«Available online at Linguistics and Education 21 (2010) 257–281 Positioning proﬁciency: How students and teachers ...»
Valdés, 2001; Yoon, 2008; Zentella, 1997). Following guidelines for interpretive inquiry, ethnography, and participant observation (Creswell, 2007; Emerson, Fretz & Shaw, 1995; Erickson, 1986; Hammersley & Atkinson, 1995; Miles & Huberman, 1994), my ﬁeldwork included intensive long-term participation in the school, careful recording and documentation (ﬁeld notes, observation protocols, event maps, analysis memos, interview transcripts, questionnaires, student work, digital audio and video recordings), and an analytical reﬂection about the documentary record during the data collection process. I visited the school two to three times a week, staying most of the school day, from August to June. During my school visits, I took on multiple roles as observer, participant, assistant teacher, small group leader and substitute teacher.4 Informally, I came to know students and their families by giving rides, visiting homes, and meeting with parents and students at local cafes or taquerias. In addition to classroom, lunchtime, and recess observations, I observed extracurricular events, community forums, and parent–teacher conferences in order to explore ethnographically school conditions and social contexts that supported or constrained opportunities for two-way language learning.
My interpretation of these data was triangulated with student interviews and surveys as well as interviews with teachers and parents in October, January, and May.
To capture language practices inside and outside of classroom, I used methods similar to Zentella’s (1997) study and equipped focal informants with portable tape recorders to register the details of natural language use throughout the day in class, at recess, at lunch, and during other intervals. To capture data at the whole class level, I also placed audio recorders on students’ desks supplemented by a video camera in the corner of the room. I shadowed six focal students taking notes to accompany the audio recordings, following shadowing or tailing methods described in Hawkins (2005) and Olsen and Jaramillo (1999). I used Hymes’s “SPEAKING grid” (Hymes, 1972, p. 58) as an initial, crude analytical framework or heuristic to guide my observations, note taking and development of event maps. Hymes’s description of key features of a communicative event calls attention to the (1) scene/situation, (2) participants, (3) ends or goals, (4) activity sequence, (5) key or mood, (6) instrumentality or forms of speech, (7) norms of interaction, and (8) genre. An example of how I adapted these features to create event maps can be seen in Appendix A.
To capture data at the whole class level, I placed audio recorders on students’ desks supplemented by a video camera in the corner of the room. Over the year, I collected over three hundred hours of audio recordings from student shadowing and over ﬁfty hours of collaborative classroom activities, supplemented by twenty hours of video.5 All audio and video recordings were accompanied by daily, detailed ﬁeld notes and analytical memos, which served to identify areas for more detailed transcriptions and shape further data collection. I listened to each of the recordings in their entirety while making analytical notes and selected recordings for detailed transcriptions that highlighted language exchange during student interactions. Upon a second pass of the data, I focused on recordings that revealed classroom participation patterns related to perceived proﬁciency, which became salient during the analytical coding process described below.
3.2. Data analysis
Drawing upon methods from ethnography (Emerson et al., 1995; Green & Bloome, 1997; Hammersley & Atkinson, 1995; Heath, 1982; Spindler & Spindler, 1987) and grounded theory (Corbin & Strauss, 2008), data analysis and data collection were an ongoing process, one inﬂuencing the other. My analysis was constantly informed by ongoing conversations with students and teachers about how they teach and learn language and how they choose to use language with whom. These conversations opened new directions for inquiry.
To deal with a large volume of data, I used the qualitative software package ATLAS.ti to developing coding schemes that connected across ﬁeldnotes and audio recordings; and referring to methods from Emerson et al. (1995), excerpts from ﬁeldnotes were connected with analytical points and explanations. I used Functional Pattern Analysis (Rogoff et al., 2002) to reduce the complexity of data from in-depth naturalistic observations by coding observations into larger 4 As a former teacher in a dual language program in a border community, I could identify with the teachers at Escuela Unida, whose students were similar to my former students growing up in bilingual communities. Although I had the opportunity to develop close relationships with several students and their families, I was also acutely aware of my limitations, knowing that my role as an outsider in the community remained dominant.
5 6 focal students were audio recorded for full day (6–7 h per day) one day per month over 8 months = 300 h total for focal student shadowing.
In addition, recordings from audio recorders placed in small groups on desk include whole class and centers (50 h), jigsaw activities whole class (3 times for1 h = 3 h), information gap activities (10 pairs, 2 h each = 20 h recorded), writing revisions/pair writing (7 pairs, 2 times, 28 h). All recordings accompanied by ﬁeld notes and memos to guide selected transcription.
262 M. Martin-Beltrán / Linguistics and Education 21 (2010) 257–281 categories (as seen in Fig. 1), concepts, or deﬁnitions that reach across sources of data (Hammersley & Atkinson, 1995;
Miles & Huberman, 1994; Spradley, 1980; Taylor & Bogdan, 1998).
One reoccurring concept that emerged from coding ﬁeldnotes, memos, and participant interviews was talk about proﬁciency. Proﬁciency was a term used repeatedly by the participants at school, and the concept of perceived proﬁciency was also “observer identiﬁed” (Hammersley & Atkinson, 1995, p. 211) in analytical memos throughout the data collection process. Implicit talk about proﬁciency and positioning was harder to quantify, but it was clear that the data were “saturated” (Charmaz, 2006) with examples of talk about proﬁciency that positioned students as members or non-members of discourse communities.
My research process developed and deepened recursively. I re-read ﬁeld notes and transcripts and coded these data with a focus on perceived proﬁciency. During the Ethnographic Research Cycle (Castanheira et al., 2001; Spradley,
1980) and process of abstracting themes (Rogoff et al., 2002), I reﬁned my original research question – to examine school conditions and contexts that afforded or constrained opportunities for language exchange among students – and developed guiding questions to analyze the data further in order to look closely at the social construction of perceived proﬁciency. I developed the following sub-questions to guide more focused coding. These questions also guided my analytical framework, which can be seen in Fig. 1.
1. What do students and teachers say about their own and others’ proﬁciency?
2. How do students and teachers choose which language (Spanish, English or both) to use with others? How does language choice reﬂect and reify perceptions of proﬁciency and/or grant/deny access to participate?
3. How do students participate differently across languages? What are the opportunities to use target languages?
4. How do students participate differently across different contexts (i.e. classroom settings, activities), using Hymes’s framework to delineate contexts?
5. How do students and teachers recognize other students’ participation? Are students positioned as experts, novices, “legitimate peripheral participants” (Lave & Wenger, 1991) or non-ratiﬁed participants (Goffman, 1981)?
3.2.1. Levels of analysis As an ethnographic researcher, I sought to be attentive to members’ meaning-making as part of the un-parceled ecosystem of the school that is connected to local community and reﬂective of larger social structures. However, in order to begin to map out such a multilayered tangled system of actors and sociopolitical histories and environments, I found it helpful to begin using “foci of analysis” (Rogoff et al., 2002). Based on sociocultural activity theory, Rogoff et al. suggest that foregrounding the personal, interpersonal and community/institution planes at different moments in the analysis while keeping the other foci in sight.
To analyze my ﬁeld-notes, focal student audio recordings, and interviews, I began by constructing a picture of my focal students’ personal journeys throughout typical school days or daily rounds (Martin-Beltran, 2006). I considered student biographies that shaped their academic experiences and language development. I used event maps (Erikson & Schultz, 1981) to trace how students interacted with each other in various languages within a typical school day, using Hymes (1972) SPEAKING heuristic (described above in Section 3.1) and paying close attention to participation frameworks, footing (Goffman, 1981) and taking of the ﬂoor (Edelsky, 1981). An example event map of one focal student’s day (see Appendix A) illustrates how I drew upon these concepts in my analytic framework. I took note of the participation norms and participation frameworks within the classroom which Schiffrin (1994) suggested captures “the set of positions which individuals take in relations to an utterance” (p. 243). I applied Goffman’s (1981) notion of footing to delineate the participation framework and to recognize ratiﬁed participants and non-ratiﬁed participants. I also used event maps to develop a larger picture of how student experiences compared across the school year throughout different contexts. I ﬂagged data for more detailed transcription that included talk about proﬁciency or illustrated moments when students were positioned as experts, legitimate contributors, or (non) ratiﬁed participants (see Appendix A for example event map).
Rather than making the individual the focus on my analysis, the interpersonal plane of analysis was the heart of my study. I focused analysis and transcriptions on moments of participation (or non-participation) in Spanish, English, and bilingual discourse communities as I examined student interaction, dialogue, and silence. With the audio and video recordings of collaborative classroom activities, I focused on interactional patterns coordinated between speakers to coconstruct language learning opportunities. With a community and institutional focus of analysis, I examined interviews, ﬁeld-notes from school meetings, and recordings of everyday talk that referenced the school and community setting.
M. Martin-Beltrán / Linguistics and Education 21 (2010) 257–281 263 I used these levels of personal (self), interpersonal (others) and institutional (school) to examine different positioning practices across participants and contexts. In this paper I have selected data from each of these levels to show how levels are interconnected. In the analysis I examined how these multiple levels can be simultaneously indexed in the momentto-moment discourse of class participants. Fig. 1 explains forms of evidence at each of these levels that were identiﬁed during coding processes, which also served as an analytic framework to address research questions. In addition to the text and talk about proﬁciency, this analysis shed light on the social practices within the school that positioned students as proﬁcient or less proﬁcient speakers of the target languages. These acts of positioning were manifest in participation patterns, choice of language, choice of peer partners, social networks, and tracking by proﬁciency assessments.
In my analysis of ethnographic ﬁeld notes, I attended to discursive practices and chose transcripts of classroom discourse for close analysis. In order to examine how perceived proﬁciencies were enacted in moment-to-moment interactions, I analyzed discourse turn-by-turn. I analyzed talk about proﬁciency that was enacted, ascribed, and discussed in the company of others.
My approach to analyzing classroom discourse incorporates insights from ethnography of communication, interactional ethnography, and critical conversation analysis (Castanheira et al., 2001; Freeman, 1998; Garﬁnkel, 1967;
Palmer, 2009; Sacks, 1984; Schiffrin, 1994). Ethnography of speaking literature reminds us that we need to attend to an analysis of the contexts, within which utterances occur; therefore, I began with an analysis of larger patterns across the school day and year (using event maps) followed by a more detailed analysis of moment-to-moment interactions.
To examine the process of the construction of perceived proﬁciency in the larger data set, I identiﬁed the following categories of discursive practices observed throughout the school year: practices of accommodation, naming, and declarations of proﬁciency. Although discursive practices are also “acts of positioning”, I separated them for purposes of analysis (see Fig. 1) to give attention to the utterances as transcribed speech. I delineated acts of positioning to refer to extra-linguistic actions. These categories were based on participation and exclusion/inclusion patterns evident in the event maps (i.e. participation frameworks, observations of who is included/excluded). I re-coded the data for these categories and then selected excerpts of discourse to transcribe in detail and analyze moment-to-moment talk. This analysis revealed how classroom discourse reﬂected and constructed student positionality and identities as proﬁcient (or non-proﬁcient) members of discourse communities.
4. School context