But from what I understand the basis of the research is that it's wrong for us to use morals when judging perceived risk to a child. They base this on the fact that there's x% chance that a certain risk will happen to any child.

But that statistic is based on general population. My contention is that our moral judgement can be used to heighten perceived risk to a child, even though it might blow it well out of proportion. So just because the average child is at a perceived risk of being injured at x% when left alone, I don't think it's wrong to say a child of parent I judge to be amoral or willfully negligent is most likely at an increased risk of being injured.