4 ms·

The lead hypothesis is really interesting and certainly plausible, but I don't think it's been proven. I would like something more substantial than "the graphs

• slv77 11y ago
  The paper based the freakenomics study was based on only looked at US data but crime was dropping in the 90's globally. A global drop in crime would be hard to explain based on US policy changes with regards to abortion while lead was removed globally at different times. http://www.bbc.com/news/magazine-27067615 http://www.bbc.com/news/magazine-27067615 The abortion hypothesis was also correlation study (i.e. These graphs look similar) but didn't control for leaded gasoline. Also later studies failed to find more direct links such as higher rates of abortion in high crime areas that would explain the magnitude of the drop. A later study by the author of the paper suggesting lead as the driver of violent crime did a follow up study that incorporated lead, abortion and other variables found that lead was the dominant contributor and abortion a secondary contributor. http://www.nber.org/digest/may08/w13097.html http://www.nber.org/digest/may08/w13097.html
• BurningFrog 11y ago
  The stringent version of "the graphs look similar" is called "strong correlation", and there is some of that for the lead theory. If you haven't read it, the Mother Jones article is the place to start: http://www.motherjones.com/environment/2016/02/lead-exposure-gasoline-crime-increase-children-health http://www.motherjones.com/environment/2016/02/lead-exposure... That article has some critics that you can look up if you're ambitious.