On 23 Feb 2004 01:01:12 -0800, [EMAIL PROTECTED]
(Raoul Kamadjeu) wrote:

> Can someone tells me more on the design effect?

I wondered if that could really be something specific in 
statistics, so I googled on < "design effect" >.  
The first couple of dozen hits did not look very useful, but I 
suppose these, below, could be what you are looking for.  
Since they are  .pdf  files, I can't readily copy lines.

There were 7000+  hits in all, most of them irrelevant.
This time, adding "tutorial"  did not help.  Adding ANOVA
gets a narrower list that shows that the words 'design effect' 
do words together often, without people meaning something
specific.

==== a few google results
[PDF] Sample size and design effect
File Format: PDF/Adobe Acrobat - View as HTML
... by Gene Shackman, NYS DOH, Presented at Albany 
Chapter of American Statistical Association, March 24, 2001 
Sample size and design effect This presentation is a ... 
www.albany.edu/~areilly/albany_asa/confweb01/abstract/Download/shackman.pdf
 - Similar pages 


Developing Sample Sizes for New Surveys: Estimating the Design Effect 
Adjustment. Gene Shackman New York State Department of Health. ... 
www.albany.edu/~areilly/albany_asa/ confweb01/abstract/developing.htm 
 - 2k - Cached - Similar pages


[PDF] C:\SUDAAN\3_6 insert.wpd
File Format: PDF/Adobe Acrobat - View as HTML
... 3.6 Computation of Design Effects The design effect (DEFF) 
is defined as the ratio of the properly computed actual variance 
of an estimated parameter to the ... 
www.rti.org/sudaan/pdf_files/3_6insert.pdf - Similar pages 


-- 
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
"Taxes are the price we pay for civilization." 
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to