Welcome to the NavList Message Boards.


A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding

Compose Your Message

Add Images & Files
    Re: Lies, damn lies, and statistics...
    From: Robert Eno
    Date: 2013 Mar 18, 20:07 -0400
    The ongoing discussion about the cocked hat and now, statistics, reminds me of a bit of wisdom I learned from one of my high school math teachers eons ago, when he told me that, mathematically, it is impossible to close a door. "What do you mean by that sir" says I? (in those days you actually referred to your teacher as "sir" or by his/her surname)
    He went on to explain:
    A = the point at which the door if fully open.
    B = the point at which the door is fully closed.
    In order to get the door from A to B, one must first get it to a point halfway between A & B. Call it "A prime". In order to get it from A prime to B, one must first get it to a point halfway between A prime and B which we will call "A prime2"...... and on and on and on it goes to infinity; always halving the distance between where you are and your objective but never getting there.
    It was a great mental exercise but I also think that there may be a practical lesson here on theory vs. reality and further, a lesson about needlessly getting bogged down in minutiae.
    I think the finest bit of wisdom I have seen on this subject matter was offered by one of my fellow countrymen, Mr. William Hawes several days ago when he wrote:
    "when I was in the navy I was taught, that when in coastal or pilotage waters, to pick the worst scenario/most dangerous position in the cocked hat. Otherwise when well offshore I generally picked the position closest to the centre or if I had a good DR/EP, I would place the fix closest to the DR/EP position."
    I too, am not a math whiz and in any case, for practical navigation, I am more inclined to adopt William Hawes' rule of thumb and leave the statistical analysis on shore.
    ----- Original Message -----
    From: Hanno Ix
    Sent: Monday, March 18, 2013 4:43 PM
    Subject: [NavList] Re: Lies, damn lies, and statistics...

    May I add:

    There is a real problem with statistics, though:
    Its arguments tend to be expressed as denials of the opposite.
    e.g." With 95% confidence there is no reason not to believe, etc..."
    Add to this abtract concepts and terms
    and you will soon loose the CEO of the Co that paid for the analysis.

    One I know complained  once:
    " Statisticians argument with opponents by proving them  wrong long
    enough until these run out of counter arguments and/or get tired.
    They go home thinking they did their job by telling me what not do 
    while I have to make a decision what to do."

    An exaggeration of course but you get the gist.


    From: Frank Reed <FrankReed---com>
    To: hannoix---net
    Sent: Monday, March 18, 2013 11:48 AM
    Subject: [NavList] Lies, damn lies, and statistics...

    Derrick Young, in the "Darn Old" thread, you wrote:
    "You can prove anything you want using statistics - I've seen it
    done too many times. "
    That's one of those 'things people say', but of course it isn't true. You can't "prove anything" with statistics unless your audience is stupid. Mark Twain's version of this sentiment from over a century ago was "Figures often beguile me, particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: 'There are three kinds of lies: lies, damned lies, and statistics.'" It's folksy and memorable (see PS). A more correct, less pithy version might be something like this:
    "You can browbeat ignorant people with BAD statistical arguments and convince them that practically anything is true, but with GOOD statistical arguments, a competent and intelligent audience can be convinced that many subtle and counter-intuitive hypotheses are true or at least have merit and others are certainly false".
    What we have here is a mixed situation. Some members of our NavList 'audience' have only vague notions of statistics, and so they fear that they are getting the browbeating side of the story. Other members of our 'audience' believe that they really do understand statistics, thank you very much (as they are prone to say), when in reality they have some muddled notions that are getting in the way. And some members of this audience really DO understand statistics though unfortunately it's hard to make any headway when their posts are mixed into this confusing soup. And then there's just plain pig-headed-ness which afflicts nearly everybody with an interest in celestial navigation at some point in time.
    PS: So Twain attributes the "lies, damn lies, and statistics" line to Benjamin Disraeli, but apparently Disraeli never said it either. There's a long Wikipedia article on this simple phrase here: http://en.wikipedia.org/wiki/Lies,_damned_lies,_and_statistics

    NavList message boards and member settings: www.fer3.com/NavList
    Members may optionally receive posts by email.
    To cancel email delivery, send a message to NoMail[at]fer3.com
    View and reply to this message: http://fer3.com/arc/m2.aspx?i=122943

    View and reply to this message: http://fer3.com/arc/m2.aspx?i=122946

    __________ Information from ESET Smart Security, version of virus signature database 8126 (20130317) __________

    The message was checked by ESET Smart Security.

    Browse Files

    Drop Files


    What is NavList?

    Join NavList

    (please, no nicknames or handles)
    Do you want to receive all group messages by email?
    Yes No

    You can also join by posting. Your first on-topic post automatically makes you a member.

    Posting Code

    Enter the email address associated with your NavList messages. Your posting code will be emailed to you immediately.

    Email Settings

    Posting Code:

    Custom Index

    Start date: (yyyymm dd)
    End date: (yyyymm dd)

    Visit this site
    Visit this site
    Visit this site
    Visit this site
    Visit this site
    Visit this site