[Rasch] Power and Sample size for a Rating Scale Model

Svend Kreiner svkr at sund.ku.dk
Wed Dec 14 19:01:21 EST 2011


Power has to be defined in terms of a specific test of significance. 
Since we always use several tests during analysis of fit to the Rasch 
model is is impossible to come up with a specific power and a specific 
sample size. Are we talking about infits, about tests of DIF (and which 
tests of fit, there are many and some are more powerful than others and 
therefore need smaller sample sizes) about over-all tests of the model, 
about tests of unidimensionality (again there are many, some with and 
some without power).

For these reasons, I do not think that your question has an answer. 
Remember also, that there are those that claim that power can be too 
large and therefore recommend that sample size should be restricted.

Best regards

Svend

Den 14-12-2011 06:52, Juanito Talili skrev:
> Dear all,
> Any suggestion/reccomendation on the sample size for Rating Scale 
> Model?  Power analysis (in general) is very much appreciated.
> Thank you.
> J

-- 
*Svend Kreiner*
Professor

*Københavns Universitet*
Biostatistisk Afdeling
Det Sundhedsvidenskabelige Fakultet
Institut for Folkesundhedsvidenskab
Øster Farimagsgade 5, Opg. B
Postboks 2099, 1014 København K
DANMARK

TEL 35 32 75 97
FAX 35 32 79 07
LOK 10.2.09
_svkr at sund.ku.dk
___www.biostat.ku.dk_ __


_
__
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20111214/fff6b837/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lv_sund_lille_rgb.gif
Type: image/gif
Size: 2973 bytes
Desc: not available
Url : https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20111214/fff6b837/attachment.gif 


More information about the Rasch mailing list