Educational research The development of a two-tier multiple-choice diagnostic instrument for evaluating secondary school students’ ability to describe and explain chemical reactions using multiple levels of representation A. L. Chandrasegaran a* , David F. Treagust a and Mauro Mocerino b a Science and Mathematics Education Centre, Curtin University of Technology, Australia b Department of Applied Chemistry, Curtin University of Technology, Australia e-mail: A.Chandrasegaran@curtin.edu.au Received 29 May 2007, accepted 14 June 2007 Abstract: A 15-item two-tier multiple-choice diagnostic instrument was developed to evaluate secondary students’ ability to describe and explain seven types of chemical reactions using macroscopic, submicroscopic and symbolic representations. A mixed qualitative and quantitative case study was conducted over four years involving 787 Years 9 and 10 students (15 to 16 years old). The instrument was administered to sixty-five Year 9 students after nine months of instruction to evaluate their use of multiple levels of representation. Analysis of the students’ responses demonstrated acceptable reliability of the instrument, a wide range of difficulty indices and acceptable discrimination indices for 12 of the items. The teaching program proved to be successful in that in most instances students were able to describe and explain the observed changes in terms of the atoms, molecules and ions that were involved in the chemical reactions using appropriate symbols, formulas, and chemical and ionic equations. Nevertheless, despite the emphasis on multiple levels of representation during instruction, 14 conceptions were identified that indicated confusion between macroscopic and submicroscopic representations, a tendency to extrapolate bulk macroscopic properties of substances to the submicroscopic level, and limited understanding of the symbolic representational system. [Chem. Educ. Res. Pract., 2007, 8 (3), 293-307.] Keywords: chemical reactions, macroscopic, submicroscopic and symbolic representations, student conceptions Introduction The theoretical basis of this study is a constructivist approach that is grounded in the belief that what a learner already knows is a major factor in determining the outcomes of learning (Ausubel, 1968). Learners are provided with opportunities to develop new understandings with the teacher acting as a facilitator of learning rather than as a transmitter of knowledge. The complex and abstract nature of chemistry makes the study of the subject difficult for students (Ben-Zvi et al., 1987, 1988; Johnstone, 1991, 1993; Nakhleh, 1992; Gabel, 1998, 1999; Treagust and Chittleborough, 2001). As a result, students tend to hold particular idiosyncratic views about scientific phenomena and concepts that they bring with them to science lessons. These conceptions that students develop (referred to as student conceptions) are the result of several factors, such as their sensory experiences and the influence of their cultural background, peers, mass media as well as classroom instruction (Duit and Treagust, 1995). Very often, the conceptions that students develop about the behaviour of matter tend to differ from the views that are held by the scientific community (Osborne et al., 1983). It is likely that students are satisfied with their own conceptions as a Chemistry Education Research and Practice, 2007, 8 (3), 293-307. This journal is © The Royal Society of Chemistry