After receiving the results of the last people who had installed the AWARE framework, I added them to the data collection to take another look at the resulting graphs. Unfortunately the AWARE database file I got from some people wasn’t working, it was corrupt. I googled the problem myself first and couldn’t find any solution, I tried opening the files with several database managers but (obviously) none were able to open the file. So while studying and taking yet another break, I decided to send an e-mail towards the developers of the AWARE framework asking if they were aware of this problem and if they maybe had a fix. Turns out they did know and forwarded me this link: http://random.kakaopor.hu/how-to-repair-an-sqlite-database, which was quite an easy fix but somehow this never showed up in my search results. Guess after 5 years of Googling everything, I’m still bad at it.
So having fixed all the database files, I now have data of 14 students. These consist of 13 males and 1 female student. 9 are computer science students, of which 1 Phd student. As we can see in the graphs below, the most used apps are still very similar to our intermediate results. The chat app of Facebook by far generates the most notifications, which results in a lot of micro-usage sessions and a lot of longer sessions as well.
We can clearly see most of the apps are still communication apps, and a lot of micro-sessions are occurring. So I think we are still justified in our new direction of enhancing notifications to allow for handling them quicker, which hopefully could reduce the amount of micro-sessions.
The graph of session duration still looks more or less the same as the one from the intermediate results. It is obvious most user sessions are still 0 – 20 seconds.