Abstract
Increasing concerns about publication bias, p-hacking, and low power have amplified the uncertainty about which research findings are likely to be true. In this presentation, I provide a practical introduction to recently developed statistical tools that can be used to deal with these uncertainties when designing and evaluating research.
I will discuss meta-analytic techniques such as p-curve analysis and PET-PEESE meta-regression, and explain how well powered studies can be designed even when the true effect size is unknown.
Finally, I will explain the benefits of open science, and provide some examples of how this can be accomplished.