|
|
IRT Model: Estimating Group-level Abi... |
|
Message/Author |
|
Bob Houchens posted on Wednesday, September 03, 2008 - 4:47 pm
|
|
|
Can anyone refer me to an article or an example that explains how to specify and estimate group-level "ability" from two-level data? My data consist of patient scores on 4 binary items (quality scores), and patients are nested within physicians. Each patient can respond to 1, 2, 3, or 4 items. I want to estimate a parameter for each physician that estimates the latent quality for that physician. |
|
|
I don't know about an article off hand. Multilevel IRT would be a good Google term. In Mplus this is handled in line with UG ex 9.7 (simplified to having no covariates), where you can request estimation of factor scores for the group-level "fb" factor. The different numbers of items responded to is handled via missing data. |
|
Bob Houchens posted on Thursday, September 04, 2008 - 12:39 pm
|
|
|
Thanks. I'll look at ex. 9.7. |
|
Back to top |
|
|