[MUSIC] Hi there. Thanks for joining again. In previous video, I introduced data coding for qualitative data and explained why it's important. This video, we'll discuss coding in more detail and I will provide you with some tips. One thing to keep in mind when coding is that you make sure that the meaning of the code is always clear to you, and especially stays clear during the project. When a meaning of a code is not clear, there's a chance that you end up with quotations that are wrongly coded. To remember the meaning of codes, you can make so-called memos in your software program. If you are analyzing hundred of pages and creating many, many codes, it's good to record your thoughts during the process, and you can record your thoughts with memos. A memo is basically a comment field to make notes for yourself. You can link a memo to a piece of text to a code or to a quotation. And another thing to keep in mind while coding is to keep context in your coding. Now take a look at this interview fragment. If you take, for example, the code participation in projects, and your code or label only the word no. Then afterwards, you don't know what this no refers to. So when analyzing the code, participation in project, you only get to see the word no. In this case, you also have to include the question of the interviewer. So what about the number of codes? Student who start coding for the first time often ask the question how many codes are needed to analyze my data? Well fortunately, there is no rule for this. Although, it will depend a bit on the amount of data you have. Of course, the more data, the more codes you may need. On the other hand, you have to decide for yourself how many codes you can handle. If you have the feeling you are losing over your codes, you might want to change your list of codes. You could recode parts of your data, or maybe you can reduce your codes by connecting quotations of a certain codes to another existing code instead. Or you could cluster individual codes into a group of codes. I think the most important thing to keep in mind, that your code list is not fixed from the start. So during the coding process, the code list is like a dynamic entity. It might expand, it might shrink, or maybe it remains the same. A good technique for structuring your code is so called axial coding. Axial coding means that you use theory to bring order into your codes. You start to look for patterns and codes and try to make categories of codes. I will give you an example of axial coding. In the literature, you may have found that social capital has to do with the following three indicators. One, number of people you know in the neighborhoods. Two, frequency of contact you have with neighbors. And the last one, type of support you receive from neighbors. Now you can structure your own codes based on these indicators of social capital you just found in the literature. You can use three codes that relate to the indicators, and then cluster these codes to run group code called social capital. Schemes like this where you create some kind of ordering among codes, are also known as taxonomic code schemes. It helps you to get a grip on the often large number of codes in research project. You're clustering of codes, the interconnection between codes becomes visible, and this facilitates the further analysis of the data. I will now give you an example of an imaginary taxonomic code scheme with different levels of codes and group codes related to the literature. The main code in this example is motivation to participate. This code consists of three different group codes, political, personal, and financial. We divided the group codes personal into family connections and serving the public good. Now the last distinction we made is that family connections can be of positive and of negative influence. And with the clustering of codes and code groups, we made sure that these theoretical ideas on the motivation to participation are now related to your own code list. Now, coding seems like a lot of work, and actually in fact, it is. Interview transcripts normally consist of ten pages or even more. And for every text part or quotation, you have to decide if its code worthy or not. And if it is, which code will you need to use? Or do you have to create a new code maybe. And how can you cluster your codes into a group code that make sense? All these questions will arise during the process. But you will find that more interviews your code the faster it will go because you will get more and more experienced during the coding process. So yes, coding is time consuming but it's worth the effort. Because when everything is coded in a structured way, you can analyze your data in really efficient way. In the next video, we'll talk about the analyzing process. But for now, thank you again for your attention and we will meet again in the next video. Bye. [MUSIC]
