[Rasch] Sample size?
trevor.bond at jcu.edu.au
Wed Sep 30 06:06:23 AEST 2015
The only issue is the density of the data.
If both raters rate all persons on all items, the only issue will be that the no of responses in each category is likely be too small to estimate their difficulty.
Mike uses the very small data set from Guilford as an example. I cover that example in Chap 8 as well. Pls take a look at it.
Sent from 007's iPad
> On 30 Sep 2015, at 12:32 AM, Shirin Shirazi <shirin.shirazi at gmail.com> wrote:
> Dear List Members,
> Good day. I have a question and I wonder if you can afford some time to read and respond. But please take it from a person who does not know the ropes of MFRM.
> If we assume the model is item and sample independent, then does it matter if we have a very limited sample size with limited related facets to run Rasch model? I mean have the minimum and maximum number of facets (examinees, items, examiners, etc) been determined the violation of which hinders us from running the program. For example, two raters, 10 testees, a 15-item questionnaire?
> I would be mostly appreciative of your kind response to my question.
> Shirin Shirazi
> PhD in Applied Linguistics
> Rasch mailing list
> email: Rasch at acer.edu.au
> web: https://mailinglist.acer.edu.au/mailman/options/rasch/trevor.bond%40jcu.edu.au
More information about the Rasch