Sunday, 20 May 2007

Qualitantative

I've been marking the exams from the second year research methods course I'm teaching. The students had to answer three questions out of 12 to pass the exam. Here are some rough statistics: 75 percent of the 27 exams I got for this course contained at least 2 same questions. They only differed for the third one. But, 50 percent of all exams had answered the same set of questions. Maybe in a different order but I've been reading 14 times the same answers. 2-3 questions were very popular and one of them almost were answered in 90 percent of all exams. Only one exam had answered a question which no one else had answered. Overall, from the 12 original questions only seven were answered but with a highly unequal distribution. And apart from that, the answers mainly used the same sources and arguments. I could now probably tell you the books which will be used for referencing before I even read the exam answer.
This made me thinking how much difference between qualitative and quantitative research there really is. Each question had a chance of 1/12 to be answered in the first instance and 1/4 in the second (as there were three to be answered). Taking this to 27 exams, there were roughly 1000 possible combinations (to be honest, I'm not quite sure about the number because I didn't really listen when I was taught probability calculations...). Whatsoever, this never materialised. At the end, it were more or less all the same. The thing which struck me was that most students believe qualitative research (roughly, everything that is not numbers, so to speak, interviews, observations, documents and the like) is the best method to research what humans do and why they do it. With interviews it will be possible to account for the diversity of human behaviour. Hm. When I look at the exams the diversity is rather limited. Humans are not so different as it is assumed. I could do 10 interviews and would probably get 10 different answers, yes. But I guess most of it is just a variation. Not more. I could do the same with statistics. I will get a variation with some main answers that always come up. To analyse diversity I would just have to concentrate on the margins and see who answered there. To cut this thing short now, there is apparently less diversity in human behaviour as it is believed. I dont know really, but in the end it's all the same, isn't it? This doesn't mean that there is no real use for qualitative. There isn't real use for quantitative and statistics either. None of both is better or worse than the other. They just don't fully explain anything. In the end, we all buy at Tesco, have a Nokia, and use toilet paper. What we use these for might be different. But the outcome is the same. I don't starve, I can call, and I can whipe up something.
Well, I kind of drifted away again and I intended to write more about individuality when I started on this post. But let's get rid of hokus-pokus interviews and non-sense statistics.

No comments: