Don't put faith in that Chinese pollution study. Their "statistically significant" discontinuity in life expectancy only happened because they chose to fit the

Don't put faith in that Chinese pollution study. Their "statistically significant" discontinuity in life expectancy only happened because they chose to fit the data with a strange third order polynomial plus step function. Take a look at the data behind that China pollution study and see if you would draw the same fit line: http://andrewgelman.com/2013/08/05/evidence-on-the-impact-of-sustained-use-of-polynomial-regression-on-causal-inference-a-claim-that-coal-heating-is-reducing-lifespan-by-5-years-for-half-a-billion-people/ http://andrewgelman.com/2013/08/05/evidence-on-the-impact-of... There's so much bad science in the world. :(