1
$\begingroup$

I have 500,000 students dichotomy responses for a 45 item test. I know the ability of the this students, that this ability was calculated from this responses using 3PL model by Expected a Posteriori (EAP) and I know the prior mean and standard deviation used. Having this information it's possible in R to know or at least get a approximation of the 3 parameters of each items that were used to calculated the students ability?

I am not interested in recalculating the value of the parameters, but rather to know the values of the parameters used by those who prepared the test.

Improve this question
$\endgroup$

1 Answer 1

Reset to default
2
$\begingroup$

I don't think there is an easy way to calculate them. You can only estimate them. Assuming that the $\theta$ is known, you can write a function to optimize the loglikelihood function for a given response vector of each item.

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.