Re: Variance of an index of dispersion
- From: "John Uebersax" <jsuebersax@xxxxxxxxx>
- Date: 16 Dec 2005 08:19:38 -0800
> 1. Does it make sense to have a variance of a measure of dispersion,
Yes
> 2. How would I calculate a variance [of the dispersion statistic]
Couldn't this be done using jackknife methods? The jacknife is very
easy given categorical data. With K ethnic categories, you'd
re-calculate the dispersion statistic K times, each time deleting a
single observation of one ethnic group; weight these results (so that
it has the effect of deleting *every* observation in the sample once);
and apply the usual jackknife formula.
--
John Uebersax PhD
.
- References:
- Variance of an index of dispersion
- From: weare
- Variance of an index of dispersion
- Prev by Date: Re: paired observations
- Next by Date: Re: Connotation of the word Treatment
- Previous by thread: Re: Variance of an index of dispersion
- Next by thread: Need Help for a Statistical problem
- Index(es):
Relevant Pages
|