2 ms·

Obligatory note that it’s unethical to do a real control experiment at society wide levels, so we will never untangle causation/correlation.

• jgalt212 3y ago
  natural experiments to the rescue!
• Retric 3y ago
  Which is why I was suggesting widows, they should be more directly comparable to single mothers after you correct for income.