Displaying 1 result from an estimated 1 matches for "dichotmously".
Did you mean:
dichotomously
2007 Dec 12
0
IRT Likelihood problem
...items are scored
dichotomously (x_{ij} = 0 or 1) and p(x) is Muraki's generalized partial
credit model when items are scored polytomously (x_{ij} = 0 \ldots J).
Now, I wrote the following two functions to maximize the likelihood. The
first one is for the 3PL and works when all items are scored
dichotmously. The second is for the GPCM and gives the MLE when all
items are scored polytomously.
In the code below, this requires that I first have in hand estimates of
the item parameters so that the maximization is only w.r.t. theta.
# x = response pattern
# b = vector of location parameters
# a = vector...