A Paradox between IRT Invariance and Model-Data Fit When Utilizing the One-Parameter and Three-Parameter Models

Autor/inn/en	Custer, Michael; Sharairi, Sid; Yamazaki, Kenji; Signatur, Diane; Swift, David; Frey, Sharon
Titel	A Paradox between IRT Invariance and Model-Data Fit When Utilizing the One-Parameter and Three-Parameter Models
Quelle	(2008), (42 Seiten) PDF als Volltext kostenfreie Datei Verfügbarkeit
Sprache	englisch
Dokumenttyp	gedruckt; online; Monographie
Schlagwörter	Error of Measurement; Item Response Theory; Models; Goodness of Fit; Evaluation Methods; Grade 2; Grade 3; Grade 4; Grade 5; Grade 6; Academic Ability; Correlation; Computation; Age Differences + Suchen Sie Ihr Suchwort? Messfehler; Item-Response-Theorie; Analogiemodell; School year 02; 2. Schuljahr; Schuljahr 02; School year 03; 3. Schuljahr; Schuljahr 03; School year 04; 4. Schuljahr; Schuljahr 04; School year 05; 5. Schuljahr; Schuljahr 05; School year 06; 6. Schuljahr; Schuljahr 06; Korrelation; Age; Difference; Age difference; Altersunterschied
Abstract	The present study compared item and ability invariance as well as model-data fit between the one-parameter (1PL) and three-parameter (3PL) Item Response Theory (IRT) models utilizing real data across five grades; second through sixth as well as simulated data at second, fourth and sixth grade. At each grade, the 1PL and 3PL IRT models were run with each of three ability groups; low, middle and high utilizing PARSCALE Version 4.1. Results were compared in terms of item fit as well as Pearson and Spearman rank-order correlations between estimated item and ability parameters. At each grade, the 3PL exhibited the best model-data fit. However, the 1PL produced a greater degree of item and ability invariance across the three ability groups. (Contains 13 tables.) (Author).
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2017/4/10