Alan - You've already received several promising replies to your inquiry,
but one that no one mentioned may be worth noting:
Agresti, A., & Finlay, B. (1986). Statistical methods for the social
sciences, 2nd ed. San Francisco: Dellen Publishing Company.
This was Agresti's unusually sophisticated intro-stats text. His
next-to-the-last chapter on Models for Categorical Variables makes
reference to adjusted residuals and standardized residuals. On p. 492 he
writes that "adjusted residuals ... make a cell-by-cell comparison of fo
and fe. These adjusted residuals behave like standard normal variables
when the model truly holds. Hence, a large adjusted residual (say,
exceeding 2 or 3 in absolute value) gives evidence of some lack of fit in
that cell. ... Some computer packages also report for each cell a
residual, which equals fo - fe, and a standardized residual, which equals
(fo - fe)/sqrt(fe). These are more difficult to interpret inferentially
than adjusted residuals, because their sampling distributions are not
standard normal."
Best wishes for your work.
Gordon Bear School of Science Ramapo College Mahwah NJ 07430-1680 USA
-----------------------------------------------------------------------------
Voice: 201 684-7754 Fax: 201 684-7637 E-Mail: [EMAIL PROTECTED]
On Thu, 6 Jul 2000, Alan McLean wrote:
> Hi to all,
>
> For some years I have been teaching a technique which I know as testing
> the components of chi square in a standard contingency table problem. If
> you calculate the standardised residual
>
> SR = (fo - fe)/sqrt(fe)
>
> for each cell, these residuals are approximately normally distributed
> with mean zero and standard error given by
>
> SE = sqrt((1 - rowsum/overallsum)*(1-columnsum/overallsum))
>
> provided the expected frequencies are large enough (as for the use of
> chi square itself).
>
> My problem is that I have no source for this technique. I have never
> seen it in a textbook. (I have no doubt about its validity, and frankly
> don't understand why textbooks do not refer to it.)
>
> Can anyone give me a reference to it? Ideally, a reference to its
> original publication.
>
> My thanks in advance.
> Alan McLean ([EMAIL PROTECTED])
> Department of Econometrics and Business Statistics
> Monash University, Caulfield Campus, Melbourne
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
