NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED260094
Record Type: Non-Journal
Publication Date: 1984-Nov
Pages: 23
Abstractor: N/A
Reference Count: N/A
ISBN: N/A
ISSN: N/A
Implementing Full Information Factor Analysis: TESTFACT Program.
Muraki, Eiji
The TESTFACT computer program and full-information factor analysis of test items were used in a computer simulation conducted to correct for the guessing effect. Full-information factor analysis also corrects for omitted items. The present version of TESTFACT handles up to five factors and 150 items. A preliminary smoothing of the tetrachoric correlation coefficient may be needed before the principal factor analysis is carried out; the new matrix is then analyzed by the MINRES method. In the simulation, stepwise full-information factor analysis of one or two factors was carried out on a 25-item data set. One model corrected for guessing and one did not. Item difficulty and chi square statistics were computed for a one-factor and a two-factor solution. Difficult items were most affected by guessing, and the guessing correction raised the tetrachoric correlation coefficients. Goodness of fit statistics were affected by additional factors or dimensionality. It was recommended that researchers using full-information factor analysis choose both the step-wise option and the guessing model in the TESTFACT program. Also the TESTFACT option which imposes constraints on item parameter estimates in maximum likelihood factor analysis should be used to avoid the Heywood case. (GDC)
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Audience: Researchers
Language: English
Sponsor: N/A
Authoring Institution: N/A