Teaching probability: effects of task frames and training on misconceptions

The aim of the present work is to improve knowledge about how students think about probability in order to find educational methods intended to eliminate misconceptions. The role of task frames and training on reasoning about sequences of independent outcomes was investigated. Three types of frames have been compared: abstract, concrete, and concrete plus reference to stereotypes. Tasks were administered to two groups of students: one group had followed a brief course on probability, the other had not. Frames influenced the use of normative strategies as well as training, whereas some students, even after the training, continued to hold misconceptions. In particular, the use of normative strategies was more likely in abstract frames. This effect held both for training and non-training group. However, training group was more prone to avoid mistakes in concrete plus reference to stereotypes frames. Results suggest that these tasks can be used to illustrate normative responses and biases in teaching probability