5 steps to revealing your data’s hidden secrets
28 October 2013 - IN analysis, annotations, coding, data, matrix, memos, models, query
When you start your research project, you probably first read some literature, then make a decision on your methodology, and then design your data collection.
After the hard work of data gathering, you sit there with your material and wonder:
"What on earth am I supposed to do with this?"
You might try to meditate in front of your data, or put it under your pillow to catch the deepest secrets hidden in it. Maybe your data will speak for itself?
I am sorry to disappoint you, but this will never happen. Instead, you will need to work with your data: read it, code it, read it, code it, read it—and this is where NVivo bursts onto the scene, like a prince on his white horse :)
Here's the path I followed to analyze my data and write-up my findings (with just a little help from that prince and his horse):
Step 1: Organizing your data
You have gathered different kinds of data and need to think what data goes to Internal Sources and what you might put into External Sources in NVivo.
In my study, I put my interview audios and transcriptions (plus the participants’ information sheet) into Internals and photos into Externals. In nodes, I created a ‘people’ node which allowed me to combine all the information about participant in one place and add the node attributes (like age, gender, country of origin etc.).
Step 2: Initial coding phase
You might start by reading and annotating your data. While reading, annotations are easily added to your data in NVivo. It is also wise to record your thinking in memos. In my project I had memos for my research design, reflective journal, coding, methodology and analysis.
Gradually by reading and re-reading the data (including annotations and memos), your analytical approach begins to emerge and you can start your initial coding.
After completing some coding you might make a model to visualize your thinking:
Step 3: Coding-on
Initial codes can be coded again, coded-on. In my study, I coded-on to nodes for the people and emotions involved in the initial codes. I also auto-coded the free-telling and prompted parts of the interviews and had them as 'on-codes'.
Step 4: Exploring the data
After coding and coding-on you can use queries to explore the patterns in your source materials and data. For example, in my study I could see from the matrix coding query that being in a new place was often mentioned during the free-telling phase. Based on my theoretical framework, this indicated that it was an important thing for the participant.
Step 5 : Drafting the first paper and writing up
Memos written during the analysis (along with the process recorded in them) can be used when drafting the first paper. You can do your writing in NVivo—I used my analytical memo for this. I finished my paper with Word because NVivo doesn't have 'cite while you write' features like my referencing program.
See, you can just kick back and relax. NVivo is doing all the work for you!
Well, maybe not. You still need to think, code and write but NVivo can make it clearer and more visible, step by step.
What sort of path have you followed?
Watch Tuija's webinar for more detail
You can watch a recording of Tuija's eSeminar where she demonstrates how NVivo helped her to plan, analyze and report on her study: