Thematic analysis was employed for the present study. Noticeably, analysis of narratives and collection of further narratives were carried out simultaneously. Put it differently, whenever a narrative was returned, it was read thoroughly for initial coding as well as for a decision on how much data should be further collected.
According to Barkhuizen et al. (2014), there are two approaches to analyzing narratives – that is, thematic analysis and discourse analysis. Polkinghorne (1995) postulates that thematic analysis “is largely a matter of categorization and classification, in which particular instances of phenomena are linked to more general concepts” (p.74, as cited in Barkhuizen et al., 2014). These authors point out that one risk of content analysis is that the researcher may fail to make full use of the collected data but focus on the pre-determined themes, which is likely to limit the value of the research. However, Duff (2008) argues that although the themes are data driven, it is still useful to decide the categories before analyzing the data as long as the categories are in accordance with the topic of the study as well as the research questions. In order to make good use of the priori codes as well as reduce the risk, Gao (2010) suggest that apart from adopting the themes determined in advance he makes an effort to find out the subthemes emerged from the data (p.76). Further, Barkhuizen et al.
(2014) posit that thematic analysis is suitable for multiple case studies as it “opens up the possibility of comparing the narratives in a data set, of establishing shared themes, as well as highlighting individual difference” (p.77)
93
While thematic analysis focuses on what participants say about their matters, discourse analysis focuses on how participants say (Barkhuizen et al., 2014, p.75).
When adopting discourse analysis approach, the researcher can use one of the following three strategies: metaphor analysis, narrative structure analysis, and interaction analysis. With the first strategy, the interpretations are based on the metaphors that participants use to talk about the matters. Importantly, metaphor analysis has a certain relationship with participants’ cultural background (Oxford, 2002, as cited in Barkhuizen et al., 2014, p.82). The second strategy, narrative structure, pays attention to structural features of narratives, for example, how participants construct the openings and closing of the narratives, or how different aspects described in the narratives are linked together. In so doing, the researcher is likely to interpret about participants’ identity as well as the way they use or learn a language (Coffey, 2010; Menezes, 2008, as cited in Barkhuizen et al., 2014, p.82).
The last strategy, narrative in interaction, is used for short narratives as natural spoken interaction. The research using this strategy follows three-step analytical procedure based on three positioning levels: “(1) how the characters in the story are positioned in relation to each other, (2) how the speakers position themselves in relation to each other, and (3) how speakers construct themselves and others in terms of teller roles and dominant discourses or master narratives” (Bamberg and Georgakopoulou, 2008, p.385, as cited in Barkhuizen et al., 2014, p.84). This implies the role of the social power in the analysis of oral interaction between different stakeholders.
Considering the focuses of different strategies discussed above, the study adopted thematic analysis. The primary aim of the study is to explore what teachers learn through their involvement in course design and the factors that affect their learning.
It, therefore, has nothing to do with how cultural background affects the use of language, how a person learns a language, or how social powers interfere their views and sharing, which is the focuses of discourse analysis approach.
3.2.5.2. Data analysis procedures
Following Duff’s (2008) proposal of an overall process of conducting a qualitative
94
research, data analysis and interpretation of the current study followed a number of steps: (1) transcribing the oral narratives; (2) coding (both written and oral narratives); (3) inducing themes (pre-determined and emerged themes); (4) member- checking. As data collection involved both written and oral narratives, transcribing was still a necessary step to facilitate the coding process. All the interviews were transcribed in verbatim; however, as the study did not follow discourse analysis approach the exact time of pauses or pitches were not of critical importance.
Importantly, the data analysis and interpretation were guided by activity theory.
Particularly, course design was a collective activity driven by a collective motive, which was embedded in the object of the activity. In order to achieve this collective motive, the involved teachers worked with a variety of available materials and/ or attended a number of workshops and trainings, which were considered as the mediation within the activity. Moreover, in such a collective activity, the involved had to fulfill different tasks as well as obey the requirements of their leader, their Faculty Board, and the Governing Board; in other words, the labor force involved in course design was divided and operated under certain rules. Moreover, these teachers simultaneously belonged to the community of course designers and that of classroom teachers who taught the designed course(s). These characteristics of the course design activity reflect the three components of activity theory, namely labor division, rules, and community. As discussed in 3.1, contradictions within each component of the activity system as well as those between different components are a drive of change.
Therefore, in this study, after the data were coded in accordance with each component of the activity theory, the interaction within and between the components were examined so as to identify whether there occurred any contradictions and whether those contradictions were resolved. In short, the data analysis and interpretation in the present study went through the following steps:
First, the coding process started with open coding which aimed to identify any useful segment of data (Merriam, 2009, p.204). In open coding, key words in relation to the hierarchical levels of the activity of course design (i.e. activity – motive, actions –
95
goals, and operations – conditions) as well as the outcomes (i.e. what the teacher learned) were searched line by line.
Second, for each activity system, a priori coding using the six components of the activity system as predefined categories was conducted. The codes and subcategories to identify course-design activity system, which were built upon Dang’s (2013) codes and subcategories and Lohman’s (2000) framework for analyzing environmental inhibitors to informal learning (Tables 3.7).
Third, contradictions are identified according to the coding rule by Murphy and RodriguezManzanares (2008) and put into a table as in the following example:
Table 3.6: Coding example
Contradiction Definition Evidence Resolution
Subject – division of labor
Unequal power relationship
She (Hong) named the skills that were needed for the 3B Listening-Speaking sub-course, the topics (which were the same or similar to those of the 3B Reading-Writing sub-course), and the list of materials that she already had. What we needed to do was to find more materials that matched with the given skills and topics…Hong probably had a lot of experience in course design.
Additionally, I found her reasons [for her initiated ideas] quite convincing. She talked gently, but the arguments were rational. (Follow- up interview)
None
Fourth, the outcomes, which are considered to be what teachers learned through their involvement and derived from the resolution to the contradictions, were coded into two categories based on curriculum development theory by Graves (1996), Nation and Macalister (2010) and Richards (2001): (1) curriculum development knowledge, including curriculum development process, approach to curriculum development, material selection and development; (2) Knowledge of subject matter; (3) knowledge of testing and assessment. As qualitative analysis is interpretive (Barkhuizen et al., 2014; Dửrnyei, 2007; Merriam, 2009), member-checking was undertaken to avoid bias during this step. As the former narratives were quickly analyzed before the later were conducted, member-checking was carried out during the process of oral narratives in which the participants were asked to clarify the unclear ideas and
96
confirm the interpretation of the data they provided in the previous narratives.
Table 3.7: Codes and subcategories to identify course-design activity system Codes Sub-categories from the data set
Level 1
Sub-categories from the data set Level 2
Subject Educational background Qualifications Expertise
Learning strategies Teacher’s prior teaching
experience
Years of teaching Target students Subject matters Teacher’s previous
professional development (PD) opportunities
PD activities
Lessons learned from the previous PD activities
Teacher’s social background
Characteristics Family background Relationships at work Teacher’s previous design
experience
Working in a task force with the current team members
Working in a task force with the teachers other than the current team members Object Raw object Accomplishment of an assigned task:
developing EAP course(s) Culturally more advanced
object Mediational
tools and artifacts
Teacher’s practical knowledge
Classroom image
Image of classroom teacher cohort Course design experience
Teacher’s theoretical knowledge
Subject content knowledge
Competence-based teaching and assessment Curriculum development knowledge Materials of the subject
matter
Available commercial course books Other self-designed course books Available online resources Tool for team work Email correspondence
Face-to-face meeting (regular team meeting & appraisal meeting) Other PD activities Workshops
Mentoring Short trainings Community Course-designer team
Other teacher-colleagues Target students
Experts
Head of Divisions Faculty Board
97 Division of
labor
Contribution to environment analysis
Contribution to need analysis
Contribution to course outline
Course overview Assessment intentions Developing principles Contribution to material
development
Material evaluation and selection Material adaptation and development Task design
Contribution to
development of assessment tools
Assessment tasks and criteria Test specifications and tests Contribution to teacher
orientation
Course introduction and explanation Teachers’ guides
Power relationship among group members
Who initiates ideas?
Who controlled the process?
Who made major decisions?
Who gave feedback?
Rules
Cultural norms Nature of innovation
Organizational rules Timelines
Financial resources Incentives and rewards University workload Collaborative rules Mutual support
Trust
Leaders’ expectations Member’s expectations Members’ personal
obligations