May 4, 2010

So, in my last post I talked about coding and no I don’t mean coding as in scripting or writing programming code (though it is just as tedious), I’m talking about coding qualitative data.

For those non-anthropologist among my readers, what this amounts to is tagging subsections out of large amounts of textual data with a taxonomy you develop along the way. This taxonomy is based on the themes that emerge as you sift through all of the data. This, if done meticulously (as of course it is), can take quite a bit of time and requires a bit of resifting of the data to capture and tag everything correctly.

Speaking as not only an anthropologist, but also an information architect, I actually really like this part! It’s to anthropology what sifting through the dirt at a dig is to an archaeologist. Each time something unique emerges from the sand/data there is a little *squee* that occurs. Each bit is just as precious as the last no matter how big or small. We collect each and every one of these unique themes and catalog them for reference later.

What happens after all the sifting has been completed is actually very interesting for dataphiles like me. You can then select any tag you used and see all of the text among all of your data sources that matches that tag. Or, you can line up all your tags as see which tags show up most often. Consider doing that to your blog, or Delicious Bookmarks.

You can tell a lot about an avid Delicious user, or at least what interests them on the web, by looking at their tags as a whole and then at their most (and even least) used ones.

Tagging data and looking at the results of that process is very similar.

Basically, it is in a sense both organizing your qualitative data and quantifying it. As you can imagine, it’s very revealing!

How else do you quantify qualitative data? Well you make a survey of course! Why would you do that? Well it helps triangulate your data. In other words, it helps you generalize your qualitative findings against a broader sample in quantifiable terms.

In order to do this properly we take those themes that emerged through our coding or tagging process then use that data to form appropriate questions and answers to be asked of the same audience but on a larger scale. Thus, we use both qualitative and quantitative methods in the research process giving it both depth and validity.

Could you do one without the other? Well, of course you can and many anthropologists tend to prefer one over the other. I’ll not step into the qual vs quant debate here, except to say that it exists. What I will say is that the goal of the anthropologist is to approach research and the research questions from a holistic perspective. Using both methods, as I am doing here, helps to do this.

Why am I going into all of this detail? Well, if you’ve reached this point you are in for a special announcement! My survey has been approved by my committee and is now in the hands of the IRB for review. Once they approve it, I’ll be releasing it to all Fedora contributors to participate in.

It is about 28 questions (depending on how you answer) and will take most people less than 10 minutes to fill out. I say most, because I do give you places for long text descriptions if you so desire, so those people who have a lot to contribute may take a little longer. From my perspective, the more data the better!

I have lofty hopes of getting about 100 of you excited enough to participate. So, once I release it, I will need the help of all of the Fedora contributors that read this to pass it along and to tell those people you pass it along to, to pass it along as well!

An interesting side note to the survey is that it was created with the free and open source LimeSurvey! So, by taking it you’re not only helping Fedora, but you’re also supporting a pretty cool FLOSS project!

During the process of creating the survey I’ve become quite the LimeSurvey user. So, if anyone needs help with it, please feel free to ping me. When I’m done with this project I may even find a way to lend some of my user interface design and usability skills to the LimeSurvey project, if they want them. I have a few ideas that could make it a bit easier to use after having been a user myself.

I honestly cannot wait to get the survey out and start getting the data back. This part of the process is always so exciting!