This visualization illustrates the appropriate tests to use when your research hypothesis is that two treatments are equally effective, or that a new treatment is no worse than the current gold standard.
Here's a new visualization that shows the p-curve distribution when comparing the means of two independent samples for varying effects. Many know that the distribution is uniform when the null is true, but what about when it isn't?
The notion is fairly well spread that wait-lists could act as a nocebo condition in psychotherapy trials. In this post I write about some recent results from a network meta-analysis that investigated this.