Literaturnachweis - Detailanzeige
Autor/inn/en | De Champlain, Andre; Gessaroli, Marc E. |
---|---|
Titel | Assessing the Dimensionality of Item Response Matrices with Small Sample Sizes and Short Test Lengths. |
Quelle | (1996), (33 Seiten)
PDF als Volltext |
Sprache | englisch |
Dokumenttyp | gedruckt; online; Monographie |
Schlagwörter | Adaptive Testing; Chi Square; Computer Assisted Testing; Factor Analysis; Goodness of Fit; Item Response Theory; Matrices; Sample Size; Simulation; Test Length; Law School Admission Test |
Abstract | The use of indices and statistics based on nonlinear factor analysis (NLFA) has become increasingly popular as a means of assessing the dimensionality of an item response matrix. Although the indices and statistics currently available to the practitioner have been shown to be useful and accurate in many testing situations, few studies have investigated their behavior with small sample sizes and short tests, conditions that are usually encountered with computerized adaptive testing and computerized mastery testing. The purpose of this investigation was to compare the empirical Type I error rates and rejection rates obtained using two NLFA fit statistics with conditions simulated to contain short tests and small sample sizes. The behaviors of an approximate chi-square statistic, the LISREL8 (computer program) chi-square statistic, and the likelihood ratio chi-square difference with unidimensional data sets were examined with simulated data sets of 20 and 40 items and 250, 500, and 1,000 examinees for the Law School Admission Test. Preliminary findings with respect to the approximate chi-square statistic were encouraging in that it appeared to have low Type I error rate and rejection rates that were very high with two-dimensional data sets. The statistic was relatively unaffected by the sample size, test length, and latent trait correlation levels simulated. (Contains 5 tables and 75 references.) (SLD) |
Erfasst von | ERIC (Education Resources Information Center), Washington, DC |