all of this is assuming of course, that some extreme value ... by ANY
definition ... is "bad" in some way ... that is, worthy of special
attention for fear that it got there by some nefarious method
i am not sure the flagging of extreme values has any particular value ...
certainly, to flag and look at these ... makes no more sense to me than
examining all the data points ... to make sure that all seem legitimate ...
and accounted for ...
actually, the more i think about boxplots ... since one can't get any
notion of frequency PER score value ... i like the notion of just drawing
the whisker to the endpoints ... and be done with it ...
boxplots give you limited data in any case ... why make them work more than
they are worth?
what would be better would be to have a dotplot formation on top of the
boxplot ... or, a dotplot with ... Q1, Q2, and Q3 ... indicated by some
notch technique ...
hmmm ... anyone know of a package that does this kind of a diagram???
doxplot????
At 09:45 AM 8/30/01 -0300, Robert J. MacG. Dawson wrote:
>I wrote:
>
> > Er, no.
> >
> > Q1 ~ mu - 2/3 sigma
> > Q3 ~ mu + 2/3 sigma
> > 1 IQR ~ 4/3 sigma
> > 1.5 IQR ~ 2 sigma
> >
> > inner fence ~ mu +- 2 2/3 sigma which is about the 0.5 percentile.
>
> -right so far -
>
>and then burbled
>
> > The inner fences are selected to give a false positive rate of about 1
> > in 1000.
> >
> > I suppose that if we take into account the Unwritten Rule of
> Antique
> > Statistics that all data sets have 30 elements, this *does* give
> > a "p-value" of (1-e)*30*0.001 = 5% <grin>
>
>which is obviously wrong. The false positive rate is about 1 in 100
>and my fanciful 5% calculation is unsalvageable.
>
> -Robert Dawson
>
>
>=================================================================
>Instructions for joining and leaving this list and remarks about
>the problem of INAPPROPRIATE MESSAGES are available at
> http://jse.stat.ncsu.edu/
>=================================================================
_________________________________________________________
dennis roberts, educational psychology, penn state university
208 cedar, AC 8148632401, mailto:[EMAIL PROTECTED]
http://roberts.ed.psu.edu/users/droberts/drober~1.htm
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================