"Doug Anderson"  wrote in message
news:37eifuF5c1svvU1{at}individual.net...
> "John Royer"  writes:
>
>> "Doug Anderson" 
wrote in message
>> news:37dm2fF5cui3kU4{at}individual.net...
>> > "John Royer"  writes:
>> >
>> >> "Doug Anderson"
 wrote in message
>> >> news:37a7qaF5a4ktvU1{at}individual.net...
>> >> > "dwacon"  writes:
>> >> >
>> >> >> "Hardpan" 
wrote in message
>> >> >> news:bc5u019ka5udj175go4b966brfialtqc3p{at}4ax.com...
>> >> >> >
http://www.stuff.co.nz/stuff/0,2106,3171380a10,00.html
>> >> >> >
>> >> >> > Sly DNA tests show 1 in 3 dads duped
>> >> >> >
>> >> >> > 30 January 2005
>> >> >> >
>> >> >> > By TIM HUME
>> >> >> >
>> >> >> > Hundreds of Kiwi men are paying out almost
$900 for secret tests
>> >> >> > in
>> >> >> > Australia to determine whether they are really a dad.
>> >> >> >
>> >> >> > And the company which runs the tests says
that in one in three
>> >> >> > cases,
>> >> >> > the man finds he's been duped.
>> >> >>
>> >> >> Wow -- what does that say about Kiwi women?
>> >> >
>> >> > Nothing really.
>> >> >
>> >> > It says that among Kiwi men who believe they
_aren't_ the father of
>> >> > a
>> >> > child, 2/3 of those who go ahead and test turn out
to have been
>> >> > wrong.
>> >>
>> >>  Yup gee, only 33% of the population is illegitimate. Thats not too
>> >> bad
>> >> now,
>> >> IS IT?
>> >
>> > You really have a problem understanding this?  Or are you just
>> > pretending?
>> >
>> > This group is not a random sample, and is not representative of the
>> > population.  How you can be confused about this is beyond me, unless
>> > emotion is clouding your intellect.
>> >
>> >> Another way of looking at it DOUG is that there is a ONE in THREE
>> >> chance
>> >> you
>> >> are an illegitimate child as well. Me too! that sucks,
wouldn't that
>> >> just
>> >> turn your whole world around?
>> >
>> > Actually, chances that I'm illegitimate are zero.  But that isn't the
>> > point.  This report we are discussing says nada about the chances of a
>> > random individual being illegitimate.
>> >
>> > You sometimes have valid points John.  But you reduce your credibility
>> > by maintaining that a report like this has anything to do with a
>> > general population statistic.
>>
>> Doug.I'm not loking for a fight but your last comment essentially says
>> that
>> you are right until I prove you wrong.
>
> Nope.  It says a self-selected sample is far too biased to generalize
> to the population.  I can't believe you really don't get that.
>
>> Discussion with such a person is
>> futile and wasted. I am always willing to listen and learn.
>
> Read some very basic book that discusses proper statistical sampling
> and biased samples then.

For example, one of the more common unanticipated consequences of tracing
the inheritance of DNA markers within a family can be the inadvertant
disclosure of mis-identified paternity. This is not a new issue for clinical
geneticists, who encounter it in the context of prenatal testing and carrier
screening for recessive diseases. However, for clinicians from other
specialties, interested in late onset diseases of adults like breast cancer,
this risk is still under-appreciated. Long-standing recommendations to
address that risk prophylactically in the informed consent process
(President's Commission, 1983) are rarely followed outside of traditional
medical genetic settings. In 1991 one NIH research team published a
disguised pedigree of a family that disclosed the existence of two children
in which "paternal genotype inconsistencies" were identified (Compton,
1992). The sequelae of this disclosure within the immediate family were
disastrous, and led to new guidance to NIH investigators to include the
risks of mis-identified paternity in the informed consent process for these
studies, and not to publish unexpected results that compromised the
confidentiality of research subjects.(Austin and Kaiser, 1993).
Unfortunately, clinicians faced with explaining the risk status of family
members will not have the luxury of simple non-disclosure when unexpected
findings like mis-identified paternity emerge.

http://www.cwru.edu/med/bioethics/pubetj1.html




--- UseNet To RIME Gateway {at} 2/15/05 9:23:57 PM ---