Learning Resources

Introduction to p-values

p-values are a key concept for publishing scientific research. Unfortunately, they’re non-intuitive and confusing for many. This video teaches the fundamental ideas of p-values in a relatable, easy-to-understand way in just three minutes. Best of all, it features puppies!

 
 


 

Visual Explanation of Multicollinearity

Multicollinearity happens for example when you are looking at two or more predictor or risk factors at the same time. For example, to calculate age-adjusted odds of a surgery complication. Multicollinearity is when the 2 or more predictors are correlated which can make it more difficult to tease out the separate effects of each. A little correlation is ok, too much and you can’t tease them out and worse statistical software output can’t always be trusted and there may not even be an error message. This is a great visual showing the little correlation vs too much.