Buy

Books
Click images for more details

Twitter
Support

 

Recent posts
Recent comments
Links

A few sites I've stumbled across recently....

Powered by Squarespace
« Recycled policy | Main | Full retreat? »
Saturday
Apr062013

Briggs on statistics

Wonderful lecture by Matt Briggs on the absurdities of statistics as applied in the social sciences.

And climatology.

PrintView Printer Friendly Version

References (1)

References allow you to track sources for this article, as well as articles that were written in response to this article.
  • Response
    Response: eye exercises
    - Bishop Hill blog - Briggs on statistics

Reader Comments (13)

Excellent way to spend a cold Saturday morning.
I would recommend this lecture to all government ministers and senior civil servants — except that I doubt their little brains would cope.
It was good to be reminded of John Brignell's list of all the things that will result from global warming though I suspect quite a few of the links on that list will no longer work.

Apr 6, 2013 at 11:02 AM | Registered CommenterMike Jackson

Wonderful is correct.

Spread it widely...

Apr 6, 2013 at 11:16 AM | Unregistered CommenterDodgy Geezer

What a brilliant lecture. Thanks for the post.

Apr 6, 2013 at 12:11 PM | Unregistered CommenterJohn Marshall

The thing that strikes me about Briggs is his wry acceptance of fallibility, and humans' unending capacity for self-delusion.

His weary amusement at the knots we are capable of tying ourselves into represents a kind of reality and humility which is utterly absent in the ranks of the 'science-is-settled' alarmists.

A great lecture.

Apr 6, 2013 at 1:05 PM | Unregistered CommenterRick Bradford

We had a large survey for a highly visible client. I had just retired so I had Matt work on it. He was easy to work with and seemed to get along well with the client. Unfortunately we only communicated by email so I have yet to meet him in person.

He always keeps in mind that our models always tend to have unknown variables and that the addition of a single previously unknown variable can change all the existing "highly significant" variables. Paleoclimatology and climatology have too many known unknowns and unknown unkowns to be confident about anything.

Apr 6, 2013 at 2:45 PM | Unregistered Commenterbernie

Excellent lecture - we can see why government departments are so keen on statistical studies!!

And John Brignell's list is excellent -

http://www.numberwatch.co.uk/warmlist.htm

Apr 6, 2013 at 3:12 PM | Unregistered CommenterMarion

I liked it too. Thing is, this kind of easy to understand critical review can be presented to young people in schools.

...

...

Andrew

Apr 6, 2013 at 3:45 PM | Unregistered CommenterBad Andrew

Great fun - loved the bit about metastudies and subprimes!! :-D

Briggs is fun here too:

"It is Time to Stop Teaching Frequentism to Non-statisticians"

http://arxiv.org/abs/1201.2590 (pdf link at the top right)

Apr 6, 2013 at 5:03 PM | Unregistered Commenternot banned yet

Might I recommend some one has a look at the latest data on the epidemic of human 'mad-cow' disease

http://www.cjd.ed.ac.uk/documents/figs.pdf

It is not only Thermogeddon that has failed to come, we are not losing our brains, our minds perhaps, but not our brains.
The vCJD cost the UK billions of pounds and was all complete statistical bollocks. Note how in the peak vCJD years the crash in Iatrogenic and genetic numbers. Also note the slow rise that we observe in Sporadic
cases that were missed in the past.
At least one person who got wealthy and knighted based on predictive models.

The numbers

Year Sporadic Iatrogenic Genetic vCJD Total Deaths
1990 28 5 0 0 33
1991 32 1 3 0 36
1992 45 2 6 0 53
1993 36 4 7 0 47
1994 54 1 8 0 63
1995 35 4 5 3 47
1996 40 4 6 10 60
1997 59 6 6 10 81
1998 64 3 5 18 90
1999 62 6 2 15 85
2000 50 1 3 28 82
2001 58 4 6 20 88
2002 73 0 5 17 95
2003 79 5 6 18 108
2004 50 2 6 9 67
2005 67 4 13 5 89
2006 68 1 9 5 83
2007 64 2 10 5 81
2008 86 5 5 2 98
2009 80 2 8 3 93
2010 85 3 7 3 98
2011 90 3 14 5 112
2012 82 5 10 0 97

Apr 6, 2013 at 6:59 PM | Unregistered CommenterDocmartyn

And most of these absurd statistics are paid for by taxpayers!

Apr 6, 2013 at 10:30 PM | Registered CommenterRobert Christopher

“The noble art, which had once been preserved as the sacred inheritance of the patricians, was fallen into the hands of freedmen and plebeians, who, with cunning rather than with skill, exercised a sordid and pernicious trade. Some of them procured admittance into families for the purpose of fomenting differences, of encouraging suits, and of preparing a harvest of gain for themselves or their brethren. Others, recluse in their chambers, maintained the gravity of legal professors, by furnishing a rich client with subtleties to confound the plainest truth, and with arguments to colour the most unjustifiable pretensions. The splendid and popular class was composed of the advocates, who filled the Forum with the sound of their turgid and loquacious rhetoric.”

E. Gibbon (“The Decline And Fall Of The Roman Empire,” 1776)

Apr 7, 2013 at 3:36 AM | Unregistered CommenterNikFromNYC

For anyone like me who has zero statistical training it is still enlightening and (almost) entirely understandable, so no need to be put off.

Apr 7, 2013 at 9:11 AM | Unregistered Commenterartwest

Garbage in, G O. The issue with most science these days is everyone is a stato. Not many of them are very questioning statos.

Apr 8, 2013 at 11:00 PM | Unregistered CommenterDavidCH

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Post:
 
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>