Simon,Steve wrote
> There is one solution, though it is unlikely to be adopted. When sample
> sizes are limited by the nature of the disease (childhood cancer is an
> example), then we should consider using a larger alpha level. Think in terms
> of new drugs. A Type I error is allowing an ineffective drug onto the marker
> and a Type II error is keeping an effective drug off the market. If the
> current statistical standard effectively prevents most effective drugs from
> getting to the market (high Type II error rate), then make it easier for ALL
> drugs to get onto the market by increasing the Type I error rate.
>
>
To some extent the use of confidence intervals avoids much of this issue and
the EBM community should be applauded for pushing for their use as standard. I
think it would be a shame if the issue of changing the "standard" levels of
significance for studies to suit the situation was completely lost in this. I
have always found the argument that studies should be analysed in such a way as
to balance type I and type II errors very convincing. It might also qualm my
concerns regarding the simple rejection of low power studies since it seems to
me that when we arrive at a study that is designed with type one and type two
errors of 50% (alphan .5 power.5) we can probably genuinely dismiss that as
unethical as the result cannot possibly be more informative than tossing a
coin. Any statisiticians care to comment or am I missing something?
Peter
--
Peter Griffiths
Lecturer, Research in Nursing Studies Section
Florence Nightingale Division of Nursing and Midwifery
King's College London
Waterloo Rd
London, UK
SE1 8WA
+44 171 872 3012 (DDI)
+44 171 872 3219 (Fax)
[log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|