Enhancing the Transparency of Qualitative Data Analysis:a Case Study using Software for Qualitative Research : Enhancing the Transparency of Qualitative Data Analysis:a Case Study using Software for Qualitative Research Simon Maxwell
Environment and Society Research Unit
Department of Geography, University College London, UK
Presentation to the Third ASC International Conference
University of Edinburgh
22nd-24th September 1999
Introduction : Introduction Analytic processes in qualitative research are often mystical
Software for qualitative research can help enhance transparency in analytic processes
Grounded theory approach to research
This combination can aid ‘rigour’ and quality in qualitative research
These issues are of paramount importance as qualitative methods are increasingly used more directly in decision-making
Structure of presentation : Structure of presentation Background
A grounded theory approach to qualitative analysis
Software for qualitative research
Making data and analytic processes accessible
Case study: public meanings of car use
Discussion
Conclusions
Background : Background New methodologies to address exclusion from decision-making and lack of trust in public institutions
Better understandings of the public’s own priorities are central
Qualitative research more open to bias?
Baxter and Eyles (1997): emphasize the importance of detailing the analytic process
Geertz (1973): ‘detailed, thick description’
Grounded theory 1 : Grounded theory 1 Based on the work of two sociologists: Barney Glaser and Anton Strauss (e.g., Glaser and Strauss, 1967)
An approach to an entire research project:‘In this method, data collection, analysis, and eventual theory stand in close relationship to one another. A researcher does not begin a project with a preconceived theory in mind (unless his or her purpose is to elaborate and extend existing theory). Rather, the researcher begins with an area of study and allows the theory to emerge from the data’ (Strauss and Corbin, 1998)
Each step of a grounded theory analysis requires careful documentation
Grounded theory 2 : Grounded theory 2 Line-by-line analysis; constant checking of data against other data; open, axial and selective coding; extensive use of diagrams; theoretical memos
The importance of staying close to the data
The process is creative
Aims to recognize the complexities, contradictions and ambiguities of human actions
Software for qualitative research : Software for qualitative research Many packages designed around the concepts of grounded theory
Core functions for analysis of textual data- easy and flexible access to transcribed data- variety of means for coding data- recording notes- building links
Audio recordings contribute to ‘thick description’
ATLAS.ti used in case study
Making data and analytic processes accessible : Making data and analytic processes accessible To the researcher or research team
To principal users of the research
To wider users
Case study: public meanings of car use 1 : Case study: public meanings of car use 1 Case study project titled Creating places: using deliberative and inclusionary processes to explore public meanings of car use
Funded by the UK Economic and Social Research Council (ESRC)
Grounded theory approach
In-depth discussion group methodology
Each group had between eight and ten members and met once a week for five weeks
In-depth group work with parents of young children and older people
Case study: public meanings of car use 2 : Case study: public meanings of car use 2 All group discussions were recorded and transcribed
Audio files were transferred to computer using Cool Edit software
All transcripts were coded on a line-by-line basis
Codes also developed for group dynamics
‘Emic’ codes represent the first level of analysis made available to research users
Codes printed from ATLAS.ti with details showing location of all corresponding primary text
Transcript extract with emic codes from ATLAS.ti : Transcript extract with emic codes from ATLAS.ti
Extract of code list from ATLAS.ti : Extract of code list from ATLAS.ti Code: It was burnt out: a good reason not to have a car {1-0}P 3: Parents 3.txt, 232 - 235Code: The bus service was good {1-0}P 3: Parents 3.txt, 238 - 240Code: Has to be a balance between stick and carrot {1-0}P 3: Parents 3.txt, 228 - 229Code: We never had a car when we lived in Liverpool {1-0}P 3: Parents 3.txt, 230 - 232Code: Inconvenience of looking after a car {1-0}P 3: Parents 3.txt, 240 - 241
Case study: public meanings of car use 3 : Case study: public meanings of car use 3 ‘Emic’ codes and theoretical memos cut up and sorted and sifted into categories by hand
ATLAS.ti used to explore additional dimensions of coding
An intensive, dynamic and iterative process
Categories represent the conceptual categories and theoretical constructions emerging from the research (‘etic’ codes)
‘Etic’ codes relate back to empirical work through ‘emic’ codes
‘Etic’ codes represent the second level of analysis made available to users
Transcript extract with etic codes from ATLAS.ti : Transcript extract with etic codes from ATLAS.ti
Discussion : Discussion The analytic process represented aims to be a creative mix of paper-based and computer-based methods
Aims to acknowledge the benefits of more traditional and technological approaches to qualitative analysis
The process outlined is just one way of using software for qualitative research
Computer-based approaches are central to enhancing the transparency of the analytic process
Computerized techniques can appeal to a desire to impose control on qualitative data
Very difficult to retain the creativity essential to qualitative analysis if the process is carried out entirely on computer
Conclusions : Conclusions Software for qualitative research can:- open analytic procedures to critical evaluation by other researchers and a wider audience- improve 'rigour' and quality in qualitative research
Greater emphasis is needed on the transparency of analytic processes as qualitative techniques are used more directly in decision-making