NotesFAQContact Us
Collection
Advanced
Search Tips
Peer reviewed Peer reviewed
Direct linkDirect link
ERIC Number: EJ856070
Record Type: Journal
Publication Date: 2009
Pages: 3
Abstractor: ERIC
Reference Count: 9
ISBN: N/A
ISSN: ISSN-1536-6367
Some Observations on the Identification and Interpretation of the 3PL IRT Model
Azevedo, Caio Lucidius Naberezny
Measurement: Interdisciplinary Research and Perspectives, v7 n2 p89-91 2009
The paper by Maris, G., & Bechger, T. (2009) entitled, "On the Interpreting the Model Parameters for the Three Parameter Logistic Model," addressed two important questions concerning the three parameter logistic (3PL) item response theory (IRT) model (and in a broader sense, concerning all IRT models). The first one is related to the model identifiability aspects, and the second is related to model interpretation. This author contends that one of the most important and not completely clarified aspects concerning the using of IRT is model identification. There is no agreement about the most suitable way to identify an IRT model. However, when all parameters are unknown, it seems that to impose restrictions on the latent traits assuming a known (not only the shape, but also the population parameter values) distribution for them, is the approach that has been most often used. Azevedo states that assuming a distribution for the latent traits is a better way to identify the model than to impose restrictions on the item parameters, especially when considering the interpretation of the parameters (items and latent traits).
Psychology Press. Available from: Taylor & Francis, Ltd. 325 Chestnut Street Suite 800, Philadelphia, PA 19106. Tel: 800-354-1420; Fax: 215-625-2940; Web site: http://www.tandf.co.uk/journals
Publication Type: Journal Articles; Opinion Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A