The comments are from this presentation )(http://www.stata.com/meeting/s...
Conference San Diego 2012. There were some errors related to the discussion of the 2PL and 3PL model as well as some omissions in terms of tools.
This article is easy obtained via search so it is a good idea to fix these errors to avoid accidentally confusing users.
Slide 12: (the labeling of a parameter as 2nd or 3rd does not change the model,, but most who use IRT think of the parameters as described below).
Two parameter logistic (2PL) model has a second parameter account for discrimination
Three parameter logistic (3PL) has an additional 3rd parameter,
referred to as guessing to accounts for a non-zero lower asymptote of
the item response curve
Slide 13:
Also, Gllamm as shown is Stata Journal Article
http://www.stata-journal.com/a...
I hope this helps!
Leah
Research Scientist, NWEA