Dear John,
just a few thougths I had after reading your mails.
Merging - at least for the people who work more often with QDA packages - if most often used
with reference to merging two qualitative projects. Thus at a first glance it is a bit
confusing to use the term mergin when talking about combining quantitative and qualitative
data in one project.
The kind of quantitative data you can use are descriptive values, e.g. like gender, age,
education, income, profession (there are other variables of that kind apart from
demographics, I know - but these are easy examples). In other words, you must be able to put
things into categories, thus you cannot use continues variables - at least if you are
looking at the exporting and importing functions of tables in Nud.ist and winMAX. One other
package that can deal with such data in similar ways is The Ethnograph.
Within Atlas.ti you cannot import quantiative information but - if you code in a particular
way - you can create a continues variable from your coding scheme (in addition to ordinal
type variables) in form of a SPSS syntax.
However, one also has to raise a word of caution any time when the use of such exporting
functions in QDA packages is discussed - since coding schemes in most cases do not fulfill
the criteria that are needed to run a statistical analysis.
I am not opposing the possibility of being able to combine qualitative and quantiative data
- I think it is very helpful, it strenghtens and facilitates the analysis, but I think one
should be aware of what one can and what one cannot do and what kind of quantitative
information can be used in combination with QDA packages.
John, I am glad to hear that nowadays a more wholistic approach is used to approach the
problems of development, e.g. the consideration of the various components : PA, RH, SI, N,
and I am also glad to hear that decisions are not only made based on numbers (since it is
people after all who contribute to the process of development) and qualitative data are
collected as well. I am certain that there are appropriate ways for combining qualitative
and quantiative material also in your area of work - but there are also limits.
You didn't mention what kinds of statistical analysis processes are usually used. If it is a
lot of econometrics (regression analysis, etc.) - like I know is used a lot here at my
university by people who work in the area of development helping poor
localities - this is something beyond what you can combine with qualitative data in QDA
packages.
Maybe this is not new to you and maybe it was not your intention anyway - these are just the
thoughts that occured to me when reading your message.
kind regards,
Susanne
////////////////////////////////////////////////
Susanne Friese
Qualitative Research & Consulting
http://www.quarc.de
Alte Dorfstr. 71A
70599 Stuttgart, Germany
Lehrstuhl für Haushalts-und Konsumökonomik
Universität Hohenheim (530/1)
70593 Stuttgart, Germany
Tel: +49 (0)711 45 00 21
Fax: +49 (0)711 459 2870
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|