Developing and Evaluating a Survey for Representational Fluency in Science

Abstract

Various representations, used for communication and problem solving in science, are an unspoken prerequisite for learning, understanding and participating in scientific communities. Work has been done highlighting the importance of competence in particular multiple representations in science learning, the specific representational practices for the different disciplines and to translating between representations. However, limited attention has been paid to obtaining a threshold level of ability in, not only one, but some combination of representations for a discipline. This notion leads to generic fluency with various representational forms used in science, with discipline specific expertise – representational fluency nuanced for a particular discipline. The aim of this study is to examine representational fluency nuanced for physics. This is achieved through the development of a survey instrument, the Representational Fluency Survey (RFS), consisting of representationally rich multiple choice items obtained predominantly from various validated sources. The survey was implemented with 334 students from first year to postgraduate at an Australian university to capture a cross-sectional snapshot of representational fluency nuanced for the specialization of physics. Reliability and validity were determined through standard statistical analysis and through consultation with experts. The results show that representation fluency develops across the years, and that there is a threshold associated with fluency. However, our study does not comment on causality. We demonstrate that in coalescing existing research on multiple representation while paying attention to disciplinary differences is a potentially fruitful pursuit. The RFS test of representational fluency in science is tailored to be used with university physics students but illustrates that adaption for other specializations may be possible.