So it becomes database type. We are still experimenting with interface design however while we are going through it all, we are using three important (yet equally frustrating tools).
First, I took the data from the survey results, put it into Excel and cleaned it up. By clean, it’s the process that we use to remove any of the unnecessary information, or incomplete results. This process is often easy, but when you are handling large datasets, it can become one of your worst nightmares. Especially, if you’re not careful and take care of versions. Remember always keepn the original file in its original state. Also, each time you delete a huge chunk of data, for example, ALL of the incomplete responses, save it and call it with a unique name such as results_post_inc_rsp (inc = incomplete, rsp = responses).
Then, once the data is clean, we take it into the wonderful SPSS, statistical analysis program. Depending on your experience with statistics, and programs like this, this experience is either one of routine or one that is traumatic. Getting your head around can be a daunting task, I have been there, but in some cases, a lot of what I used to do in Excel, I now do in SPSS. Not only does it save time with finicky equations that you write in excel, it keeps things tidy (at least for me, this work process works!).
Once we got the game database sorted, it was time to import it into Parse.
Now, we had something more adequate to test with, we began iterative testing. Link a few things and hey presto! You’re set.