Conditional independencies come in two forms.

1. They are statements of which variables should be associated with one another (or not) in the data.
2. They are statements of which variables become dis-associated when we condition on some other set of variables.

The implication that some variable Y is not associated with some variable X, after conditioning on some other variable Z is written: Y ⊥⊥ X|Z. D ⊥/⊥ A means D “is not independent of” A

If we now look at the data and find that any pair of variables are not associated, then something is wrong with the DAG