Based on your screen shot you are referring to the same score. If the two scores are independent of each other, then create two separate categories to track them. 

Thanks Ahmed,

Just to clarify - are you saying that the 2 embedded data variables (MatrixScoreBel & MatrixScoreUni) are the same?  I can see that the code ${gr://SC_e4X0Q4hEGTXoYke/WeightedMean} is the same but my assumption was that the weighted mean referenced the matrix block immediately above it.  

In addition, when Matrix 1 is below 2 and Matrix 2 is above 2, I get the follow-up question for the first and not the second, which to me suggests they are different numbers.

Sorry if I am missing the point.

Rob

Lets say y = x1 + x2. Till x2 is not defined, the value of y will be equal to x1. However, if x1 is above the threshold, irrespective of the value of x2, y will always be over the threshold. 

This is what is happening in the case you've mentioned and why you're facing issues. 

An easy way to visualize this would be to pipe the value of the score in a question before the branches and see what they are. 

Thanks Ahmed.  

I’m afraid I don’t follow your thinking -  I really am a novice at Qualtrics.  

What steps should I take to get the desired survey behaviour?

@RobBar - Currently, the uniqueness score is being added to the belonging score because you are using the same scoring category for both. To make them independent, each should have its own scoring category.

Thanks Tom.  I see now that the weighted mean was being calculated by the overall score.  I was unaware that you could have different scoring categories, that can be created in the scoring menu.

My mistake was thinking the embedded data weighted mean was referenced to the matrix question that it follows.

It now works