Qualitative data analysis
- Nicole Mc Neilly
A lot of the data you’ve collected might be qualitative - from text and written responses, to observations you have documented, recordings you have collected and more. This can be initially very overwhelming; a short interview can still create pages of data. While analysing qualitative data can be the subject of an entire degree course, here we outline the core steps of analysing this:
Review the data you have collected.
Creating a set of references you are looking for in the data, e.g. learning or network outcomes, that are drawn from your change pathway outcomes and indicators. If you are analysing text, assigning colours to each reference can be helpful for this, and you can use the comment function in Google Sheets.
Mark (or ‘code’) any interesting points that emerge (e.g. interesting points of view or quotes) on your first read.
Formulate a general description of the data. This is your first step of drawing conclusions, which we go into in more detail in the next step.
Review the data again, and mark any new information that didn’t seem so important or that you missed the first time. The more you read the data, the more you might find, especially if you have collected multiple viewpoints that you can compare.
Come back to the data with fresh eyes and repeat!
Tip.
It’s helpful to take notes as you’re reviewing the data to document your thoughts and observations that aren’t directly related to your analysis - this is common practice in the social sciences.
Your analysis centres on identifying and describing patterns.
Do you see any patterns? Are they expected or unexpected?
Do you see any changes in patterns over time e.g. by comparing new data to the baseline set?
Can you explain what each pattern means or why they are important?
What is missing from your analysis that you might have expected?
How did participants respond to certain questions? For example, did they hesitate in an interview when they were talking about whether or not they enjoyed something? Did questionnaire respondents avoid one particular question?
Tip.
Be ready from the beginning to document any recommendations on how to improve your work that might emerge from the data analysis stage. This is a great tangible output for your colleagues and it will help convince them of the benefit of investing in this type of research.
Before focusing on conclusions, you want to be as certain as possible that you have drawn all relevant information out of your data. It never hurts to have another look at the data or to ask someone else to do so. Taking a break from the data is also beneficial - you can come back with fresh eyes.