Re: Statistics and Probability in Genealogy

Allen <allen@xxxxxxxxxxx> wrote:

Bob - remove cap to reply wrote:
On Sat, 17 Mar 2007 "Don Moody" <dpmoody@xxxxxxxxxxxxxx> wrote:
But suppose in spite of everything you were able to say that N% of
death ages were inaccurately given by an average of M% a century (or
whenever) ago. That tells you absolutely nothing about whether ONE
particular death was or was not accurately reported or by how much it
was wrong if inaccurately reported. YOU CANNOT MAKE STATS OUT OF ONE
EVENT.
Dead right - unfortunately this misconception seems to be very prevalent
across society. I've even seen reports where the majority of doctors (who
should be intelligent) don't really know what the statistics mean once they
have a positive test result.
Bob

Amen. My son-in-law teaches advance placement calculus (AP courses are
high-school courses i.e. generally taken in the 11th or 12th year, for
college credit). He has also taught an AP statistics course a few times.
I took a large number of math courses in college, but none are as
important in my opinion as an understanding of statistics in today's
world. Over 50 years ago an American author wrote a book titled How To
Lie With Statistics which showed how data can be presented to prove or
disprove almost anything (note--still in print). One example that I
recall is that in the first 20 years or so of the 20th century the stork
population closely correlated with the birth rate in Scandinavia--a
correlation, but hardly a causal relationship.
Allen

I too have the book... the author was Darrell Huff.
One of my favourites is the Gee-Whiz Graph, that can (by careful choice
of scale and origin) change "National income rose 10%" into "...climbed
a whopping 10%".

Recommended reading for everyone - and amusing with it, too.

"Statistical thinking will one day be as necessary for efficient
citizenship as the ability to read and write" - H. G. Wells (from the
first page of the book).
--
Please reply to john at yclept dot wanadoo dot co dot uk.

--
Posted via a free Usenet account from http://www.teranews.com

.

Relevant Pages

  • Re: Statistics and Probability in Genealogy
    ... death ages were inaccurately given by an average of M% a century ago. ... I've even seen reports where the majority of doctors (who ... He has also taught an AP statistics course a few times. ... I took a large number of math courses in college, but none are as important in my opinion as an understanding of statistics in today's world. ...
    (soc.genealogy.britain)
  • Re: Experienced Statistician to help decide whether a regression is legitim
    ... Most malpractice ARE based on "ignorance", ... eager-to-learn EXECS (most CEO and high level managers) who were NOT the ... courses of education on how statistics is properly used. ...
    (sci.stat.math)
  • Re: Experienced Statistician to help decide whether a regression is legitim
    ... Most malpractice ARE based on "ignorance", ... courses if it might enlighten you somewhat doesn't make you all of a sudden a "magus of statistics".. ... education on how statistics is properly used. ... even taught principles management that are much more ...
    (sci.stat.math)
  • Re: Randomness and linguistic development
    ... and use modern probability and statistics. ... phoneticians, and computational linguists. ... You need to have a background in probability and statistical ... kind of "methods" courses, can give you the necessary understanding; ...
    (sci.lang)
  • Re: why is probability and statistics a hard subject?
    ... of upper division / first year graduate. ... is that some statistics courses are poorly taught. ...
    (sci.stat.math)
Loading