In analysing the data, I turn to the theory of communities of practice introduced by Wenger (1998/2003) and use multi-methodological lenses. Central to my work is the concept of a learning community and the ways that one identifies oneself (identification) and one’s capacity to negotiate meanings (negotiability) as a member of such a community.
As discussed earlier, I aim to understand the dynamic social relations in which learner identities (e.g. Gordon), agencies (e.g. Wenger 2006; McNay 2004) and gender (e.g. Butler
Table 1: Data collection
Phase 1 Phase 2 Everybody working together
(8 lessons)
Video recordings of the lessons (8 lessons)
Recorded stimulated recall-interviews:
1 group interview with girls 1 group interview with boys 2 individual interviews with girls 2 individual interviews with boys
Teacher diary
Work divided into groups of 7 girls and 9 boys (16 lessons)
Video recordings of the lessons (16 lessons)
Recorded stimulated recall-interviews:
1 group interview with girls 1 group interview with boys 1 individual interview with girls 1 interview in pairs with girls 2 individual interviews with boys Teacher diary
and to observe and explore the practices in which agency is either limited or excluded (Lappalainen 2007), or supported. Further, I discuss both the complications and even the inequalities that children’s gendered interaction may cause in learning, as well as the situations in which the social boundaries were crossed and multi-voicedness promoted.
Hence, my focus is in the ways that students construct identities and memberships, and participate in the mutual shaping of meanings in a BoM classroom. As is common in an ethnographic study, the phases of conducting my inquiry, including the data production, interpretation and theorizing, overlapped each other (Lappalainen 2007).
My analysis draws on Braun and Clarke’s (2006) thematic analysis method (TA).
TA is an empirically-driven approach that identifies and analyses the most fundamental patterns of meaning in a data set (Braun & Clarke 2006; Joffe 2011). Themes can involve both manifest content and latent, more tacit content. This method often views both types of content simultaneously, using the manifest themes as a path to understanding the more latent content. TA uses existing theoretical constructs to examine data but allows:
“emerging themes to ‘speak’ by becoming the categories for analysis” (Joffe 2011, 209).
As explained earlier, when implementing my inquiry, I had three sources of data:
observation data, stimulated recall interviews and a teacher diary. After transcribing the video recordings and familiarizing myself with my field notes and stimulated recall interviews, I first approached the data by looking for all possible, various aspects of learner identity. I was particularly interested in how gender, being a constitutive source of identity, operates as a vehicle of for inclusion and exclusion in the children’s classroom interaction.
(See Appendix 3 for example of transcription)
So the starting point for my analysis was to use the data in order to identify various modes of participation; events that could be decoded as promoting participation; and events that could be decoded as compromising participation (Wenger 1998/2003) in the student-student and student-teacher interaction. I first used open coding (Kvale &
Brinkmann 2009, 202; Bold 2012, 130), coding very concrete and hands-on notions about the actions and interactions that took place during the lessons, such as “boys run around”;
“girls/boys select place next to other girls/boys”; “boys sword fight with the dish brushes”;
“girls wait silently in their seats”; “girls and boys play warm-up games together”; “girls and boys debate topic for group-work task” and so on. I then started to organize, group and conceptualize the codes by exploring the data and identifying what could be related to what.
In the next stage of analysis, I constructed categories of meanings (Kvale & Brinkmann 2009, 202; Gibbs 2007, 47-48) that related to the actions and events I had identified. In constructing the categories, I used both concept-driven and data-driven approaches (Kvale and Brinkmann 2009, 202). On one side, I placed modes of participation that in my mind were connected to the processes of negotiating identities and memberships (identification) in the BoM classroom. I identified these categories as “use of student voice”; “mobility and use of physical space”; and “emphasizing/de-emphasizing gendered groupings”.
On the other side, I placed modes of participation that I thought were connected to the processes of negotiating and shaping shared meanings and creating ownership in learning (negotiability). For this side, I selected categories such as “having/not having ones ideas accepted”; “accepting/not accepting ideas of others”; and “contributing/not contributing to shared enterprises”. Then I went back to my initial codes and organized them according to my categories; I noticed that many of the happenings both promoted and compromised participation depending on from whose perspective their were viewed. (See Table 2)
Promoting
When coding (Kvale & Brinkmann 2009; Bold 2012) and categorizing the data (Kvale
& Brinkmann 2009; Gibbs 2007), I noticed that the student voice operated as a strong legitimator in the BoM classroom interactions (Arnot 2007), regulating access to participation. To get a closer look at the use of student voice, I went back to my observation data, and first, counted the number of times that the students engaging in conversation during the lessons. I was interested in who was talking and what was the conversation about. Hence, I wanted gain knowledge about both the amount and the quality of the conversations. These efforts provided me with an overview of the gendered nature of the use of student voice in the BoM classroom. (See Table 3 in 4.2.1)
In order to move ahead with the analysis, I selected episodes from the observation data that seemed significant for meaning condensation (Kvale & Brinkmann 2009, 205-207).
The selected episodes, teacher diary and stimulated recall interviews were understood as accounts (Atkinson & Coffey 2002). I then used these accounts to organize, re-narrate and retrospectively construct the events of interaction, including choices that were made and activities in the BoM classroom; further I interpreted the meanings connected to the observed happenings (Josselson & Lieblich 2003). The narratives provided select sources for representations of experiences that offered insight “into characters, events and happenings central to those experiences” (Coffey & Atkinson 1996, 68).
I then aimed to build overarching themes within the data that would open up multifaceted insights to the research topic (Braun & Clarke, 2006). I was able to construct three umbrella themes, namely 1) modes of participation; 2) holding gendered beliefs and preconceptions; and 3) shifting between ‘good student’ and ‘rebel student’ identities.
I mentioned above that when conducting the analysis, I approached it by using multi-methodological lenses. In Chapter 5, I present the characters of ‘Emma’ and ‘Amos’. These narratives are representative constructions (Bold 2012) based on the reflections of several children, participating in the activities in the BoM classroom and later in the stimulated recall interview sessions. The video documentation of the interviews allowed me to return to the interview ambiances including tones of voice, facial expressions and feelings. This turned out to be helpful particularly, when constructing the narratives, since sometimes the ways things are said express more than the actual words. Leavy (2009) suggests that using such constructions or fictions in analysis “allows the researcher to re-examine his or her findings in a new context”, which can be beneficial in trying to elaborate the theoretical or other insights (p. 46). Taking such a standpoint, Frank (2000) identifies fiction as a problem-solving strategy (see also Leavy 2009, 44). According to Leavy (2009), in terms of practice, fiction can be used as a part of the methodology that is “consistent with feminist,
that the texture of human experience, in my inquiry informed by the classroom discourse, can be accessed and expressed by using the tools of fiction (ibid.). The aim of including the narratives in the results chapters was, for one, to attend to the voices of the children participating in BoM, and second, to give the reader a taste of the ambivalent nature of this research topic. Hence, the narratives can be viewed as an attempt to represent, through the characters of ‘Emma’ and ‘Amos’, a set of events, ambiances, feelings and a particular type of phenomenon (ibid.), and they can be understood as “a form of analysis and reporting of the research data” (ibid., 145). Bold (2012) suggests that using representative constructions is justified in situations “where the researcher seeks to make sense of diverse realistic data through analysing the parts and then synthesizing them into a realistic framework – a narrative that is readable and meaningful – in preparation for further analysis” (p. 146).