Messages

1

Questions and Answers / sufficient statistic

« on: May 16, 2012, 05:07:16 AM »

hi
When we use of latent regression analysis and use of some variables such as sex or age in population model then vector of raw scores isn’t sufficient statistic for vector of theta estimates.
Is it true?
I wonder if I can ask you to help me.
Best regards
E.sanei

--------------------------------------------------------------------------------

2

Questions and Answers / interpret

« on: May 13, 2012, 07:22:13 AM »

Hi

I have two questions, I wonder if I can ask you to help me.

1.      In EAP estimates for theta, if I have one dimension, software gives me four columns. The First column is individual number and the second column is EAP estimate, I don’t know third or fourth column is standard error of measurement.

2.      I don’t know in what manner I can interpret thetas, for example if person A has theta=0 and person B has theta=1, who can I compare them?

Best regards

E.sanei


--------------------------------------------------------------------------------

3

Questions and Answers / Re: confidence interval

« on: May 08, 2012, 07:12:52 AM »

you are wright.
thank you so much for your help.

4

Questions and Answers / confidence interval

« on: May 08, 2012, 06:43:57 AM »

Dear professor

For the MNSQ fit statistics you provide a ninety-five percent confidence interval for the expected value of the MNSQ in ConQuest. Is it possible I provide confidence intervals with other percent?

I need a confidence interval that equal to (0.7 , 1.3).

I wonder if I can ask you to help me.

Best regards

E.sanei


--------------------------------------------------------------------------------