Belgium: The Most Dangerous Place in the World

Robert Pelton Young is apparently full of shit, because Belgium is the most dangerous place in the world (at least in terms of the number of people killed per capita by natural disasters). I always had a suspicion that the peaceful countryside was a facade.

Somewhat more seriously, because I don’t want to just crack wise about human tragedy, this is pretty surprising stuff. Assuming the figures are correct, and given Belgium’s tiny population of 10.4m people, this strikes me as a classic example of high variation in a small sample.

[Update] Turns out the Belgium may not be nearly as dangerous as this study says. Matter of fact, as a commenter on the above article points out, the most dangerous place — at least for mathematicians and statisticians — may be the University of Leuven from which these figures originated. It looks like they may have screwed some simple math.

[Annual Disaster Statistical Review 2006 via ResourceShelf]

Related posts:

  1. Edge.org: The Most Dangerous Idea? Letting Pundits Prattle
  2. Lifestyles of the ARM-ed and Dangerous
  3. Calgary is Healthiest Place to Live?
  4. Mobile Porn and Dangerous Lads in Pubs
  5. Apple VC Deemed Dangerous

Comments

  1. gg says:

    um. is there some explanation for this data?

  2. olivier says:

    I can explain: The linked pdf mentions 940 people dying in Belgium during the 2006 heat wave. That could very well be… But there’s about 10 million inhabitants in Belgium, so that makes 9 people killed out of 100,000 in this disaster, not 9000. That’s the blunder. 9000/100,000 calculates to 900,000 deaths on our total population. Now, I’m from Belgium, and I haven’t heard of a disaster that killed 900,000 people in 2006. By the way, the authors made the same dumb mistake with the Dutch figures … these should also be divided by 1000.

  3. olivier says:

    And additionally: The French figures are also 1000 times too high. Boy oh boy. I think I know what I’m going to leak to the Belgian press tomorrow morning… Btw, Paul, I love how your tone suddenly became so diplomatic on the “underlying human tragedy” of the Belgian people. Hell, even the Belgians don’t care about the Belgian people, so your gentleness was nice, but really not necessary.

  4. olivier says:

    Closing comment: I’m going to leak to the Belgian press what a bunch of incompetent people are working in that Epidemiological Institute at the University of Leuven. In case somebody thought I was in any way annoyed by Peter, who was also amazed by these numbers. Love your blog, peter! ;-)

  5. olivier says:

    I mean, Paul, not Peter. (It’s 4am here… must go to bed and stop working NOW)

  6. No problem. Thanking Peter to please Paul is fine by me :-)
    And thanks for doing the number-crunching. Looks like we thereby have an even better story than the one I first posted!

  7. Felix Salmon says:

    Um, I don’t see the problem here: the math looks perfectly fine to me. If you look at Table 1 on the previous page, the total number of deaths ranges from 940 in the Belgian heat wave up to a maximum of 5,778 in the Indonesian earthquake. If you use CIA factbook populations, that works out to 9.045 deaths per 100,000 people in Belgium, and 2.462 deaths per 100,000 people in Indonesia. Evidently the authors of the paper are using slightly different population figures, but there doesn’t seem to be anything wrong here — and certainly nothing off by three orders of magnitude.
    If anything, the screwup seems to be by your commenter Olivier, who doesn’t recognise a decimal point when he sees one. That 9.057 figure means just a hair over 9, it doesn’t mean something over nine thousand. Which is why when he started checking the other numbers, he “discovered” that ALL of them were out by a factor of a thousand! Ahem.

  8. Avi Bryant says:

    This is a localization issue – in Belgium, where Olivier is commenting from, 9.057 means nine thousand and fifty seven and 9,057 would mean nine and 57 thousandths. In some other countries, of course, the opposite is true. The confusion is understandable but Paul, I think you need to retract your update.

  9. My update is hereby retracted, and Belgium is back to being dangerous. Thanks Felix and Avi.
    Having now also had a chance to rescan the original numbers — I was on Blackberry making updates last night — everything also looks okay to me.

  10. fewquid says:

    It’s the french fries. The Belgians make some of the best in the world. And they sell them willy-nilly by the side of the road to anyone that’s interested. Must lead to some kind of disaster somewhere ;-)
    But seriously… Belgium dangerous? Seems like a stretch. Every time I’ve been there, there has been a lot of horse manure (really, I kid you not), but other than the nasal and shoe-related dangers of manure, it seemed pretty safe to me…

  11. olivier says:

    I stand corrected, which shows one shouldn’t comment on posts in the middle of the night. As Avi Bryant pointed out, 9.057 indeed reads as nine thousand fifty seven down here, and a sleep-drunk (not drunk on belgian beer-drunk) head makes you forget to read commas where there’s points, and vice versa. So apparently the 940 deceased during the heat wave of 2006 (http://en.wikipedia.org/wiki/2006_European_heat_wave#Belgium) catapulted us to the number one spot after all. And it’s still rare to find air conditioning in Belgian houses! These temperatures are -historically- too rare to justify the investment. Although two heat waves (2003 was also hot) does not a climate change make, the amount of record breaking temperatures have been on a roll lately. The past winter was also weirdly warm (barely one week of freezing).