[Rasch] A FACETS problem

Mike Linacre (RMT) rmt at rasch.org
Wed Jan 27 10:11:02 EST 2010


That sounds good to me. Your situation appears to be a practical 
application of Occam's Razor. There is no need to perform a complex 
(error-prone, difficult-to-interpret) analysis if a series of simple 
(robust, easy-to-verify) analyses will accomplish your objectives.

Mike Linacre

At 1/26/2010, Petroski, Greg wrote:

>Hi Mike,
>Thank you for the reply.  My feeling right now is that FACETS is 
>mind-boggling complicated, or at least requires more than I am willing to 
>invest, and I am not sure it really gets at inferences I am interested in.
>A more simple approach would be to calibrate the items for each subscale 
>using data from all staff types and time points, get a measures for each 
>individual respondent, and then take this as the input for the analysis of 
>variance.  I sure there are holes in this plan so invite objections.

Please consider the environment before you print

More information about the Rasch mailing list