Those data points don't disagree with wyager's point. Both the the Signalling Theory they are advocating and the Human Capital Development (HCD) that I assume

Those data points don't disagree with wyager's point. Both the the Signalling Theory they are advocating and the Human Capital Development (HCD) that I assume you are advocating would make the same predictions on those life indicators you listed. The difference between the theories is which causes which: does consuming education cause people to have more of the traits that cause good life outcomes OR do the traits that cause good life outcomes also cause people to consume more education? Pure HCD theory says it is only the first. Pure signalling says only the second. The two can be mixed. What mix people believe in is important as the two theories result in very different policy goals for our education system. But both agree that people with more education are going to earn more, commit fewer crimes, experience fewer crimes, be healthier, live longer, etc, etc, etc.