Data, Data, Data

I’m now several weeks into the data analysis portion of the ride.  Choosing to focus on a variety of factors that could have an impact on communication has meant choosing to analyze several different types of data.  While at times the different programs and kinds of data felt overwhelming, overall I think I’ve been happier to be able to mix it up a little from day to day.

One of my first data analysis tasks was to write up every bit of potentially useful information I could remember from the trip that didn’t fit neatly into an interview protocol.  Everything from the agenda of any meetings we attended to who made small talk and about what.  Preserving this qualitative info will help ensure that the team members who were not there this summer can have as comprehensive idea as possible of what happened.

Another data project was putting together a family tree based on who respondents identified as their relatives.  This endeavor proved to be a much harder than it sounds, in part because of the lack of any free, user-friendly, convertible, and sharable genealogy program out there, and because of the cultural differences we encountered in what I had thought would be a very straight-forward question.  Often asking, “Do you have any family in Chaguite?” elicited a name of a cousin that turned out to be a 5th cousin, or a sister that turned out to be a sister-of-the-Church.  Nonetheless, in a region where family ties clearly affect anything from housing to schooling, and certainly communication, a family tree was a necessity.  In the end, I think the final trees show several connections that the team had not previously expected, which should certainly inform our plans and actions.

After coding and formatting all of the interview data into SPSS, I finally ran some tests.  Seeing significant results arise, for both some relationships I expected to see correlated and some surprises, was an exciting feeling.  Especially when strong trends emerged despite small sample size and interview-team errors.  Seeing these significant results both encourages me that we’re on the right track, and inspires me to want to know more, to ask improved questions, and to discover ways to use this knowledge to benefit the community.

The social network dataset is also yielding significant results.  Several key actors have arisen from the data, individuals who community members identified as potential project organizers working for the good of the community.  The network shows others as good communicators, maybe who link otherwise separated parts of the network or link otherwise unconnected important actors.  These data provide empirical support for any hunches of who seems to be a trusted leader, or who is highly involved in community events.

Now that the data have been analyzed and most conclusions are drawn, the next task is writing.  I won’t lie, I’m a bit (ok, a lot) daunted at the thought of how much there is to tell, especially for a non-MANOS audience, where the background information of what I’m doing and why seems endless, not to mention the multiple facets of the current data alone.  Fortunately, my excitement for the project hasn’t diminished, so I know that wanting to share this information will drive me over my writing struggles to a final product.