Think I just finalized my “vision” for what my database will look like for my research here. It helps me to visualize it:

School here starts September 7^th, which would worry me, but none of the kids really show up until the 2^nd or more likely 3^rd week of September, so I still have time to work on my database before all the little pieces of data starts flowing in. To give you an idea of what I mean:

I’m monitoring 29 schools (it was 37, but we decided to cut all the ones in Port au Prince, as they’re so different from most of our schools that it would throw off all our results—tuition is higher, attendance is higher, teachers are more qualified, travel time from home to school is shorter—urban and rural schools are just completely different animals here). Each school will be (for the first time ever) sharing its daily attendance data with us. Schools directors will add up the absences in each grade, and each month we’ll collect the forms and give them new ones for the upcoming month. So, using quick math, with  7 grades in a primary school and 7 grades in a secondary school, I’ll be collecting 203 attendance forms (7 x 29) each month. Each school director will also be tracking his/her daily local weather, broken down into simplistic categories like sunny, cloudy, rainy, stormy, hurricany. If you can’t tell I haven’t really figured out the wording for them to describe the weather yet.

If I were just putting this all into an excel spreadsheet it would be more straight-forward, but since it’s going into a relational database, every day can be seen as an individual unit of information. So October 3^rd, in the School X’s 2^nd grade, is one piece of data. Thinking about it that way, we just ballooned from 203 attendance forms to 6,090 pieces of data (203 attendance forms x 30 days on each form).

That actually isn’t as scary as it sounds, because I only have to build one template that I can enter this stuff into. One template I can use for each school’s attendance, broken down by month, week, day, school grade.

What DOES scare me is that if I don’t build that one template correctly, in 5 months I’ll be screwed. It’s cliché, but it’s like building a house. If you can build a solid foundation, you can always change things later on. But you don’t start with something solid? Forget about it. So I’ve got a few weeks to do some test runs and make sure what I think will work will actually work.

I’ll also be following 100 teachers chosen randomly at different schools, following their individual classes performance and their students’ attendance (instead of having it added in to all the other 2^nd grade classes, for example, keeping particular teachers attendance data separate for closer examination).

And I’ll be doing all the stuff I just talked about with 12 schools not in our program, so we have a decent sized control to measure all this stuff against.

I’m a little nervous. Not about Haiti, or visiting schools, or talking to teachers, or conducting surveys. I’m nervous about this monstrosity of a database. It’s the first one I’ve ever built from scratch. If there are inherent flaws in its foundation it could make life very unpleasant for me down the line.  I need to get back to work.