Cum hoc ergo propter hoc can be a fallacy which happens often in economic and health data. In the mere fact that two different sets of information differ proportionally, the summary of the causal relation involving them can't be identified without even more info. If Eliphaz's statement is