can learn and practice clinical skills in a safe and supervised manner.

Research example The example study was doctoral research exploring the role of the CSL in preparing nursing students for the real world of practice. This qualitative investigation used a multiple case study design.

Five case study sites were selected from 13 higher education institutes that offer the bachelor of nursing degree programme in Ireland (Houghton et al 2013a). The aims of the research were to explore the strategies used for teaching and assessment in the CSL, and how these were perceived. The factors that help or hinder students’ learning and implementation of clinical skills in practice were also described. The role of the CSL was then clarified, highlighting strategies that can prepare nursing students for their experiences in clinical practice.

The case study was instrumental and embedded. An instrumental case study investigates the case to provide insight into a particular issue (Stake 2000, Casey and Houghton 2010). This case study was instrumental, in that its purpose was general understanding rather than understanding of a particular case. An embedded study has specific units of analysis (Yin 1994, Casey and Houghton 2010). The subunits in this example included the teaching strategies in the CSL, the factors that influenced students’ implementation of skills in practice, and also staff and students’ perspectives of clinical skills education and practice.

Data were collected using semi-structured interviews (n=58) and non-participant observation of students implementing their skills in the clinical setting at each of the five case study sites. In addition, documentary analysis of relevant national and international guidelines for nurse education and competency assessment documents was conducted. The analysis provided the context for the presentation of the findings and was integrated in the final stages of the write up. Only the analysis of the interview and observation data are discussed in detail in this paper.

Approach to data analysis There are no systematic rules for analysing qualitative data. Thorne (2000) stated that ‘unquestionably, data analysis is the most complex and mysterious of all the phases of a qualitative project.’ However, the aim is to rigorously and creatively organise, find patterns in, and elicit themes from data (Burnard and Morrison 1994). There must be logic behind the analysis and therefore a framework (Yin 1998).

Morse (1994) provided an all-encompassing framework for analysis based on four stages: ‘comprehending’, ‘synthesising’, ‘theorising’ and ‘recontextualising’. However, this framework does not provide sufficient detail about the practical skills needed for analysis, so strategies for analysis are also necessary.

The strategies developed by Miles and Huberman (1994) have been influential in case study research (Yin 2003, Evers and van Staa 2009, Simons 2009),

Art & science | acute care

May 2015 | Volume 22 | Number 5 © RCNi / NURSE RESEARCHER10

Nurse Researcher

and have been successfully implemented in this research design (Casey 2007), Houghton et al 2013b). In this example, the stages proposed by Morse (1994) provided the framework for analysis, and the strategies outlined by Miles and Huberman (1994) were used to put the framework into practice. This approach was congruent with the principles of content analysis, which aims to identify prominent themes and patterns among data (Patton 2002).

An overview of the analysis is provided in Table 1. In this table, the framework is illustrated by the four stages of comprehending, synthesising, theorising and recontextualising. The analysis strategies recommended by Miles and Huberman (1994) – broad coding, pattern coding, memoing, distilling and ordering, testing executive summary statements, and developing propositions – are applied to their relevant part of the analysis framework. The purposes of these strategies are explained in the third column to further emphasise their relevancy.

Data management Practically, the handling of qualitative data can be overwhelming, with traditional approaches involving large amounts of paperwork. Computer-assisted qualitative data analysis software (CAQDAS) has been developed to assist in the handling, storage and manipulation of the data (Bringer et al

2004, MacMillan and Koenig 2004, Silverman 2010). This allows for quick and easy retrieval of data, and provides a comprehensive approach to management (Lathlean 2006). However, it must be remembered that the software is incapable of understanding or giving meaning to text and cannot replace the analytical skills of the researcher (Bringer et al 2004, Lathlean 2010).

QSR NVivo was the chosen CAQDAS for this research because it allowed the researcher to manage data and ideas, and query the data (Bringer et al 2004, Bazeley 2007). QSR NVivo can assist with within- and across-case analysis, and therefore it can be used appropriately for case study research (Bassett 2010).

CAQDAS can also enhance transparency and rigour (Crowley et al 2002, Richards 1999, Bringer et al 2004). For example, QSR NVivo has a variety of search and retrieval tools, called queries, that enable the researcher to ask questions or test emerging themes (Bassett 2010).

Stages of analysis Each stage of Morse’s (1994) analytical framework will be described, outlining how the analytical strategies of Miles and Huberman (1994) were implemented in the research example, and how they were handled in QSR NVivo.

Comprehending According to Morse (1994), the first stage of analysis is ‘comprehending’, which begins while collecting data. The aim is to gather enough data to be able to write a complete, detailed, coherent and rich description (Morse 1994).

Comprehending involves initial coding called ‘broad coding’ (Miles and Huberman 1994). A code is a descriptive or conceptual label that is assigned to excerpts of raw data (Gale et al 2013). In NVivo, coding is assisted by structures called ‘nodes’, which provide the storage capacity for references to coded text (Bazeley 2007). Broad coding aims to uncover and develop concepts, and the text must be opened up, so that thoughts, ideas and meanings contained in it can be exposed (Miles and Huberman 1994). It decontextualises the data, because information is removed from its context (Tesch 1990, Ayres et al 2003). Stake (1995) recommended using some pre-established codes, but also advocated searching for additional ones.

In the research example, all interview and observation data were imported into QSR NVivo 8. As there was such a large data set, to make the coding process more manageable, it was necessary to create a provisional ‘start list’ of codes based on the research questions.

Stages of analysis Morse (1994)

Analysis strategies (Miles and Huberman 1994)

Purpose

1 Comprehending Broad coding General accounting scheme that is not specific to content but points to the general domains in which codes can be developed inductively.

2 Synthesising Pattern coding Memoing

Explanatory, inferential codes to create more meaningful analysis. ‘One of the most useful and powerful sense-making tools at hand’ (Miles and Huberman 1994).

3 Theorising Distilling and ordering. Testing executive summary statements

Memos tie together different pieces of data into a recognisable group of concepts. ‘Building towards a more integrated understanding of events, processes and interactions in the case’ (Miles and Huberman 1994).

4 Recontextualising Developing propositions

Formalise and systemise into a coherent set of explanations.

Table 1 Stages of data analysis

© RCNi / NURSE RESEARCHER May 2015 | Volume 22 | Number 5 11

Data analysis

The observational data and interview data were coded into broad codes linked to the primary aims of the research. The broad codes were further distinguished by perspective: clinical staff, academic staff, students, newly qualified nurses and observational data. This provided a general accounting scheme that was not content specific, but identified general domains in which codes could be developed inductively (Miles and Huberman 1994).

To do this required the use of ‘tree nodes’ in QSR NVivo. These have the capacity to develop hierarchical relationships with other nodes (Bassett 2010). Using tree nodes assists in organisation, conceptual clarity and identifying patterns among the data (Bazeley 2007).

The clinical staff were further distinguished between clinical nurse managers, clinical placement co-ordinators and student preceptors. The academic staff were further distinguished between CSL manager, CSL technician, nurse tutor and lecturer. The students were coded as junior or senior students (Table 2). In addition, QSR NVivo’s ‘attributes’ allowed the storage of factual information about participants, such as demographic data, which could be used to seek patterns and ask questions about the data at a later stage (Richards 1999).

Synthesising ‘Synthesising’ is the merging of perceptions and cases to describe typical, composite patterns (Morse 1994). This involves organising and coding the data, and can be achieved by a process known as ‘pattern coding’ (Miles and Huberman 1994). The purpose of this type of coding is to reassemble data that were fractured during broad coding.

In the research example, this involved the use of explanatory, inferential codes to create more meaningful analysis (Miles and Huberman 1994). Using tree nodes, codes that accurately reflected the perceptions of the participants and the observations made by the researcher were created. This was a lengthy process, because each previously identified code (tree node) had a large number of codes within it. As new codes were identified, the researcher needed to return to previously coded data to ensure there were no missed codes.

Writing memos is another synthesising strategy. From the pattern codes, ‘memos’ can be formulated. Memos are summaries of key information derived from the coding system; they lay the foundation for further development of propositions regarding the data (Miles and Huberman 1994). They can also be termed ‘executive summary statements’. They were drafted in the form of a memo against each theme and each perspective.

Theorising ‘Theorising’ involves building a comprehensive and coherent account of the data (Morse 1994) by examining the relationships between the identified categories of data (Tesch 1990). Theorising need not necessarily focus on the development of theory, but rather the examination of relationships among the data. In relation to the strategies proposed by Miles and Huberman (1994), this involves ‘building towards a more integrated understanding of events, processes and interactions in the case’ by ‘distilling and ordering’ the memos and testing against the data executive summary statements made about the data. It is an active, continuous and rigorous process of viewing and challenging the data (Morse 1994).

In terms of the example study, this was important for making comparisons across case study sites, as well as comparing participant group perspectives. This process provided a clearer picture of the perceptions of the participants and how they could be linked with the observational data. For example, if an observation was made, it was possible to ascertain what participants from that site had said about a particular situation or phenomenon.

The executive summary statements connected the data into a recognisable group of concepts (Miles and Huberman 1994). The codes were arranged into two main themes – ‘creating a bridge to practice’ and ‘the real world of practice’ – both of which had three subthemes. The executive summary statements were tested against the data to ensure that statements made about the data could be traced back to, and found in, the data. This allowed for findings to be confirmed and representativeness assured.

Participant type Further distinction

Clinical staff ■  Clinical nurse managers. ■  Clinical placement

co-ordinators. ■  Student preceptors.

Academic staff ■  Clinical Skills Laboratory (CSL) manager.

■  Clinical Skills Laboratory technician.

■  Nurse tutor. ■  Lecturer.

Students and newly qualified nurses

■  Junior student. ■  Senior student. ■  Newly qualified staff nurse.

Table 2 Participant type, coding by perspective

Nurse Researcher Data analysis

May 2015 | Volume 22 | Number 5 © RCNi / NURSE RESEARCHER12

This was done manually by the researcher, but the process also used QSR NVivo query tools, which are described in more detail in relation to rigour (Houghton et al 2013a).

Recontextualising This involves the development of propositions that may be applicable to settings and populations. It is the elegance of these propositions that makes qualitative enquiry transferable (Morse 1994, Ayres et al 2003). The researcher can begin to sharpen and shape the propositions (Miles and Huberman 1994). Furthermore, the findings can be compared with those of previous research, thus enhancing the rigour of the research (Eisenhardt 1989).

The strategies for ensuring the rigour and quality of the example study were discussed in another paper (Houghton et al 2013a). Once the executive summary statements had been tested, they were formalised and systemised into a coherent set of explanations, linked as memos to the subthemes

in QSR NVivo. They formed the basis for the final presentation of the findings.

Conclusion There is a lack of literature outlining qualitative data analysis processes in QCSM. This research example outlines a framework for data analysis in case study research that can be adopted by other researchers. This framework provided a comprehensive and practical approach to analysis, yet maintained the artistry and intuition necessary for insightful qualitative research.

It is important that qualitative case study researchers are systematic in handling large data sets from multiple sources of evidence. This systematic approach allows for cross-case comparisons to be made and also provides evidence for ensuring the quality of the research. By using a framework such as the one described, portraits of phenomena can be most rigorously and articulately displayed.

References Anthony S, Jack S (2009) Qualitative case study methodology in nursing research: an integrative review. Journal of Advanced Nursing. 65, 6, 1171-1181.

Ayres L, Kavanaugh K, Knafl KA (2003) Within-case and across-case approaches to qualitative data analysis. Qualitative Health Research. 13, 6, 871-883.

Bassett BR (2010) Computer-based analysis of qualitative data: NVivo. In Mills AJ, Durepos G, Wiebe E (Eds) Encyclopedia of Case Study Research. Sage Publications, Thousand Oaks CA.

Bazeley P (2007) Qualitative Data Analysis with NVivo. Sage Publications, London.

Berg BL (1995) Qualitative Research Methods for the Social Sciences. Second edition. Allyn and Bacon, Boston MA.

Bringer JD, Johnston LH, Brackenridge CH (2004) Maximizing transparency in a doctoral thesis 1: the complexities of writing about the use of QSR*NVivo within a grounded theory study. Qualitative Research. 4, 2, 247-265.

Burnard P, Morrison P (1994) Nursing Research in Action: Developing Basic Skills. Second edition. Macmillan, Basingstoke.

Casey D (2007) Nurses’ perceptions, understanding and experiences of health promotion. Journal of Clinical Nursing. 16, 6, 1039-1049.

Casey D, Houghton C (2010) Clarifying case study research: examples from practice. Nurse Researcher. 17, 3, 41-51.

Creswell JW (1998) Qualitative Inquiry and Research Design. Choosing Among the Five Traditions. Sage Publications, Thousand Oaks CA.

Crowley C, Harré R, Tagg C (2002) Qualitative research and computing: methodological issues and practices in using QSR NVivo and NUD*IST. International Journal of Social Research Methodology. 5, 3, 193-197.

DeSantis L, Ugarizza DN (2000) The concept of theme as used in qualitative nursing research. Western Journal of Nursing Research. 22, 3, 351-372.

Eisenhardt KM (1989) Building theories from case study research. Academy of Management Review. 14, 4, 532-550.

Evers JC, van Staa AL (2009) Qualitative analysis in case study. In Mills AJ, Durepos G, Wiebe E (Eds) Encyclopedia of Case Study Research. Sage Publications, Thousand Oaks CA.

Gale NK, Heath G, Cameron E et al (2013) Using The Framework Method for the Analysis of Qualitative Data in Multi-Disciplinary Health Research. tinyurl.com/p532uxl (Last accessed: April 2 2015.)

Hancock DR, Algozzine B (2006) Doing Case Study Research. First edition. Teachers College Press, New York NY.

Houghton C, Casey D, Shaw D et al (2013a) Rigour in qualitative case study research. Nurse Researcher. 20, 4, 12-17.

Houghton CE, Casey D, Shaw D et al (2013b) Students’ experiences of implementing clinical skills in the real world of practice. Journal of Clinical Nursing. 22, 13–14, 1961-1969.

Keyzer DM (2000) Nursing research in practice: the case study revisited. Australian Journal of Rural Health. 8, 5, 266-270.

Lathlean J (2010) Qualitative analysis. In Gerrish K, Lacey A (Eds) The Research Process in Nursing. Sixth edition. Blackwell Publishing, Oxford.

Lathlean J (2006) Qualitative analysis. In Gerrish K, Lacey A (Eds) The Research Process in Nursing. Fifth edition. Blackwell Publishing, Oxford.

Luck L, Jackson D, Usher K (2006) Case study: a bridge across the paradigms. Nursing Inquiry. 13, 2, 103-109.

MacMillan K, Koenig T (2004) The wow factor: preconceptions and expectations for data analysis software in qualitative research. Social Science Computer Review. 22, 2, 179-186.

Miles MB, Huberman AM (1994) Qualitative Data Analysis: An Expanded Sourcebook. Second edition. Sage Publications, Thousand Oaks CA.

Morse JM (1994) ‘Emerging from the data’: the cognitive processes of analysis in qualitative inquiry. In Morse JM (Ed) Critical Issues in Qualitative Research Methods. Sage Publications, Thousand Oaks CA.

Patton MQ (2002) Qualitative Research and Evaluation Methods. Third edition. Sage Publications, Thousand Oaks CA.

Pontin D (2000) Observation. In Cormack D (Ed) The Research Process in Nursing. Fourth edition. Blackwell Publishing, Oxford.

Richards L (1999) Data alive! The thinking behind NVivo. Qualitative Health Research. 9, 3, 412-428.

Salminen A, Harra T, Lautamo T (2006) Conducting case study research in occupational therapy. Australian Occupational Therapy Journal. 53, 3-8.

Silverman D (2010) Doing Qualitative Research. Third edition. Sage Publications, London.

Simons H (2009) Case Study Research in Practice. Sage Publications, London.

Stake RE (1995) The Art of Case Study Research. Sage Publications, London.

Stake RE (2000) Case studies. In Denzin NK, Lincoln YS (Eds) Handbook of Qualitative Research. Second edition. Sage Publications, Thousand Oaks CA.

Stake RE (2006) Multiple Case Study Analysis. Guilford Press, New York NY.

Tellis W (1997) Introduction To Case Study. tinyurl.com/3oke8oe (Last accessed: April 2 2015.)

Tesch R (1990) Qualitative Research: Analysis Types and Software Tools. RoutledgeFalmer, New York NY.

Thorne S (2000) Data analysis in qualitative research. Evidence Based Nursing. 3, 3, 68-70.

Yin RK (1994) Case Study Research: Design and Methods. Second edition. Sage Publications, Thousand Oaks CA.

Yin RK (1998) The abridged version of case study research. In Bickman L, Rog DJ (Eds) Handbook of Applied Social Research Methods. Sage, Thousand Oaks CA.

Yin RK (2003) Case Study Research: Design and Methods. Third edition. Sage Publications, Thousand Oaks CA.

Conflict of interest None declared

Online archive For related information, visit our online archive and search using the keywords