I mentioned in my last post that I was knee deep in reading for my dissertation proposal. I’m still knee deep, but the ideas are starting to become a little better formed now. One thing I can confidently say is that all of the questions I’m interested in researching involve understanding the way that language is used in the context of business to business selling – something I’ve been involved with for most of my professional career. This means that I’ll almost certainly be using a qualitative methodology, one of the many variants of discourse analysis, to undertake my empirical research into the topic. This is not the conclusion that I wanted to come to, based on my past experience of how much time qualitative projects consume!
As part of my undergraduate degree, I undertook two full-scale psychology projects. One of these was an experiment, which gratifyingly gave me and my partner in crime two significant and one insignificant result after the stats had been crunched. As far as I was concerned, this kind of ‘split decision’ was brilliant as you can learn just as much, if not more, from experimental results that don’t conform to your expectations. It also makes producing an interesting write-up straightforward, particularly if you follow the standard psychological report-writing conventions. In the grand scheme of things, it really didn’t take that much effort to produce something that I was happy with.
The other project was a qualitative one requiring the use of critical discourse analysis. For all kinds of reasons (transcribing interview data is one – I figured out that I can only manage this at a rate of about 16 hours of effort for 45 minutes of data), this type of research takes far, far longer to do well. Probably the most challenging part of a qualitative project is interpreting the data. With experimental research, there’s usually only one (correct!) way you can crunch the data through a stats software package. With data from a discourse analysis project, there are endless ways to analyse how the participants are using language to take positions to either justify themselves or to blame others. Analysis is hard work, and crucially, you also have to consider the way that your own experiences and interests influence your conclusions. The researcher is explicitly present in the data, rather than a factor that has been assumed to be ‘controlled out’ through good experimental design.
If you’ve had experience of quantitative and qualitative research, which do you prefer – and is it the same approach that you feel gives you the greatest insight? For my own part, I certainly prefer quantitative research from the point of view of simplicity, but the insight that rich, qualitative data gives is often worth the effort.