Jaqueline17w

Given two disagreeing polls, a single compact & imprecise but taken at deal with-benefit, and the other huge & exact but with a superior prospect of being fully mistaken, what is the ideal Bayesian product to update on these two datapoints?