What I wish I knew when starting NVivo

Like many of you, I found learning NVivo tough. I started using it last year, and since then, I’ve gone on to run the research community for NVivo at the University of Melbourne.

I get to hear from a lot of researchers about what it’s like to use NVivo. Hearing all their stories and getting to teach the tool is a great privilege. However, with all this experience, I don’t feel like an expert. As I’m sure you all know, trying to figure out what NVivo is all about can be difficult. Just because I’m standing at the front of the class teaching it doesn’t mean I haven’t struggled with NVivo myself. Indeed, I still do today! This post is about failure, and why it’s okay. So, as a way of demonstrating that, I thought I’d cover the four things that I wish I knew when I started NVivo, and how it didn’t destroy my research when I got them wrong.

Dragging and Coding

Yeah this one hurts. When I started using NVivo for my research, I thought that I had nailed the basics. All I had to do to code was right click and select the code button, right?

Wrong.

Had I been bold enough to watch their (very) helpful tutorial videos, I would’ve known straight away that there is a MUCH easier way to code your files. Simply select the text or region and drag it over to the code it. Easy. I did find this process VERY difficult on a laptop’s track-pad, so I made a sensible investment in a wireless mouse. So many hours wasted, but hopefully you won’t ever find yourself in my situation.

Iterating your Coding

After I had finished an initial draft of my literature review and methodology, I thought I was set to start coding the themes that had emerged from my research. Not only that, I thought these themes would be the ONLY interesting themes and that they would be easily discernible and distinct from each other.

So, after a few hours of coding, I thought I was finished. Yep. It was as easy as that.

It dawned on me pretty quickly after a few meetings with my supervisor that I had a lot more work to do. When I learnt this, I was pretty bummed out. Surely I had done something wrong if I had to go back into my code and do it again? Wrong.

Reiterating your coding is one of the most useful things I did as part of my research. Reiteration is research. It gives you space to refine and reflect on your research and the relevance of it to your research question. I was able to go back into my codes, reorganise them, and discover new themes and relationships that had never crossed my mind during the early stages of analysis.

NCapture

If you use Zotero then you’ll like the look of NCapture. NCapture is a cute little extension that allows you to capture web pages to import into NVivo. This doesn’t sound like a whole lot of functionality, but I can help out a lot.

I was having to save my newspaper articles manually by grabbing them from the web (scraping wasn’t working). This took a lot of time, and I wasn’t happy with the end result a lot of the time. Text would be missing, and the pages would be poorly formatted.

And along came NCapture. With a simple click, I was able to capture all the information I needed for NVivo. Easy.

File and Case Classifications

What is the difference between a file and a case? Files are essentially the data that you put in NVivo – interviews, newspaper articles, images – that sort of thing. Cases are units of observation – people, organisations, places – that are in your data and that have characteristics that might be really important for your research question.

When I started using NVivo, I didn’t really have much of an idea. I was working with newspaper articles, and so I was using file classifications.

To give a clearer example, I was looking at articles written by Malcolm Farr. Those articles were files, which I classified as news articles. Malcolm Farr is a case, a person with specific attributes (for example age).

Turns out that I had been a lot more interested in the author than the information about the articles and I had multiple articles by the same author.

What I should have done? Case classifications. Going back into my data, I coded the articles as cases, and found it a lot easier to work with the data to help answer my research question.

It’s Okay to Fail

Each time I learnt that I’d made a mistake in how I’d used NVivo in my research – from spending too long getting my files and coding them, to mixing up file and case classifications, to taking more time with my coding – I thought that my research was all the worse for it. But it wasn’t, and I got there in the end.

NVivo makes research easier, but it doesn’t do your research for you.

Because of that, you have more space to fail. Every time you make a mistake, it’s an opportunity to learn something new about NVivo, or about the research process. My failures in NVivo made me a better researcher.

Your failures will too.