Survey of classroom use of representations: development, field test and multilevel analysis

Abstract

Because of the multimodal nature of learning, doing and reporting science, it is important that students learn how to interpret, construct, relate and translate scientific representations or, in other words, to develop representational competence. Explicit instruction about multimodal representations is needed to foster students’ representational competence in the classroom. However, only a handful studies have surveyed how representations are actually used in science classes. This might be because of the fact that economical instruments for assessing the use of representations in classrooms are not available. To bridge that gap, an instrument was developed, field-tested in biology classes with 175 and 931 students, respectively, and analysed using exploratory and (multilevel) confirmatory factor analyses. Results supported an instrument with six scales and 21 items at the individual and classroom levels covering the following dimensions: (1) interpretation of visual representations, (2) construction of visual representations, (3) use of scientific texts (verbal representations), (4) use of symbolic representations, (5) number of terms used in class, and (6) the extent to which active social construction of knowledge is possible in the class. The scales showed satisfactory discriminant validity and reliability at each level. Further applications of this instrument for researchers and teachers are discussed.