Multimodal transcription of video: examining interaction in Early Years classrooms

Video is an increasingly popular data collection tool for those undertaking social research, offering a temporal, sequential, fine-grained record which is durable, malleable and sharable. These characteristics make video a valuable resource for researching Early Years classrooms, particularly with regard to the study of children’s interaction in multiple modes, which might include gaze, facial expression and movement in addition to language. Collecting video and carrying out multimodal analysis presents both possibilities and challenges, not least because multimodal methodologies are still in relative infancy. One particular challenge is developing forms of multimodal transcription. This paper seeks to reflexively explore the effect of applying different transcription conventions to the same section of video to investigate how two children interact at a computer in an Early Years classroom. It is suggested that the modes to which the researcher attends, and how these are (re)presented in transcripts, shape the analytical and rhetorical interpretations of children’s interaction, with multimodal transcription potentially enabling particular insights into patterns within and between modes as they are orchestrated in interaction.