RESEARCH IN SOCIAL CARE: GUIDELINES FOR RESEARCHERS
8 ANALYSE THE DATA AND INTERPRET THE FINDINGS
There are many different ways of analysing the data coming from social care research, but a distinction is often made between qualitative and quantitative data.
The analysis of qualitative data
- Qualitative data analysis collates and summarises the mass of information generated by interviews or observation
- It involves researchers seeking relationships between different themes that have been identified for the research and creating patterns out of the data
- The behaviour, experiences and attitudes of respondents may be related to their biographical characteristics, such as age, sex, income level and so on
- Implications for policy or practice may be derived from the data
- Theory may be tested or developed
- Although methods of analysis vary greatly, the following steps are typical:
- Tape recorded interview material is transcribed
- Anonymising of data
- Familiarisation with the data through repeated reading
- Organising of the data for easy examination, for example, by setting key facts or comments out on a spread sheet or framework
- Use of a specialist computer programme such as NVivo or NUD*IST
- Identification of themes related to the original research question
- Development of provisional categories
- Exploration of relationships between categories
- Refinement of themes and categories
- Choice of examples to illustrate themes and categories
- Comparison of new data with original research questions
The analysis of quantitative data
- Quantitative data analysis collates and summarises the mass of information generated by interviews and other forms of data collection
- In the initial analysis the characteristics of the data may be described and explored by using cross tabulations, creating graphs and charts, and calculating means and standard deviations
- Further analyses will build on these initial findings, seeking patterns and relationships in the data by comparing means, exploring correlations, and performing multiple regressions or analyses of variance
- Advanced modelling techniques may be used to build sophisticated explanations of how the data addresses the original questions
- Although methods may vary greatly, the following steps are typical:
- Identifying a data entry and analysis programme eg SPSS
- Coding the data and reviewing it for completeness
- Entering the data and checking
- Carrying out simple exploratory analyses
- Reviewing the results
- Performing more sophisticated analyses as appropriate
- Development of theory
- Comparison of new data with original research questions
Using existing data
There are many sources of data already in existence which could be used to explore social and health care issues. Many of these data sets contain information from thousands of people, are based on nation-wide random samples and are processed in an extremely professional way. They thus offer data in quality and quantity which far surpasses anything which can be obtained in the average project. The two points to consider are whether a particular data set contains the information that is required and whether the researcher has the skills to manage the analysis.
Existing data sets can be obtained from the ESRC Data Archive at the University of Essex. For example:
- The English Longitudinal Survey of Ageing (ELSA) collects data about 12,000 people over 50 and about how their health, social and economic circumstances change over time
- The General Household Survey contains data from nearly 13,000 households about their use of social and health services and quality of life
- The Millennium Cohort Study collects longitudinal evidence about 19,000 children born in 2000 and about their birth and early years, their family circumstances and their physical, emotional and social development
- The National Child Development Study follows the lives of 17,000 people who were born in 1958; it contains information about health and social care, family relationships, and other topics
