John Whittington wrote (in small part):
>
> Sure, average age at death clearly tells us 'something' ... but
> exactly what (apart from 'the obvious'!)? I would suggest that, in
> the absence of other information or assumptions, we really would not
> have any way of knowing how to interpret it.
[snip]
>
> My personal feeling is that those who think they can derive useful
> conclusions, in which they can have reasonable confidence, from
> changes in 'average age at death' in Iraq are relying on assumptions
> (e.g. that most of the 'excess deaths' were in pretty young people)
> which, although 'reasonable' assumptions, are far from certainties
> (unless one has data on which to base those assumptions). ... and, of
> course, if one actually had that 'additional data', one probably would
> not want/need to use 'average age at death' of a surrogate for what
> one really wanted to estimate.
>
I tend to agree with all this, but I'd argue that it is not very
relevant to Ray Thomas's suggestion on what one might do with such data.
Ray was talking about using it essentially to corroborate (or otherwise)
the results of the survey published in the recent Lancet article. They
did collect data on age at death, and indeed published some analysis of
it in the article. If one could get data on age at death from some
central registration, or indeed from some different sample, one could
compare the two sets of results (in several plausible ways). If they
were very different, that would leave some important questions to be
answered about the validity of the Lancet survey and/or the other data
on age at death. If they were similar, well, of course that doesn't
establish for sure that the Lancet survey was OK --- nothing is certain
in that sense --- but could increase confidence in the survey results.
This sort of 'triangulation' use of age at death data would be very
different from using it as the only source of information on mortality.
I generally agree with John's previous post that it's numbers of excess
deaths that are the key thing of interest here, not so much the ages at
which those deaths occurred, but if the main reason for looking at the
ages is to give us information about how good the survey sample was,
then it would be useful to look at the ages anyway.
But, that said, I still don't know where some 'independent' source of
age at death data in Iraq is actually going to be found.
Regards,
Kevin
******************************************************
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]
Disclaimer: The messages sent to this list are the views of the sender and cannot be assumed to be representative of the range of views held by subscribers to the Radical Statistics Group. To find out more about Radical Statistics and its aims and activities and read current and past issues of our newsletter you are invited to visit our web site www.radstats.org.uk.
*******************************************************
|