In this lesson we will look at the methods that can be used to evaluate error correlation.

In the previous recipe, we discussed the origins of error correlation and saw that it is introduced whenever there is a common error between measured values due to a common measurement. In this recipe, we’ll discuss practical ways of estimating the error correlation.

In the first recipe in this series, we saw that we can use Type A or Type B methods to evaluate measurement uncertainty. This is also true when we are evaluating error correlation, and as with uncertainty analysis:

- Type A methods are based on a statistical approach
- Type B methods are based on other forms of knowledge

In this recipe we’ll examine both of these methods, and see examples of their application. We’ll start this process on the next page, where we will examine how correlation can be evaluated using a statistical (Type A) approach.