 |
 |
Esa Pikkarainen wrote: > Now here has been interesting questions and talk about basic > question. It seems that there is an essential difference between > qualitavite analysis (QA) and content analysis (CA). They both can be > computer aided but still their diffrence remains. In CA you are > describing the text material empirically and in QA you are > interpreting it. I don't think you can put it that simple. Interpretation is always necessary in both approaches. And the times when CA was used to describe only (e.g. Berelson's definition in 1952) is over. > In previous the statistical features among and > betveen accounted occurrences are important - like in any empirical > study. In latter the interpretation can be quite independent from the > empirico-statistical features of the text material. Did you mean this, > Harald? Yes to a certain extend. I think the main different is that in CA you have hypotheses to test, and you take a selective view on the material, where in QDA you probably want to find hypotheses and want to interpret all of the text. CA is selective rsp reductive, QDA is not. In CA you want to find differences, in QDA you look for regularities. Well, this list does not claim to be complete, and there are some book that cover the topic, but the problem is that both approaches are often regarded as contrary (the old quarrel between quantitative and qualitative). > The most important problem for me here is that after all you cannot > (usually) make that straight through division: In QA you have to back > your interpretations by statistical features and in CA you have to > have a pre-interpretation about your material. Yes, that's is true. I took more simplified terms to point to the beef. > Of course the order how you are proceeding can differ and thesee > approaches feel > different. But is this division fruitful? Is it possible to make a > non interpretative analysis of the _content_ of the text? No. The decision what to count and how to count are qualitative. > Or non statistical interpretation? The QDA people will tell you they don't want statistical interpretation. Or have it in a simple joke: 10 people are attending a lesson. Two people are asked on the number of visitors. quantitative: it were 10 people qualitative: too few The difference is the Gehalt of such statements. Which one contains more information? > And what is the role of the computer here between me as an interpreter > and the text as interpreted? > How programs differ in this direction? Questions, questions... You can write a book on that, Sage will publish it immediately if you have it ready. At least Simon Ross told me last year on a conference. It is like living on different planets: the CA and the QDA planets. There are only very few scientists who know the other planet. > just a thought... Not only one thought, also very interesting topics you arise, worth a discussion. Harald %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%