Literaturnachweis - Detailanzeige
Autor/in | Rudner, Lawrence M. |
---|---|
Titel | A Short and Simple Introduction to Tailored Testing. |
Quelle | (1978), (15 Seiten)
PDF als Volltext |
Sprache | englisch |
Dokumenttyp | gedruckt; online; Monographie |
Schlagwörter | Ability; Adaptive Testing; Bayesian Statistics; Computer Assisted Testing; Individual Testing; Item Analysis; Item Banks; Mathematical Models; Test Construction; Test Items; Test Length; Testing Problems |
Abstract | Tailored testing provides the same information as group-administered standardized tests, but can do so using fewer items because the items administered are selected for the ability of the individual student. Thus, tailored testing offers several advantages over traditional methods. Because individual tailored tests are not timed, anxiety is reduced and examinee motivation is improved. Economic advantages involve reduced test time, immediate availability of results, and reduced personnel requirements. Effective tailoring occurs at the item level, involving two steps: estimation of the examinee's ability from his or her previous responses, and selection from an item bank of the item likely to measure most effectively. Five methods of estimating item difficulty or appropriateness are: (1) Robbins Monro procedure; (2) fixed step size; (3) flexilevel, which requires a smaller item pool; (4) Bayesian procedures; and (5) stratified-adaptive or stradaptive procedures. (Author/GDC) |
Erfasst von | ERIC (Education Resources Information Center), Washington, DC |