I would need to know more information. Are we referring to a rasch model? What is the nature of the scale items? How many subscales are there? How many
items, and how many items within the scales (an average is fine). What type of sample is this intended to be used with? Are there subsamples within that sample? What is that subsample like?
As you probably know, if this is a rasch model, then the probabilistic nature of the model will allow you to utilize smaller and potentially non-representative
samples and still get meaningful information.
http://www.rasch.org/rmt/rmt74m.htm
I found this, which certainly does a better job than I can for figuring this out.
Also note, this is really only a power analysis in regard to the item stability, it has little bearing on power to detect effects in any given use of the measure.
That would require a separate power analysis based on a separate set of factors specific to the study itself, and you would ideally need to know the metric properties of the scale already.
Matthew J Poes
Research Data Specialist
Center for Prevention Research and Development
University of Illinois
From: SPSSX(r) Discussion [mailto:[hidden email]]
On Behalf Of Juanito Talili
Sent: Tuesday, December 13, 2011 11:59 PM
To: [hidden email]
Subject: Power analysis for Rating Scale Model
Any suggestion/reccomendation on the sample size for Rating Scale Model? Power analysis (in general) is very much appreciated.
Locked
