Bachelor Thesis Log 2

July 27th, 2010

Today's post is about empiricism and data collection. I would like to show you which considerations I have made regarding my data sources so far and what kind of data I would like to collect.

Bachelor Theses and Data Collection

A bachelor thesis in computer science typically deals with a problem that can be theoretically analyzed or even implemented. Somewhat rarer are topics that actually require field research, which can be done through surveys or similar methods. My topic belongs to the latter.

Data Sources

For my research question(s) on the versatility of teachlets, there is only a small set of data sources that may be able to help me.

• First, there is the teachlet archive, which preserves a large number of existing teachlets from past teachlet workshops. Each ideally contains the complete teachlet choreography, all slides, systems, and materials, and with any luck, experience from previous runs. For the question of what constitutes teachlets at their core, it is certainly useful to work with the definition not only from the side of theory, but also to examine existing teachlets for commonalities and general structures.
• Further, I have hopes of interviewing some experienced teachlet presenters. In the environment of our university, there are perhaps half a dozen people who have used teachlets repeatedly and reflectively. I would like to get hold of these people and interview them on the topic, in the hope that actionable insights can be gleaned from the interviews.
• Lastly, there is first-hand observation. There is a teachlet that is to be carried out here at the university at the end of August with over 100 people, and I will be observing it – “track” it, as the teachlet connoisseur would say; this refers to the minute-by-minute logging of which section took how long, especially in order to be able to compare the actual course of events with the plan after the fact. It is particularly interesting to see how the teachlet concept works with such a large number of participants.

Obstacles

Data collection is already limited by the fact that a teachlet cannot simply be carried out “on a trial basis” because a certain number of participants is required. This is the point at which the trial implementation of a teachlet following an OOPM lecture failed last semester – no one wanted to stay longer for it, which is of course understandable. So I have to be very careful to use all available sources, because it's a very limited number.

Another “problem” is that the data I collect is largely qualitative rather than quantitative. There won't be too much to compare and put into pretty diagrams, but there will be all the more different statements and evaluations from which a coherent overall picture has to be extracted. This will be a tough job, especially for a quantitative scientist like me.

Organizational Aspects

Since last week I have a second supervisor. Prof. Oberquelle has agreed to do the job. Now that this position is filled, nothing stands in the way of officially registering the thesis. I will discuss this with Axel.

Conclusion

Hopefully my presentation of the data collection concepts is reasonably understandable. What do you guys think of it? Is it written in a readable way, or should I go into more detail about typical teachlet terms when I use them?

Next week I will report about the brainstorming meeting with Axel and Christian, where hopefully some basic knowledge about the definition of teachlets can already be gained. The meeting will take place on Thursday, next Tuesday around this time you can see the results here.