1
$\begingroup$

To my R angels,

After a recent voyage of discovery, I have been using IRT to develop a set of GRM derived item parameters on a large dataset of 40000 questionnaire responses.

I would like to use these item parameters to estimate theta (ability) of the whole dataset. Thereafter, I would like to use the same parameters to estimate theta on a subset of items.

I've been trying to use the thetaEst function in CatR.

thetaEst(it, x)

Where "it" is a matrix of my item parameters. "x" is a vector of item responses.

As a vector, I can calculate a single participants theta estimate.

I would like to expand x from a single vector, to be able to input a matrix of 40000 responses.

I can't see how this is possible - any help would be very gratefully received.

Jon

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

Why don't use the mirt package to fit the Graded Response Model? Then, you can use the fscores() function from the mirt package to extract the person parameters for all individuals.

Another option is the 1factor.scores()1 function from the ltm package after fitting the model using the grm() function.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.