Summary

Many statisticians now recommend for us to move away from the use of null hypothesis testing and focus more on estimates and intervals. To do this, it’s necessary to understand the different types of intervals that exist and how they relate to each other.

Working with estimates and intervals can help keep us closer to the science we’re interested in. While intervals have some of the same disadvantages as significance tests, they can help avoid the dichotomous thinking that results are either statistically significant or non-significant. It’s also important to present estimates and intervals for meta-analysis, which is one of our main tools we use to identify general and reproducible results.