A Multidimensional Diagnostic Assessment of Preservice Mathematics Teachers’ Conditional Probability and Bayesian Reasoning

This study develops and validates a multidimensional diagnostic test to examine preservice mathematics teachers’ strengths and weaknesses regarding conditional probability and the Bayes rule. We estimated preservice teachers’ mastery of four fundamental skills (i.e., attributes): understanding and applying, i. basic probability concepts, ii. conditional probability, iii. the Bayes rule, and iv. dealing with difficulties and biases. We administered the test to a national sample of 750 preservice mathematics teachers and analysed their responses using the log-linear diagnostic classification model. The analysis revealed that 22.7% of preservice teachers mastered all four attributes, while 21.8% did not master any attributes. The average mastery percentages of the four skills were 74.3, 68.3, 47.8, and 53.2, respectively. We also report the diagnostic quality of the individual test items, mastery classifications, and the relationships among the assessed skills. Implications of the findings and further research suggestions are discussed.