"John Knight" <johnknight at usa.com> wrote in message
news:P9149.3409$eb.176341 at news2.west.cox.net...
>> "Parse Tree" <parsetree at hotmail.com> wrote in message
> news:dlV29.4417$sI2.1798259 at news20.bellglobal.com...> > "John Knight" <johnknight at usa.com> wrote in message
> > news:2mT29.58427$Fq6.5084565 at news2.west.cox.net...> > > "Parse Tree" <parsetree at hotmail.com> wrote in message
> > > news:nCL29.3972$9g7.936910 at news20.bellglobal.com...> > >
> > > Which PROVES, once and for all, that you know NOTHING about
probability
> > and
> > > statistics.
> > >
> > > No wonder you couldn't do the simplest of calculations.
> > >
> > > Why don't you tell us, Oh Great White Hope For Women, exactly how many
> > > Americans The Gallup Organization needs to poll in order to get an
> > accuracy
> > > of plus or minus 2%?
> > >
> > > If you can't answer that question, you will PROVE, once and for all,
> that
> > > attempting to educate women is a HUGE mistake, and that spending an
> EXTRA
> > $8
> > > trillion only to get THIS is even worse than a HUGE mistake.
> > >
> > > John Knight
> >
> > Gallup polls are political, and don't have teh complexity of an academic
> > test.
>> This isn't a true statement, and even if it was, this would have nothing
to
> do with the accuracy of polls or studies. Gallup's political polls are a
> small part of all of their polls. And even if they were political, the
> exact same principles of probability and statistics apply to Gallup that
> apply to TIMSS.
"The biggest errors in the Gallup Polls aren't statistical, but the bias
introduced because they're an obvious advocacy group.
The above disclaimer has to *really* be applied to their polls."
You've already agreed with me. There is an obvious bias because their polls
are subjective matters.
Also, though the exact same principles apply, the same values do not.
