NotesFAQContact Us
Collection
Advanced
Search Tips
Peer reviewed Peer reviewed
Direct linkDirect link
ERIC Number: EJ859864
Record Type: Journal
Publication Date: 2009-Sep
Pages: 25
Abstractor: As Provided
Reference Count: 69
ISBN: N/A
ISSN: ISSN-1082-989X
Estimation of IRT Graded Response Models: Limited versus Full Information Methods
Forero, Carlos G.; Maydeu-Olivares, Alberto
Psychological Methods, v14 n3 p275-299 Sep 2009
The performance of parameter estimates and standard errors in estimating F. Samejima's graded response model was examined across 324 conditions. Full information maximum likelihood (FIML) was compared with a 3-stage estimator for categorical item factor analysis (CIFA) when the unweighted least squares method was used in CIFA's third stage. CIFA is much faster in estimating multidimensional models, particularly with correlated dimensions. Overall, CIFA yields slightly more accurate parameter estimates, and FIML yields slightly more accurate standard errors. Yet, across most conditions, differences between methods are negligible. FIML is the best election in small sample sizes (200 observations). CIFA is the best election in larger samples (on computational grounds). Both methods failed in a number of conditions, most of which involved 200 observations, few indicators per dimension, highly skewed items, or low factor loadings. These conditions are to be avoided in applications. (Contains 2 figures, 9 tables, and 20 footnotes.)
American Psychological Association. Journals Department, 750 First Street NE, Washington, DC 20002-4242. Tel: 800-374-2721; Tel: 202-336-5510; Fax: 202-336-5502; e-mail: order@apa.org; Web site: http://www.apa.org/publications
Publication Type: Journal Articles; Reports - Research
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A