[Tango-L] Tango Teaching

[Trini y Sean (PATangoS)]

Yes, I should have said that one "looks" for things that
are significantly different.  So one says that two data
sets are or are not significantly different as opposed to
saying that they are the same.  I was confusing how one
would normally interpret the data with the actual test.

Trini de Pittsburgh

--- Tango Society of Central Illinois
<tango.society at gmail.com> wrote:

> On 5/11/07, Trini y Sean (PATangoS) <patangos at yahoo.com>
> wrote:
> > Hi Steve,
> >
> > Actually, one normally tests for things that are
> > "significantly different" rather than looking for
> sameness.
> 
> Steve, Trini
> 
> Actually, technically Steve is correct. The oddity of
> statistics is
> that one tests (null hypothesis) whether groups are the
> same, but one
> hopes to 'reject' the null hypothesis and 'accept' the
> alternative
> hypothesis that the groups are different.
> 
> Ron Weigel
> Statistician by day
> Tanguero by night
> 
> 
> > --- steve pastor <tang0man2005 at yahoo.com> wrote:
> >
> > > The use of the term "statistically significant"
> caught my
> > > eye, too. Based on my limited knowledge,
> "statistically
> > > significant" refers to data sets which are being
> > > evaluated for
> > >   sameness. This involves quantative, rather than
> > > qualitative data, and the generation
> > >   of a test statistic, etc.
> > >   I think we all knew what was meant by the use of
> the
> > > term, but your words
> > >   "with a reasonable degree of confidence" would have
> > > been more appropriate.
> > >   Respectfully,
> > >   Steve
> > >
> > >





 
____________________________________________________________________________________
Never miss an email again!
Yahoo! Toolbar alerts you the instant new Mail arrives.
http://tools.search.yahoo.com/toolbar/features/mail/