3 ms·

That chart shows a decline trend from 1990-2009, and then an increase every single year after that. It’s been 13 years, at which point do we call it a trend? E

• Retric 4y ago
  That basic question is why people started using statistics to analyze data. Your article is talking overall deaths not pedestrians, which is a useful point of comparison but doesn’t address pedestrian fatalities. It’s easy to say something is a trend, much harder to say if there is some specific case or if it’s random variation. Do something like compare rates of smartphone adoption vs pedestrian deaths between countries and look for correlations and you don’t want to go with gut feel. Also, look at that chart again, 2 years after 2009 were lower than the year before. I only bring it up because of how easy it is to see trends that aren’t there.