Video recording and the research process

RESEARCH ISSUES TESOL Quarterly publishes brief commentaries on aspects of qualitative and quantitative research Edited by CONSTANT LEUNG Kings College London Video Recording and the Research Process CONSTANT LEUNG King9s College London London, England MARGARET R HAWKINS University of Wisconsin–Madison Madison, Wisconsin, United States doi: 10.5054/tq.2011.254526 his is a two-part discussion Part 1, English Language Learning in Subject Lessons, is authored by Constant Leung, and Part 2, Video as a Research Tool/Counterpoint, is authored by Margaret R Hawkins Working with different research concerns, the two researchers attempt to draw attention to a set of methodological and theoretical issues that have emerged in the research process using video data The broad aim is to signal the need to address these complex and intriguing issues explicitly, as video material is used in a great deal of educational and social science research T PART 1: ENGLISH LANGUAGE LEARNING IN SUBJECT LESSONS (LEUNG) A good deal of recent qualitative research in English as an additional language (EAL) in naturalistic settings has drawn on the principles and 344 Project title: Modelling for Diversity: Academic Language and Literacies in School and University Funded by the Social and Economic Research Council (UK) Grant no RES062-23-1666 Research team: Constant Leung, Brian Street, and Sabina Sica TESOL QUARTERLY Vol 45, No 2, June 2011 techniques of ethnography In this discussion I raise two conceptualcum-methodological issues: the affordances and limitations of video(and audio)-recording of events, and the role of theory in constructing data out of events recorded in this way The points I make here have grown out of a research project on language and literacy practices in ethnically and linguistically diverse high schools and universities in London, England.1 In the National Curriculum in England, EAL is not regarded as a distinct subject area Students with EAL backgrounds, irrespective of their English language proficiency, are expected to participate in regular subject lessons at their age grades In English universities, teaching is generally conducted through the medium of English (with the exception of some language programmes) Broadly speaking, the curriculum at both school and university does not provide any sheltered EAL provision This curriculum arrangement suggests that the learning and teaching of EAL should take place as part of the everyday classroom activities across all the subject areas In this research project we have adopted an ethnographic approach and, among other things, have made extensive video–audio recordings of lessons and tutorials in both schools and universities in biology, business studies, English, and history as part of the study For reasons of scope and space, this discussion is restricted to issues related to video–audio recordings Facilitation for Building Data Ethnographic research generally involves field work at the research locale In the pre-video–audio era much of the field work observations would be recorded in the form of field notes, drawings, and still photographs, supplemented by artefacts associated with the events under investigation These modes of recording information were clearly limiting when it came to noting down language and other transient semiotic signs The advent of video–audio technology has, as DuFon (2002) puts it, made it possible to record ‘‘denser’’ linguistic information—it is now theoretically possible to record every utterance within its multimodal environment This, in principle, will provide researchers with a higher degree of fidelity in their records of the flow of action and interaction being studied The replayability of video–audio recordings, in addition to allowing researchers to transcribe the language-in-interaction to a very high level of detail, offers researchers the facility of repeated viewing This opens up possibilities of researcher collaboration by sharing data vicariously By viewing (and listening to) video–audio recordings collaboratively, different researchers can add layers of perceptions and widen the angle of approach in ways that a researcher working alone (or a small team) might not be able to achieve As Goodman-Segall (1995) suggests, working on data RESEARCH ISSUES 345 collaboratively with other researchers (particularly those not on one’s own team) can generate a broader basis for validity claims This potentiality is connected to issues concerning the role of theory in working with data ethnographically (this point is discussed again in the next section) That said, video–audio recordings are in themselves only partial records of an event in two senses First, in a narrow technical sense, a recording of, say, a lesson is likely to capture a partial view of what is occurring It is a generally observed practice that ethnographically oriented studies should preserve the naturally occurring flow of events as far as possible In pursuit of this, most researchers would only use one camera (or two at most) in any classroom This necessarily limits what is captured, whether the camera is fixed in one position or not A good deal of what occurs is often not within the frame Second, any recording of a lesson is likely to contain events that are part of a wider set of activities spanning over several lessons or even terms and that are not necessarily themselves available in any form of recording For instance, a revision exercise is likely to refer to earlier lessons and other teaching–learning activities on the same topic To understand participants’ perceptions and actions, the researcher would therefore have to look beyond what is immediately available in audiovisual form All of this amounts to saying that video–audio recordings in fact represent mere observational fragments of what is occurring (albeit seductively fulsome); a good deal of work has to be done before video–audio material can be regarded as fit for description, let alone analysis and interpretation—terms that we explore further below, drawing upon Wolcott’s (1994) helpful account Working With Data Few, if any, nonexperimental investigators would suggest that any directly observed event in and of itself would provide sufficient information to account for itself as social action or interaction One may say that an observed and recorded event is no more than raw data Such data have to be analyzed Wolcott (1994) suggests that the process of constructing data from experience in qualitative research involves three conceptually different kinds of analysis (although he is careful to point out that this distinction is a heuristic to help with mapping the research process, and in fact, the three categories continually overlap): Description addresses the question, ‘‘What is going on here?’’ Analysis addresses the identification of essential features and systematic description of interrelationships among them—in short, how things work [At this stage the researcher will draw upon analytic terms to help extend the description, e.g., literacy events and practices] 346 TESOL QUARTERLY Interpretation addresses processual questions of meanings and contexts: how does it all mean?’’ ‘‘What is to be made of it all?’’ [Here the description and the analysis are related to larger theoretical questions derived from fields of study, conceptual frameworks, etc.] (p 12, original italics) These three kinds of analysis are not necessarily sequential, and the boundaries are not rigid; researchers can and go back and forth in a recursive manner and bring elements of one level into another Each level, however, involves some measure of selective attention and framing For example, to describe students sitting in groups presupposes that this is a worthwhile feature of the event being studied (and other equally observable features not of interest) To make the analytic point that the interactions between the teacher and the students in a lesson have been driven by an initiation–response–evaluation (IRE) sequence is to signal a particular theoretical interest (and intellectual tradition) This almost unavoidable selective attention and theorising when working with data on a moment-by-moment basis is made more complicated by another methodological consideration The ethnographic imperative of the constant comparative (e.g., Heath & Street, 2008) means that researchers are minded to uncover as many possible ways of understanding a focal phenomenon as possible, that is, learning as much about the phenomenon in question as possible to achieve a thick description and being able to make links to other data that might (provisionally) be seen as comparable That, however, is only one side of the comparative ; the researcher is also expected to make the familiar strange in a bid to make sure that no stone is unturned; this requires the researcher to entertain ideas and perspectives that may at first seem unfamiliar or even antithetical to their own views This aspect of ethnographic analysis is undoubtedly time consuming and intellectually taxing, and the end point is not always clear (e.g., how many different view points should one consider?) There is also the sense of the researcher engaging in a lone struggle against the odds: one can never know and understand enough The replayability of video–audio recordings, with its affordance for detailed transcription of both speech and action, has added new dimensions to the already complex process of working with data The technical facility to make repeated passes at a dataset, either by oneself or in collaboration with others, has made the analytic process perhaps even more multilayered and multidirectional We illustrate this by drawing on our current work In our London-based school-university– focussed project one of the research interests is how EAL issues are dealt with by teachers and students in the mainstream subject classroom One way of handling this is to look for actions that may be construed as EALoriented In a 3.5-year ethnography of EAL students’ English language learning experiences in mainstream and EAL classes in a high school, Harklau (1994) examined, inter alia, the extent to which pedagogically RESEARCH ISSUES 347 motivated speech adjustments such as reducing speed and complexity by teachers were evident when speaking to EAL students For Harklau two of the comparative points were (a) the potential differences between what went on in the mainstream lessons and in the EAL lessons, and (b) the known and theorised knowledge that certain kinds of spoken language used by teachers were more likely to facilitate English language learning than others What if such comparative points are not available? From the outset of the research project we have been aware that this particular interest in EAL is likely to be more salient to the research team than to the participant teachers and students in the study After all, the participant teachers and students in the subject lessons were there to work on the subject content, not the English language In many such lessons the recordings suggest that the teachers and students just got on with doing the lesson The talk and the teaching materials in the lessons did not betray any feature of EAL-oriented activity (such as those examined by Harklau) Repeated viewing of (and listening to) the recorded material provides, in principle, limitless opportunities for describing, analysing, and interpreting the data For instance, one could look at the ways in which students’ speaking turns are orchestrated by the teacher, and begin to develop an account of how EAL students were being inducted to this facet of language use in a classroom context Or one could examine how teachers introduced content vocabulary items (e.g., by referring to written materials or by using spoken IRE routines or by reformulation, e.g., from students’ more colloquial speech to specific technical vocabulary associated with the subject in question) and begin to work out what kind of language learning this may mean for EAL students The list can be extended almost infinitely, especially if we take account of collaborating researchers’ points of view In a sense, our dataset can be described, analysed, and interpreted in a variety of ways, and much depends on researchers’ interests and their propensity to the constant comparative The key question here is, Has this technological affordance created a new issue for ethnographic research—the creation of an omni-purpose dataset? And how might this affect the accounts that the researchers eventually give? In the usual research tradition, the researchers may report on their findings at conferences and take account of audience reaction as they draft further accounts; or they may write articles and again receive feedback from journal editors, reviewers, and readers But the increased availability of audiovisual accounts, and the possibility of shared sessions at the point of contact with these data, may mean that such outside perspectives are built in from a much earlier stage Might research accounts begin to look more hybrid as a result, and might individual/team authorship be harder to define and track? This has considerable implications for what counts as academic literacies—a 348 TESOL QUARTERLY theme that is both part of the content of the research described here and also, it would appear, a key element of the reporting PART 2: VIDEO AS A RESEARCH TOOL/COUNTERPOINT (HAWKINS) Although video recording is now a commonly used mode of data collection in ethnographic and other language and educational research, it is quite striking that it is not explicitly addressed or discussed in most research methods publications in the fields of EAL/ ESL/applied linguistics, or in those specific to literacy and education A discussion of affordances and limitations seems long overdue, as does serious engagement with considerations of the role and place of video data in the research process In the discussion above, Leung has contextualized his discussion in a current research study of high school and university classrooms in England I provide a counterpoint to that discussion, using exemplars from a project that focused on the language, literacy, and academic development of English learners (kindergarten through grade 10) in U.S schools Our primary concern was to identify what (all) teachers need to know to adequately support the schooling of English learners In order to that, we followed five diverse English learners (different age/grade levels, ethnicities and languages, geographic locations, length of time in the United States, and English proficiency) to videotape a full day of their lives both in and out of school We then had a team of four people separately code the data for salient themes, after which we came together to compare, contrast, and discuss Ultimately, we identified 13 key themes that we feel are crucial for teachers’ professional development Data Building: What Gets Represented? As Leung has pointed out above, video data allow for affordances not available through audio recording alone From an applied linguistics standpoint, we have long recognized that meaning is made not just from speech alone, but from all of the extralinguistic, or paralinguistic, cues that always accompany speech (Although, as Cameron, 2001, points out, care must be taken as to how to transcribe such gestures and cues.) Clearly, video enables us to capture the full range of tools directly employed by interlocutors in situated interactions This, however, is not enough for accurate comprehension of the communication that is occurring From a sociocultural standpoint, meaning is indeed made, and languages acquired, between people, rather than as an in-the-head isolated activity However, the underRESEARCH ISSUES 349 standings people construct are shaped by more than words, gestures, and physical signs at play in specific interactions They are mediated by spaces, cultural models, and resources that may not be readily visible or recognizable in a video recording In addition to Leung’s excellent point in regard to the need to understand activities and interactions longitudinally—that is, as they occur within a stream of events and conversations—they must be understood as imbued with meaning from their physical, cultural, social, and historical contexts Video is not able to capture this Thus, for example, an understanding of a particular event for a particular learner at school relies on understanding the school and community environments (including histories of and attitudes toward immigrants), teachers’ views and understanding of good pedagogical practices for English learners, school and district policies, and so forth In our project, we captured an entire day in learners’ lives, thus giving a more nuanced view than capturing a single event might, but we still were not able to fully contextualize what we captured And what, specifically, you are looking for, in terms of theme or topic, will influence what you see, as will the dispositions and understandings of who is doing the looking Whose Voices? It seems to me that the key drawbacks are not, in fact, in the video data itself—surely it is a luxury to have too much data, though it may perhaps be overwhelming and confusing, especially for novice researchers—but they are located in the limitations of the tools and modes we have available for analysis A key concern is that, as has been well documented in ethnographic research, no analysis can be neutral or objective (e.g., Duranti, 1997; Ochs, 1999) The very focus of the research, and what is taped, is dependent on the interests and perspectives of the researcher, or research team Thus, in our project, we felt that we must understand multiple aspects of learners’ lives and experiences in order to understand language, literacy, and academic development in school, and that determined what we taped We felt that too many cameras and videographers would be overwhelming in immigrants’ homes and classrooms, and thus we had one camera person and one sound person, which determined what we could (and could not) capture And some members of the team came immersed in educational literature and experiences regarding immigrants and schooling, whereas others did not (one came with an abiding interest in social justice issues globally), and that too determined what we saw Others surely would have picked up on other aspects of learners’ lives and schooling 350 TESOL QUARTERLY Equally important, there has been attention in educational research to issues of research on as opposed to research with Videotape captures events but not participants’ understandings or explanations of events Thus, consistent with issues discussed in critical applied linguistics, researchers must safeguard in whatever ways possible against attributing meaning to, or imposing meaning upon, others (e.g., Pennycook, 2001) Attention must be paid to capturing the perspectives and understandings of participants Put another way, this has been discussed in the research literature as validity/reliability, and the ethnographic research literature promotes member checks Researchers are certainly responsible for both analysis and interpretation (though, as Leung suggests above, the boundaries between these are quite blurry, if they exist at all) but must include the additional step of ensuring that participants’ (in the video events) voices are represented, either in the research itself or post-research as confirming/validating interpretations In our case, we included interviews with school staff, students, and parents as part of the video data, specifically asking them to clarify our understandings of what we were seeing, and ascertaining how they each viewed schooling for the focal students In cases where we had interpretive questions, we asked post hoc As a perhaps interesting note, we came to think of the camera itself as another eye, or viewpoint Even though, for example, I was present throughout all of the recording, what I saw when reviewing the video was often different from my memory and my field notes At times the camera and recorder caught interactions that I did not see or hear (a matter of focus at any given moment), or caught interactions from a different angle which suggested alternate interpretations It speaks to the affordances of video that the camera serves as an additional viewer, thus more robustly capturing events and interactions, and enriching the capacity for meaning making Constraints Leung, above, alludes to the limitations of the camera to fully capture events, based on the limited scope of observation This is also discussed by Zuengler, Ford, and Fassnacht (1998) and Dornyei (2007) Zuengler et al point to blind spots, referring to what the camera does not pick up They also point to the camera as distractor, which indeed it is It must be acknowledged that the act of videotaping does, to some extent, influence the events and interactions that are captured While in my experience ongoing data collection in a given site mitigates these issues, such that when one recurrently videotapes in a classroom, students and teachers get used to the camera, it is also true that certain interactions or RESEARCH ISSUES 351 activities may or may not go on, or certain things may or may not be said, given participants’ awareness of the taping process In our case, to give but one example, we very much wanted to capture the schooling experience of our focal learners However, when a middleschool student walked down the hall with a camera person and sound person in close pursuit, other students would mug for the camera, or make comments about being on television, or talk to our focal student in order to be captured on the tape, whereas they would not have otherwise The Role of Theory Leung (above) discusses how theoretical perspectives shape interpretations of video data Similarly, DuFon (2002) identifies three salient questions, the first of which is, ‘‘How should the interaction be video recorded?’’ (p 41) I will suggest that this is not, in fact, the first question (though it is an important one) and that, whereas theoretical perspectives indeed shape interpretations, they shape the research process long before the data interpretation stage As discussed above, sociocultural and social justice theories of language and learning shaped the inception and design of this project We knew we needed information for teachers; we wanted to be sure that what we represented to them was salient and indeed critical for the school success of English learners Were we aligned with other perspectives in applied linguistics and educational linguistics, we might have focused on finding information about what forms, structures, and vocabulary were needed in the classroom, and which posed challenges for learners We might, instead, have approached it from a standards/ assessment perspective, asking how learners best can indicate what it is that they know and can vis-a`-vis academic requirements Instead, we followed learners both in and out of school, believing that learning occurs through situated interactions, that these interactions are located in larger institutional and societal contexts, and that these contexts shape the language development and learning that occurs within them We included, but were not limited to, taping instructional events This theory also came to bear on how we analyzed and interpreted the tapes We examined participants’ (stated) perspectives, and contextual (school, home, institutional, and societal) factors, and how these played out in and shaped situated encounters We looked at issues of race, class, gender, and other indicators of privilege and power So we noted, for example, when a Hmong student was subject to ‘‘microaggressions’’ (Yosso, Ceja, Smith, & Solorzano, 2009) from peers, based on the portrayal of a Hmong family in a recent Clint Eastwood movie they had 352 TESOL QUARTERLY seen (The movie was entitled Gran Torino, and they addressed the Hmong student as ‘‘hey, gran Torino!’’) And last, theory came to bear on how we packaged what we learned for educating teachers, and the uses to which we have put the video Theory is, in fact, the cohesive bond that holds a research project together, but is also in a sense a source of bias, or subjectivity, on the part of the researcher Video, like all other forms of qualitative research, must be considered subjective, and this makes it imperative that discussions such as this around its affordances and limitations come to the fore in discussions of video as a research tool THE AUTHORSS Constant Leung is Professor of Educational Linguistics in the Department of Education and Professional Studies at King’s College London, in England He is Deputy Head of Department and Director of two MA programmes His research interests include additional/second language curriculum development, language assessment, language policy and teacher professional development Margaret R Hawkins is a Professor in the Department of Curriculum and Instruction at the University of Wisconsin–Madison, in the United States She researches languages and literacies, especially in relation to community-based sites of learning for immigrant and refugee youth, families and schools, language teacher education, global partnerships, and nongateway districts’ responses to new immigrant and refugee populations REFERENCES Cameron, D (2001) Working with spoken discourse London, England: Sage Publications Dornyei, Z (2007) Research methods in applied linguistics Oxford, England: Oxford University Press DuFon, M A (2002) Video recording in ethnographic SLA research: Some issues of validity in data collection Language Learning and Technology, 6(1), 40–59 Duranti, A (1997) Linguistic anthropology Cambridge, England: Cambridge University Press Goodman-Segall, R (1995) Configurational validity: A proposal for analyzing ethnographic multimedia narratives Journal of Educational Multimedia and Hypermedia, 4(2/3), 163–182 Harklau, L (1994) ESL versus mainstream classes: Contrasting L2 learning environments TESOL Quarterly, 28(2), 241–272 doi:10.2307/3587433 Heath, S B., Street, B., & with Mills, M (2008) Ethnography New York, NY: Teachers College Press Ochs, E (1999) Transcription as theory In A Jaworski & N Coupland (Eds.), The discourse reader (pp 167–182) New York, NY: Routledge Press Pennycook, A (2001) Critical applied linguistics Mahwah, NJ: Lawrence Erlbaum Wolcott, H (1994) Transforming qualitative data: Description, analysis and interpretation London, England: Sage RESEARCH ISSUES 353 Yosso, T., Ceja, M., Smith, W., & Solorzano, D (2009) Critical race theory, racial microaggressions, and campus racial climate for Latina/o undergraduates Harvard Educational Review, 79, 659–690 Zuengler, J., Ford, C., & Fassnacht, C (1998) The analyst’s eyes: Data collection as theory CELA Technical Report Albany, NY: SUNY, Center for English Learning and Achievement 354 TESOL QUARTERLY ... one consider?) There is also the sense of the researcher engaging in a lone struggle against the odds: one can never know and understand enough The replayability of video audio recordings, with... that the researchers eventually give? In the usual research tradition, the researchers may report on their findings at conferences and take account of audience reaction as they draft further... 1999) The very focus of the research, and what is taped, is dependent on the interests and perspectives of the researcher, or research team Thus, in our project, we felt that we must understand