Effects of Multidimensionality on IRT Item Characteristics and True Score Estimates: Implications for Computerized Test Assembly. Computerized Testing Report. LSAC Research Report Series.
Saved in:
| Title: | Effects of Multidimensionality on IRT Item Characteristics and True Score Estimates: Implications for Computerized Test Assembly. Computerized Testing Report. LSAC Research Report Series. |
|---|---|
| Language: | English |
| Authors: | Wang, Xiang-Bo, Harris, Vincent, Roussos, Louis, Law School Admission Council, Newtown, PA. |
| Peer Reviewed: | N |
| Page Count: | 23 |
| Publication Date: | 2002 |
| Document Type: | Numerical/Quantitative Data Reports - Research |
| Descriptors: | Ability, College Entrance Examinations, Computer Assisted Testing, Estimation (Mathematics), Item Response Theory, Law Schools, Test Construction, Test Items, True Scores |
| Assessment and Survey Identifiers: | Law School Admission Test |
| Abstract: | Multidimensionality is known to affect the accuracy of item parameter and ability estimations, which subsequently influences the computation of item characteristic curves (ICCs) and true scores. By judiciously combining sections of a Law School Admission Test (LSAT), 11 sections of varying degrees of uni- and multidimensional structures are used to assess the impact of multidimensionality on ICCs and true scores. This combination makes an artificial test for analysis purposes. It was found that multidimensionality decreased item slopes and estimated true score curves. The effect is shown to be bigger in the high ability range than in the low ability range. (Contains 23 figures and 21 references.) (Author/SLD) |
| Entry Date: | 2003 |
| Accession Number: | ED469184 |
| Database: | ERIC |
| Abstract: | Multidimensionality is known to affect the accuracy of item parameter and ability estimations, which subsequently influences the computation of item characteristic curves (ICCs) and true scores. By judiciously combining sections of a Law School Admission Test (LSAT), 11 sections of varying degrees of uni- and multidimensional structures are used to assess the impact of multidimensionality on ICCs and true scores. This combination makes an artificial test for analysis purposes. It was found that multidimensionality decreased item slopes and estimated true score curves. The effect is shown to be bigger in the high ability range than in the low ability range. (Contains 23 figures and 21 references.) (Author/SLD) |
|---|
