Strategies and models in statistical reasoning

The traditional literature did not directly investigate the role of strategies in statistical reasoning. According to the mental model theory, individuals may construct various representations of the same statistical problem and apply various strategies to them. However, the theory posits that the activation of these strategies does not essentially depends on the format in which information is presented (i.e., frequentist vs. probabilistic). In this chapter, we report evidence which corroborates the model theory's predictions. In particular, contrary to the assumptions of the frequentist theory, we show that Bayesian-like strategies can be activated even with probabilistic formats.