John Whittington wrote:

> Since no-one else has yet commented on this point I made recently, I'm
also going to 
> repeat that I'm intriqued to know why, even if the jury (and Expert
Witness) DID 
> believe that there was only a 1 in 76 million chance that the two cot
deaths had 
> arisen 'naturally', how they jumped from that to the specific conclusion
that the 
> mother was guilty of murder.  Why not the father (if one was on the
scene), or maybe 
> even other family member or friend?

In the Sally Clark case, there was evidence that her husband was out when
the first death occurred, and that she was present and no evidence was
presented that anyone else was there. He was present at the time of the
second death and was arrested (with Sally Clark) on suspicion of murder but
was not prosecuted. You have to bear in mind that there was a lot of other
evidence of various sorts (and pointing in various directions) in this case
apart from the statistical stuff. The original appeal in 2000, indeed,
concluded that the statistics (1 in 73 million etc.) was wrong and
misleading, but that the other evidence was overwhelming so that the wrong
statistics wasn't enough to overturn the conviction. (Many, many people
disagreed with this, of course, including me; I'm mentioning it only to
point out that the statistics was far from being the only evidence or the
only issue, something that seems occasionally to be ignored when we
statisticians discuss it.) The second appeal in 2003 (see
http://www.sallyclark.org.uk/Judgment03.html) was on two grounds, the
statistical one and also that microbiological evidence, based on a
postmortem, that established that one of the deaths was probably due to
natural causes, was not presented in the trial. Most of the discussion in
the judgment is actually about the microbiological point, and they didn't
hear evidence (though they did read statements) on the statistics. They make
it clear that they would have allowed the appeal solely on the microbiology,
and my reading is that this is why they didn't make a big meal of the
statistical points and didn't hear evidence. They did, however, say "Thus it
seems likely that if this matter had been fully argued before
us we would, in all probability, have considered that the statistical
evidence provided a quite distinct basis upon which the appeal had to be
allowed."

In the Angela Cannings appeal (see
http://www.courtservice.gov.uk/View.do?id=2209), which is the one that lead
most immediately to the current reviews of other cases, again there was
evidence that the mother was there at the time of the deaths and no evidence
was presented that anyone else was.  

So in both cases, the decision to prosecute the mother rather than someone
else wasn't entirely based on assumptions about probabilities and so on.

Regards,

Kevin
--------------------------
Kevin McConway
Senior Lecturer in Statistics
Department of Statistics
The Open University
Walton Hall
Milton Keynes MK7 6AA, UK

Phone: +44-1908-653676
Fax:      +44-1908-652140
email:   [log in to unmask] 

******************************************************
Please note that if you press the 'Reply' button your
message will go only to the sender of this message.
If you want to reply to the whole list, use your mailer's
'Reply-to-All' button to send your message automatically
to [log in to unmask]
*******************************************************