Multi-camera on-site video technology and post-lesson video stimulated interviews were used in a purposefully inclusive research design to generate a complex data set amenable to parallel analyses from several complementary theoretical perspectives. This symposium reports the results of parallel analyses employing distributed cognition, variation theory, systemic functional linguistics, and positioning theory of the same nine-lesson sequence in a single science classroom during the teaching of a single topic. While each analysis is demonstrably valuable in itself, in combination, these results demonstrate the inadequacy of any single theory or theoretically-driven analysis to capture the complexity of the science classroom and the corresponding need for inclusive multi-theoretic research designs.
Science classrooms offer a rich educational environment, providing recordable instances of language use, a variety of classroom organizational groupings, varied instructional practices (demonstration, lecture, whole class discussion, and collaborative group work, both experimental and reflective), the utilization of a variety of artefacts (both physical and conceptual), the potential for ontological, epistemological, ethical and moral tensions to emerge, and, arguably, a highly diverse range of learning outcomes. It is this richness and complexity that offers the opportunity for the interrogation of current theory and that also poses the greatest methodological challenges.
The focus of this symposium is a seventh grade science class of twenty-seven students aged between twelve and thirteen years. Data generation employed a four-camera approach (Teacher camera, two Student cameras, Whole Class camera), including onsite mixing of camera images into split-screen video records used to stimulate participant reconstructive accounts of classroom events in post-lesson student and teacher interviews (methodology adapted from Clarke, 2006). The data related to any one lesson consisted of seven videotape sources (including three split-screen records), together with seven audiotracks: two per student group, two for the whole class camera, and one for the teacher camera, plus copied or scanned written material, researcher observation and field notes, plus the videos of two student interviews and the teacher interview.
The research design employed multiple, synchronised recordings of classroom interaction in order to maximize the sensitivity of the parallel analyses to a wide range of classroom actions and learning outcomes, and to promote a form of reciprocal interrogation, where the theories are employed to generate and analyse the data and the comparison of the analyses facilitates the reflexive interrogation of the theories.
Multi-theoretic research designs raise several issues for educational research:
(i)    Contradictory findings generated by parallel analyses
(ii)    Explanatory differences for common findings
(iii)    Complementarity of findings
(iv)    Enhanced but necessarily incomplete coverage
(v)    The challenge of synthesis.
A previous symposium (EARLI, 2009) addressed points (i) and (ii). In addition to reporting the substantive results of each analysis, this symposium is intended to catalyse discussion around points (iii), (iv) and (v). These issues are pursued in the final paper.

Reference
Clarke, D.J. (2006). The LPS Research Design. Chapter 2 in D.J. Clarke, C. Keitel, & Y. Shimizu (Eds.), Mathematics Classrooms in Twelve Countries: The Insider’s Perspective. Rotterdam: Sense Publishers, 15-37.