[Rasch] Model Standard Errors
david.andrich at uwa.edu.au
Thu Mar 12 18:49:41 EST 2009
Colleagues. Just to let you know that the online course that is running from
July to November this year and which I mentioned a few weeks ago, has the
David Andrich, BSc., MEd. (UWA), PhD (Chic), FASSA
Chapple Chair, Graduate School of Education
The University of Western Australia
35 Stirling Highway
Telephone: +61 (0) 8 6488 1085
Fax: +61 (0) 8 6488 1052
Email: david.andrich at uwa.edu.au
From: rasch-bounces at acer.edu.au [mailto:rasch-bounces at acer.edu.au] On Behalf
Of Mike Linacre (RMT)
Sent: Friday, 23 January 2009 4:19 PM
To: rasch at acer.edu.au
Subject: Re: [Rasch] Model Standard Errors
Thank you for your question, Jun Kim.
In bias/DIF analysis (and in statistics in general), the standard error is
dominated by the number of observations. The more observations, the smaller
the standard error. Perhaps you do not have enough data for the analysis you
want to perform.
At 1/22/2009, you wrote:
My concern is that the magnitude of model standard error in the
bias/interaction report is relatively big across interactions. How do I
interpret that? What should I do if the model s.e. is big?
Editor, Rasch Measurement Transactions
rmt at rasch.org <http://www.rasch.org/rmt/> www.rasch.org/rmt/ Latest RMT:
22:2 Autumn 2008
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Rasch