Ole Steen Kristensen’s research aimed to identify the difficulties of young people transitioning from foster care to independent adulthood from their perspective. In his study of how their present life is characterized by accumulated memories of the past and fragmented ideas about the future, he faced the challenge of synthesizing thousands of articles presenting diverse perspectives and methodologies. NVivo software, combined with Endnote, enabled him to effectively manage the sorting, sifting and analysis stages resulting in a quality assessed metasynthesis of the literature.
Background
Ole Steen Kristensen’s research aimed to identify the difficulties of young people transitioning from foster care to independent adulthood from their perspective. In his study of how their present life is characterized by accumulated memories of the past and fragmented ideas about the future, he faced the challenge of synthesizing thousands of articles presenting diverse perspectives and methodologies. NVivo software, combined with Endnote, enabled him to effectively manage the sorting, sifting and analysis stages resulting in a quality assessed metasynthesis of the literature.
The Challenge - to generalize from diverse research results
How Is it possible to generalize results from a large collection of qualitative research studies on a particular topic? The diversity of qualitative research with its range of concepts, methods and designs, makes such a task very complex.
Also, in reviewing the literature on this study, Ole Steen faced a number of further challenges:
- The current, poor indexing of foster care research in bibliographic databases.
- After unexpected cancellation of access to the initial social science and sociological abstracts database, changes were required after the switch to a psychological abstracts bibliographic database.
- How to find an overarching approach or theory from the varied individual-level theorising that focuses on children in care.
- Each of the qualitative studies had their own way of doing research and accounting for their approaches. For Ole Steen, paying respects to well implemented research, while developing his own approach, proved potentially difficult.
Ole Steen decided to conduct a meta-synthesis of the qualitative studies in this area. A meta-synthesis is a systematic method to compare and generalize findings from several qualitative studies. It aims to portray an accurate interpretation of a phenomenon and to compare and contrast the constructs of individual studies to reach consensus on a new construction of that phenomenon. It involves: identifying findings, grouping findings into categories and grouping categories into synthesised findings (Kastner et al, 2012).
The Tools: The value of combining NVivo with EndNote
Given the diversity of qualitative research articles, Ole Steen began with a broad search of the online databases – PsychInfo and the Social Science Citation Index - through EndNote. The first search string contained twenty-one different terms. Criteria for article selection included that research samples be adolescents or young adults between 13 and 29 years old, in out-of-home placement, in the middle of transition in work or education. Preferred articles were in English. The breadth of the first search of sociological and psychological databases resulted in 1043 articles.
“EndNote was a relief -easy to work with and use in this kind of work"
For the initial work, EndNote simplified viewing and grouping articles and sifting out those covering irrelevant topics.
Exclusion criteria included articles on adoption, homelessness, youth crime, excess weight, sexual problems, psychiatric hospitalisation, elderly people, dementia, somatic problems (IPV, hepatitis, HIV) and out of home placements due to disabilities. Other exclusions were professionals’ perspective on transition, articles on supervision, therapy, grant proposals, policy analyses, technical papers on assessment, manuals and statistical indicators. Ole Steen also excluded research with focus groups as their research goals were not compatible with the project topic. This sorting and sifting resulted in 124 articles of interest.
Benefits of using NVivo for managing, querying and coding articles: Quickly import articles, abstracts and notes in one place
For the next step, NVivo enabled Ole Steen to import the 124 selected articles from EndNote into his NVivo project. Although it was his first experience with NVivo he was able to do this easily and quickly.
After import, the articles attached to the original EndNote references were now all easy to access in NVivo as files ready for coding and querying. The import also automatically extracted the collection’s abstracts and notes into NVivo memos linked to the relevant articles.
With NVivo searches and text queries it was easy to find and remove from the 124 articles those exclusively on childhood trauma with no direct bearing on the study. This resulted in 70 articles ready for quality assessment.
Quality assessment of articles with NVivo
For appraising the 70 qualitative articles, Ole Steen drew on current frameworks of appraisal criteria. The most comprehensive, by Spencer et al (2003), focuses on 16 different appraisal questions for assessment of qualitative studies plus quality indicators for each question.
By developing a coding scheme Ole Steen was able to sift the qualitative studies according to criteria related to:
- Aims
- Methodology and methods
- Sampling
- Data collection
- Theoretical approach
“It was my first experience of NVivo…but I had no problems [with NVivo]”
From this coding for the criteria and quality indicators, the appraisal process excluded more articles for the following reasons
- Inadequate sampling strategies
- Lack of systematic data analysis
- Quantification of qualitative information
- Results without documented support or citations
After appraisal there were 30 articles remaining for analysis in NVivo.
Ease of thematic analysis with NVivo
After more detailed, refined coding, Ole Steen worked with NVivo queries. The highly heterogenous nature of the diverse studies meant that NVivo Word Frequency queries did not reveal sufficient key terms to suggest codes. However, Ole Steen found the coding queries very helpful for robust, in-depth exploration. For example, after coding for different family and social connections, the ease of running NVivo Boolean coding queries across the different codes, enabled him to explore and expand more in-depth issues of social support networks of the transitioning youth.
“NVivo coding queries are very useful. Printing their results you can see how your codes change during your work”
Future plans for working with NVivo
From the ease with which NVivo enabled sound coding and querying in the metasynthesis, Ole Steen is now using the results to develop a hypothesis about transitioning to independent adulthood for a future publication. Based on his experience with NVivo in the metasynthesis, Ole Steen will continue to use NVivo in his research. His next study focuses on people with severe illness and how illness affects their time perspective.
With this study having benefited from NVivo’s robust coding and querying functions, in future studies Ole Steen plans to explore NVivo classifications in more detail. He will do this to compare useful article characteristics, including different research approaches within different disciplines in addition to dates of publication and countries of origin.
Furthermore, in his current teaching and supervision work, Ole Steen’s students are now using NVivo and EndNote to facilitate their research projects. They are also benefiting from the comprehensive and quality assessed usable database of literature in the foster care and transition study field that resulted from NVivo’s coding and querying functionality applied to EndNote data.
“I have discovered from the metasynthesis it is easy to generate a new hypothesis about foster youth”
For more information on the results of Ole Steen’s metasynthesis see:
Kristensen, O. Attuning to the past while aging out of care – a Metasynthesis. Online: https://pure.au.dk/portal/files/121836846/final_ECQI_OSK_at_Leuven.pdf
Reference
Kastner M, Tricco AC, Soobiah C, et al. What is the most appropriate knowledge synthesis method to conduct a review? Protocol for a scoping review. BMC Medical Research Methodology. 2012;12:114. doi:10.1186/1471-2288-12-114.
Spencer,L, Ritchie,J., Lewis, J and Dillon, L. 2003. Quality in Qualitative Evaluation: A framework for assessing research evidence. [Online: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/498322/a_quality_framework_tcm6-38740.pdf]
User information
About Ole Steen Kristensen
As professor in Educational Psychology in the Department of Psychology and Behavioural Sciences, at Aarhus University, Denmark, Ole Steen Kristensen divides his time between research and teaching and supervises postgraduate students.
The topic of his PhD (published 1989 in Danish) was "A social psychological study of adult education" and his fascination with teaching has led him to explore and develop teaching methods with computer technology. His research projects have focused on participatory research and developmental projects with current interests in:
- Transition from care to independent life of former foster care youth.
- The effect of interventions aimed at children and adolescents in foster care.
- Applications of interactive e-books in teaching.
- Studies of organisational culture and complex interventions.