How not to fool yourself with p-values and other statistics

In 2016 the American Statistical Association published its first position paper on the use of p-values, and recently this year a group of statisticians and other researchers published a paper recommending a change to the traditional definition of statistical significance. This talk discussed some of the p-values problems and solutions raised in these and other papers, illustrating the ideas through a series of examples of published studies. The need for more attention to the communication and "human factors" of statistics and data science was also discussed.