[Rasch] compare nested models in Facets

Iasonas Lamprianou liasonas at cytanet.com.cy
Wed Dec 21 17:44:33 EST 2011


Dear friends, 

I am using Facets, but not the latest version of Windows Facets. As a result, my Facets does not produce information like the following.

Data log-likelihood chi-square = 270479.6875
Approximate degrees of freedom = 62343
Chi-square significance prob.  = .0000
                                         Count   Mean   S.D.   Params
Responses used for estimation      =     73358   7.40   2.73    11015
Responses in one extreme score     =        15   0.13   0.50        5
All Responses                      =     73373   7.40   2.73    11020
Count of measurable responses      =     73373
Raw-score variance of observations   =   7.46 100.00%
Variance explained by Rasch measures =   4.58  61.35%
Variance of residuals                =   2.89  38.65%

However, I would like to be able to produce this information because it is useful  for my analyses. How can I do it without buying the new Facets?

The reason I need this, is to do the following: I run a model with 4 facets and then I used the same data but removed one facet from the analysis. Presumably, if I deduct the "Data log-likelihood chi-square" of the second model from that of the second, then this gives me the improvement of the model fit for df1-df2 where the df is the "Approximate degrees of freedom" for each model. Then, I assume that this is a chi-square with df1-df2 degrees of freedom in order to test if I should actually model the 4th facet or not. Is this procedure correct to test the two (nested?) models? Do we havea  reference for this?

Thank you for your support 

Iasonas 




More information about the Rasch mailing list