From: Harvey Newstrom (mail@HarveyNewstrom.com)
Date: Fri May 10 2002 - 15:24:42 MDT
On Friday, May 10, 2002, at 10:27 am, Spudboy100@aol.com wrote:
>
>> Source: Duke University (http://www.duke.edu/)
>>
>>
>> Date: Posted 5/10/2002
>> Study Indicates No Natural Limit To Life Expectancy
I always jump on bad statistics, and I'm not sure my previous posting
was specific enough. Let me explain the exact mathematical flaws in the
previous article.
There are two groups of deaths.
Group one is premature deaths. These are caused by nutritional
deficiencies and disease (and accidents). These are being addressed by
medical breakthroughs. Many such deaths are being prevented, so people
in this group are living longer. This makes the average lifespan go
up. However, this group is not affected by old age. They die due to
other causes before old age kills them.
Group two are non-premature deaths. Such people live a full natural
lifespan and die of old age. Old age has not been solved by medical
breakthroughs (yet). Statistics show that maximum lifespans are not
increasing (yet). People who live to the end of their natural lives are
not living longer. However, they are not decreasing either, so the
average lifespan still goes up. This group is always affected by old
age. They avoid other causes of death and always reach old age at the
end of their life.
The study combined these two groups and only looked at increases in
average lifespan. By only looking at increases, they were only seeing
data from group one. Group two did not have any increases, and
therefore did not influence the statistical increases being studied. In
other words, all the statistical analysis really only was done on group
one. Group two had no affect on the data being measured. When the
study concluded that old age didn't exist, what they really were saying
is that none of the people seeing lifespan increases (group one)
indicated any limitations from old age. This is correct. Had they
actually done statistical analysis on people who weren't seeing lifespan
increases (group two), they would have found the opposite conclusion.
All of them were being limited by old age. Within this group they would
have concluded that only old age exists (and no other causes of death
exists).
Basically, this study accidentally measured a smaller subset of people
and then misapplied those measurements to all people. They also
measured average numbers and misapplied these to all individual members
of the group. Both of these are invalid statistical methods. This
study didn't really prove that old age doesn't exist. It proved that
old age doesn't exist for people who die prematurely. It also proved
that none of our increases in lifespan that we have seen have involved
old age.
-- Harvey Newstrom, CISSP <www.HarveyNewstrom.com> Principal Security Consultant <www.Newstaff.com> >
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:13:58 MST