The ability to extract qualitative information from quantitative information, and/or to create new information from qualitative and quantitative information is the key task of statistical literacy in the 21st century. This paper presents a hierarchy of the graph interpretation aspect of statistical literacy that includes such ability. Participants from junior high to graduate students took part and some of them were interviewed. The SOLO Taxonomy is used for decoding the students' responses and the Rasch model is used for clarifying the construction of the hierarchy. Five different levels of graph interpretation are distinguished: Idiosyncratic, Basic graph reading, Rational/Literal, Critical, Hypothesising and Modelling. These results will supply guidelines for teaching statistical literacy.
The CAUSE Research Group is supported in part by a member initiative grant from the American Statistical Association’s Section on Statistics and Data Science Education