Transcript:

Hi, I'm Charlie Kasov, and this is, "What is partial correlation in statistics?" Now, in statistics, partial correlation is a method of isolating two variables to examine their relationship. So, let's call those variables x and y. Now, partial correlation uses what's called linear regression to figure out if a relationship that we thought existed between x and y really does exist or if the variable z is what actually has the relationship with x or y. Now, we can also use linear, linear regression to find the partial correlation between x and y that we didn't know existed, maybe z was masking the relationship between x and y. And by using partial correlation, we figured out that there actually is a relationship. So, those are examples of partial correlation in statistics.