Diabetes care areas - click here to navigate to care area page
Ask a question
As of April 1st 2013 NHS Diabetes became part of NHS Improving Quality. Please direct your enquiry to firstname.lastname@example.org
Thank you, your question has been sent. We aim to get back to you within five working days.
Step 3 - Interpretation > What's the story?
Analysing survey data can be challenging for those without specific skills or experience of doing so. We recommend seeking assistance from Research and Development staff within the Trust, unless this element of the survey is contracted out to an external supplier.
There are a number of techniques that can be undertaken on quantitative data to increase your understanding of what your patients think of the services you provide, over and above simple top-line percentages. The extent to which this type of analysis is possible will depend on the number of survey responses you receive. Options include:
- Cross-tabulation analysis Analysis conducted to compare views of different respondent types (e.g. demographics) to help identify groups in more or less need.
- Multivariate regression & drivers analysis - Statistical modelling techniques designed to quantify the relationship between multiple variables. Often used for drivers of satisfaction, etc.
- Cluster analysis - Used to group respondents who possess similar characteristics.
With regards to analysis of interview transcripts we recommend one individual reading through each in turn (where resources allow) to get a broad understanding of the issues being discussed. As already identified, feedback generated through qualitative techniques will not be statistically robust, and so you should look to conduct thematic analysis to highlight common factors.
The same process should be used when reading and interpreting the feedback generated through paper diaries.