There are several ways to teach the logic of hypothesis testing. I chose to start teaching hypothesis testing using margins of error and confidence intervals because I believe these concepts are more intuitive than that of a p-value. As I previously said, however, we can also do hypothesis testing by paying attention to p-values. Importantly, whether we test our hypothesis using confidence intervals or p-values, the conclusion will always be same.

It's like diagnosing a fever using Celsius or Fahrenheit: different numbers, but same conclusion. If you go back to R and re-run your difference in means and difference in proportions tests, you will notice that R also prints a p-value for you. You can be certain that if the 95% CI does not include zero, the p-value will be smaller than 0.05.

As we move onto conducting regression analysis, we will move away from confidence intervals and pay more attention to p-values. The video below is an additional way to teach the logic of hypothesis testing using p-values. Keep in mind that my videos from the last module, my videos in this module, and Alex's videos are all about bivariate hypothesis testing--we are all teaching the same thing in different ways. It is always good to teach a complex subject in different ways because different ways to teach might work better for different students.