Chicago Boyz

                 
 
 
What Are Chicago Boyz Readers Reading?
 

 
  •   Enter your email to be notified of new posts:
  •   Problem? Question?
  •   Contact Authors:

  • CB Twitter Feed
  • Blog Posts (RSS 2.0)
  • Blog Posts (Atom 0.3)
  • Incoming Links
  • Recent Comments

    • Loading...
  • Authors

  • Notable Discussions

  • Recent Posts

  • Blogroll

  • Categories

  • Archives

  • Don’t Trust Numbers Without Understanding What They Actually Mean

    Posted by David Foster on May 9th, 2018 (All posts by )

    It seems that the German railroad Deutsche Bahn excludes those trains that break down en-route (or that never even start) when calculating arrival statistics…and there are a lot of such trains.  (via Cold Spring Shops)

    So it would be pretty inappropriate to compare DB’s schedule performance, calculated in this way, with the schedule performance of a railroad that did include broken-down or never started trains in the late-arrival category.

    There are a lot of cases in which statistics may not be actually comparable in the way that they are assumed to be…for example, it seems that the US calculates infant mortality in a different way from most other countries, owing to the different treatment of premature births.

    Journalists, in general, fail completely in explaining what the numbers that they are citing actually mean…probably because in most cases, they don’t understand themselves.

     

    16 Responses to “Don’t Trust Numbers Without Understanding What They Actually Mean”

    1. Mrs. Davis Says:

      Journalists, in general, fail completely in explaining what the numbers that they are citing actually mean…probably because in most cases, they don’t understand themselves.

      This isn’t limited to numbers. Think firearms, conservatives, climate…

    2. Brian Says:

      Of course, even more of a dirty and dangerous secret is that SCIENTISTS, in general, fail completely in explaining what the numbers that they are citing actually mean…probably because in most cases, they don’t understand themselves.

    3. Jonathan Says:

      Much of what gets called science in the popular press is merely the repetition of buzz phrases and metaphors, and appeals to the authority of theories.

    4. Bob Says:

      http://www.numberwatch.co.uk/number%20watch.htm

    5. PenGun Says:

      It could be pointed out that it really depends on the purpose of the data array when you produce statistics like this.

      If you really wanted to know what percentage of trains, not in an incident, arrived on time that’s just the data set you need. The name and purpose of the collection is very important.

    6. David Foster Says:

      Pen…”If you really wanted to know what percentage of trains, not in an incident, arrived on time that’s just the data set you need” That’s true. But what most travelers want to know is almost certainly, ‘What are my odds of getting there when I’m supposed to?’, and, ditto, most people comparing RR systems would like to know the same thing.

    7. David Foster Says:

      Any number that involves *electricity*, you can count on most media…including, sadly, the business media…to get wrong. The distinction between a Kilowatt and a Kilowatt-Hour seems to be totally incomprehensible to journalists.

    8. Brian Says:

      If you talk to a mathematician, approximately 100% of scientific papers are based on incorrect use of probability and statistics. Each field has its own way of calculating confidence intervals, etc., almost none of which are ever actually done in a mathematically rigorous or appropriate way.

    9. PenGun Says:

      :almost none of which are ever actually done in a mathematically rigorous or appropriate way.”

      Pshaw … they are results oriented, modern, forward looking, results, not your silly math based stuff. That’s so last century. ;)

    10. Bill Brandt Says:

      That’s sorta like playing golf and allowing mulligans whenever you choose. Years ago when I was in the Army there (45 or so) I was always amazed at the trains that seemed to depart as soon as that minute hand hit the appointed number.

      But if they are going to cheat ;-)

    11. ErisGuy Says:

      If even the (West) German government falsifies its statistics….

      Never believe a government study.

    12. Mrs. Davis Says:

      If even the (West) German government falsifies its statistics….

      What the hell do you mean even? We’re talking the Huns, the Krauts, the STASI. Those people don’t deal in truth. Look at their exhaust numbers. Biggest mistake in American history was Johns Hopkins University.

    13. John Cunningham Says:

      Another state bowler is homicide rates in the UK vs. USA. In Britain, it is counted as a murder only after a conviction. In The US, a dead body shot for stabbed is counted as homicide imediately.

    14. John Cunningham Says:

      Make that “,howler.”

    15. Kirk Says:

      Anyone accepting the proffered numbers in some piece written to argue for a particular course of action is getting just what they asked for, if they don’t bother to try to verify and understand the described math, and the source data. Just assume that the bastards are lying to you, and try to find it. If you can’t, wellllll… You’re just going to have to look harder for it.

      Innumeracy is something I suspect has been encouraged in the general public; elsewise, the majority of our politicians would have been lynched a long damn time ago. Along with most of our “financial advisers” and “investment counselors”, the vast majority of whom are charlatans and crooks. I’ve heard of honest men in those professions, but to be quite honest… I’ve never actually met one.

    16. Grurray Says:

      A particularly egregious example

      https://qz.com/1269172/the-epic-mistake-about-manufacturing-thats-cost-americans-millions-of-jobs/

      Trump was correct that China was cleaning our clock, but they didn’t just take our jobs. They also took everything else. All the while we were lying to ourselves through bad statistics.