Initial Coding of Student Interviews

5 Coding and the Building of Conceptual Categories

Before discussing the relationship of the main analytical categories and their properties, it is necessary to show where they came from. The purpose of this chapter is to give an account of the early phases of the data analysis, which began as soon as the first data was collected and processed, and of the emergence of the main conceptual categories leading to the construction of the theoretical model of student enculturation into written academic discourse.

interviews, when it was no longer possible to transcribe the interviews due to lack of time and the increasing length of each interview, open coding was done by listening to the recorded interviews and making notes on them.) At this stage the sensitising concepts largely influenced what was discussed in the interviews and thus much of the early coding is related to these. However, even at this very early stage, other, totally new concepts could also emerge, related to what individual students were experiencing. The example extracts from the first round of interviews which follow will show both of these situations.

English at school was not good having a bad teacher

learning English at home

doing nothing in class teacher was just talking

teacher not knowing if someone was missing even in a small class

didn’t have problems with tests other classmates not on the same level

not a problem for him that classes were not useful

learned English from watching television

Extract 1: A1, Int. 1, p. 1

FP: September 26th, Interview 1, Student A1.

I’d like to ask you a little bit more about your background first of all. First of all your school – can you tell me something about your school?

A1: Well, at school I learnt Mathematics, at secondary school. Well English - well that wasn’t quite good there, because I had a bad teacher. Mainly the past 2 years. So I really learnt at home in English.

FP: You never felt that you learnt much in the English classes?

A1: Well in classes we done nothing so, the teacher just said, said, said. He didn’t even observe if anyone was missing. We were only 7 in one class and he didn’t know if anyone is missing.

So, that was a little bit of a problem. I didn’t have problems if we wrote tests or things like that but the other classmates had because they weren’t on the same level as I was. So it wasn’t a problem for me even if we weren’t learning but for others it was.

FP: So you studied yourself at home? To keep /?/

A1: Well actually I watched a lot of cartoons and movies and this way I learned English.

Figure 4. Example of line by line coding of interview data (FP are the initials of the researcher in all interview data)

In the first extract above (Figure 4), from the first round of interviews, the initial questions are about the student’s English learning in school. Not surprisingly, the initial coding is closely related to this sensitising concept. At this first stage of coding the main concern is to stay close to the data and to keep the codes simple and precise, while at the same time being open to the meanings within the data rather than imposing preconceived ideas on them. Through a close reading of the data the researcher tries to define what is happening and it is here that the process of grappling with meaning begins. As coding proceeds, some codes recur frequently and others hardly at all. By coding for action in particular, the researcher sees that some actions seem to be central to certain areas of experience – they begin to suggest that they have an important part to play in a particular process. In Extract 1, learning English from watching television was an action that actually formed a part of the English learning experience of several of the students, and therefore, it assumed an importance in explaining how some students who failed to learn in the school class, often because they found the class unsatisfactory in certain respects, but who were still interested in learning English, used the easiest means that was available to them to get exposure to the language. This subcategory of learning English from the television emerged from the data by comparison with other cases, thus demonstrating the importance of using constant comparative methods (Glaser &

Strauss, 1967).

The second extract (see Figure 5 below), shows a situation where very early on in the first interview a new concept emerged, which was related to the experience of the student when she first entered the university. Again, through constant comparison this concept turned out to be a significant one, describing a feeling experienced by several of the students. In answer to my question about her English learning in school, “And what about the English classes there – can you tell me something about those?” (C6, Int. 1, p. 1), the student, Csenge, gives a tangential and cursory answer at the beginning of a long turn in the discussion where

she talks about something that is worrying her at the moment in her studies: the feeling that her English is not good enough compared to the students around her. Several of the initial line by line codes relate to this feeling, from line 3 onwards.

started learning English in 9^th grade length of time spent learning English feeling that her English is not good enough ; working hard to keep up (with vocab learning?)

realizing that others know much more

having difficulty keeping up trying to catch up to be at the same level

feeling of not knowing enough compared to others

never lived in an English-speaking country

not having a chance to go hoping to go in the future

Extract 2: C6, Int. 1, p. 1

Yeah. I started to learn English when I was in 9^th so now I’m learning English for more than 4 years. But I think that maybe it’s not enough here – so now I have to do a very hard work because I think I’m – with all the words; it’s very difficult when for example at a Language Practice class when I’m with Teacher X and I just realize that the others sometimes know so much more than I do and it’s just, it’s very difficult to keep up with others and I think that now it’s the most difficult for me to just catch up and try to be at the same level. Because I think there are some students who are learning English for I think 14 years and me just 4 years and there are a lot of people who lived abroad and I have never been in an English-speaking area where, you know, everyday life I had to speak English and it’s not, it’s – I hadn’t got a chance. But maybe I will in the future – I do hope.

Figure 5. Example of line by line coding of interview data

Probably this feeling is something which the majority of students have felt – certainly I remember as an undergraduate quickly recognising that, having come from an environment where I was considered a high achiever in English, I was now somewhere in the middle ranks.

But what is important here is the intensity of the feeling and the effect it has on the student.

At several points in the first interview, Csenge reveals her anxiety about her ability to cope in

this new environment and expresses her doubts about having made the right choice: “I really hope it will turn out that it was a good choice” (C6, Int. 1, p. 1). By the second interview just over a month later she is seriously considering leaving the university and has already taken steps towards that end. This action is not solely because she feels that her English is not good enough, but it was this initial code which began an exploration of meaning in the data.

Through comparison with other students, and with the same student in subsequent interviews, this open code, feeling that my English is not good enough, eventually became part of a more complex category describing the sensation of culture shock experienced by some students when they first enter university. This code became one of the properties for this category and by exploring it further through analytic memo writing and further comparisons in the data, a more detailed picture of what culture shock means in this situation and what its consequences might be could be obtained.

Thus it is with thorough and rapid open coding of the early data that the analytic work of the grounded theorist begins. Initial codes are provisional and flexible – they can be reworked and reworded to better capture the ideas in the data. Those codes which best fit what seems to be happening in the data and which have comparative power are the ones which are likely to be incorporated in further analysis. Other codes which are not useful will be discarded or subsumed in more useful codes. It is through open coding and comparing of codes that the researcher begins to get an analytical grasp of what is going on in the data and also an indication of gaps that need to be filled by further data collection and comparative analysis. These gaps are questions that are raised about properties and the processes they are part of. In this study one of those questions was about how the eventual decision to leave the university was made by Csenge, and this led to the description of the exact moment when she realised that she was in the wrong place, a kind of epiphany for her. Once the work of open

coding has established some clear analytical directions of this type to explore, the next level of coding can begin.