On 23 Apr, 15:57, Peter Clinch wrote:
Mike Clark wrote:
In message
* * * * * Toom Tabard wrote:
[snip]
Not necessarily. You may not have the statistical proof, but the
empirical experience and knowledge of the practitioners in the
particular area can be strongly indicative of the true effect and
point to the flaws, and the reasons for them, in the statistical
analysis. When reality seems at variance with the numbers, it is
frequently the numbers which are flawed. There a legions of social,
health and safety studies where concentration on the numbers followng
the issues receiving concentrated attention, result in problems with
comparison with earlier data which wasn't collected under the same
spotlight, but where the empirical experience of practitioners not
involved in the statistical studies can give some, possibly the best,
evidence of the true effect.
Presumably you can cite some specific examples to substantiate your
generalisations above?
Preferably that take into account the lack of proper control data in a
comparative head injury scenario, and where the practitioners have been
bombarded with years of unsubstantiated claims (the infamous 85% figure
from TRT '89 in particular) about the effectiveness of the proposed
intervention to bias their opinion.
Pete.
--
Peter Clinch * * * * * * * * * *Medical Physics IT Officer
Tel 44 1382 660111 ext. 33637 * Univ. of Dundee, Ninewells Hospital
Fax 44 1382 640177 * * * * * * *Dundee DD1 9SY Scotland UK
net * *http://www.dundee.ac.uk/~pjclinch/- Hide quoted text -
------------------------------------------------------------------------
On 21 Apr, 21:25, Peter Clinch wrote:
But since he's established he really /can't/ tell the
difference between his opinions and fact I really ought to give up
and stop muddying urc's waters further: he's given himself enough
rope, tied the noose and jumped off the stool as far as his
pronouncements on helmets go.
Pete.
--
Peter Clinch Medical Physics IT Officer
Tel 44 1382 660111 ext. 33637 Univ. of Dundee, Ninewells Hospital
Fax 44 1382 640177 Dundee DD1 9SY Scotland UK
net http://www.dundee.ac.uk/~pjclinch/