Academic Journals Database
Disseminating quality controlled scientific knowledge

Designing item pools to optimize the functioning of a computerized adaptive test

ADD TO MY LIST
 
Author(s): Mark D. Reckase

Journal: Psychological Test and Assessment Modeling
ISSN 2190-0493

Volume: 52;
Issue: 2;
Start page: 127;
Date: 2010;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: computerized adaptive testing | item pool design | item response theory | conditional standard error of measurement | item pool size

ABSTRACT
Computerized adaptive testing (CAT) is a testing procedure that can result in improved precision for a specified test length or reduced test length with no loss of precision. However, these attractive psychometric features of CATs are only achieved if appropriate test items are available for administration. This set of test items is commonly called an “item pool.” This paper discusses the optimal characteristics for an item pool that will lead to the desired properties for a CAT. Then, a procedure is described for designing the statistical characteristics of the item parameters for an optimal item pool within an item response theory framework. Because true optimality is impractical, methods for achieving practical approximations to optimality are described. The results of this approach are shown for an operational testing program including comparisons to the results from the item pool currently used in that testing program.
Affiliate Program      Why do you need a reservation system?