Stochastic process computational modeling for learning research

The goal of our research was to compare and systematize several approaches non-parametric null hypothesis significance testing using computer-based statistical modeling. For teaching purposes, a model for simulation created. results were analyzed Fisher's angular transformation, Chi-square, Mann-Whitney, exact tests. Appropriate software created, allowing us recommend new illustrative materials expressing the limitations tests that examined. Learning investigations as technique comprehending inductive statistics has been proposed, based on fact modern personal computers can run simulations in reasonable amount time with great precision. collected revealed most often used small samples have low power. Traditional does not allow students analyze test power because true differences between are unknown. As result, Ukrainian education, including PhD studies, emphasis must shift away from toward modeling practical approach establishing scientific hypotheses. This finding is supported by papers American Statistical Association's recommendation.