Five pillars for teaching probability for linking different meanings, connecting probability with statistical inference, and establishing sustainable intuitions.
Start early and develop the ideas and concepts in a spiral way Using games intelligently to induce sustainable probabilistic intuitions Building Bayesian and risk thinking as early as possible Linking probability and statistical inference from early education Develop the twin relation between probability and risk Shaping conditional probability and a wider conception of statistical inference Conclusions Use all roots of probability to establish sustainable meaning to statistical inference Simplifying the complexity of inference leads to a caricature of the concepts. To shape wider conceptions of statistical inference – shape probability: Migon & Gamerman (1999) and Vancsó (2009) suggest teaching classical and Bayesian inference in parallel. Greater complexity allows for deeper understanding: the challenge is to develop appropriate learning paths. Either we use assumptions in the form of models or we refer to the metaphorical character of probability. We advocate a pluralistic probability perspective that implies a comparative statistical inference (Barnett, 1982), transposed to the educational corner.
Start early and develop the ideas and concepts in a spiral way Using games intelligently to induce sustainable probabilistic intuitions Building Bayesian and risk thinking as early as possible Linking probability and statistical inference from early education Develop the twin relation between probability and risk Shaping conditional probability and a wider conception of statistical inference Conclusions Use all roots of probability to establish sustainable meaning to statistical inference Simplifying the complexity of inference leads to a caricature of the concepts. To shape wider conceptions of statistical inference – shape probability: Migon & Gamerman (1999) and Vancsó (2009) suggest teaching classical and Bayesian inference in parallel. Greater complexity allows for deeper understanding: the challenge is to develop appropriate learning paths. Either we use assumptions in the form of models or we refer to the metaphorical character of probability. We advocate a pluralistic probability perspective that implies a comparative statistical inference (Barnett, 1982), transposed to the educational corner.