‹ BackHN Continuity

Thread

Burning Man death rates – A short lesson in statistics

179 points · 243 comments · viraj_shah

  1. trimethylpurine · · focus · HN ↗
    The 4.9 standardized includes deaths from extremely violent, poverty stricken parts of the country,

    You need to exclude those areas to lower that number before comparing.

    Imagine a tour of the US that includes the more dangerous parts of Oakland, Memphis, and the like.

    Most people wouldn't book that tour, so the base line risk factor you're using isn't relevant for most people.

    1. compiler-guy · · focus · HN ↗
      This is discussed in the article.
      1. trimethylpurine · · focus · HN ↗
        It's not discussed in the article. The article discussed, race, sex, etc. and then makes an assertion that it "wouldn't move much from 4.9."

        That's completely false when using geolocation. We can estimate from crime stats easily that a tour of the country avoiding the highest murder rate parts of the country would land you very close to a zero murder rate.

        Among healthy adults, that's essentially a near zero chance of death.

        That isn't surprising. Attending a festival in a desert with little oversight and lots of drugs is obviously more dangerous than buying groceries in white suburbia.

        4.9 is simply not the relevant number unless your goal is to try to say that the event is on par with taking a walk in the wrong part of town.

        But I'm certain that wasn't the point of the article. It's dishonest, is my point.

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.