University of Twente Student Theses

Login

IRT in Item Banking, Study of DIF Items and Test Construction : Item Response Theory in Item Banking, Study of Differentially Functioning Items and Test Construction

Tshering, Gembo (2006) IRT in Item Banking, Study of DIF Items and Test Construction : Item Response Theory in Item Banking, Study of Differentially Functioning Items and Test Construction.

[img]
Preview
PDF
1MB
Abstract:Classical Test Theory (CTT) had been a sole principle of measurement for the measurement specialists until 1970s. In the decades of the 1970s, Item Response Theory (IRT) began to pick up momentum among the measurement specialists. The weaknesses inherent in CTT are readily redressed by IRT. However, CTT is still used by the measurement specialists due to its simplicity compared to IRT in terms of both mathematical complexity and requirement of special computer packages. Where expertise and special computer packages are available, IRT is always a first choice. The existence of local independence and invariance of item parameters across examinee groups makes IRT superior to CTT. The limitations due to lack of adequate statistics for studying differential item functioning and item bias and incapability of producing tailored tests naturally make CTT inferior to IRT. CTT upsets the measurement specialists when they are confronted with the task of building item banks. The item parameters obtained by using CTT change when the group of examinees changes. For instance, the items are easy when examinees are able and items are difficult when examinees are not able, meaning that the difficulty of the ability being measured change as the examinees change. This is undesirable because ability should maintain its level of complexity irrespective of where, how, when and who are tested on it. Item banking is often used as a quick and efficient means to test as and when desired and this requires the items in the item bank to measure the abilities they are supposed to measure with the same level of complexity across different examinations. Once again CTT fails to serve the measurement specialists in dealing with item banking. Despite the complexity of expertise involved in it, IRT is widely applied in making paper and pencil tests, studying differential item functioning , item banking and computer adaptive testing (CAT). I have been very lucky to have the opportunity to put the principles of CTT and IRT into practice through a project at the National Institute of Educational Measurement, CITO, Arnhem, the Netherlands.
Item Type:Essay (Master)
Clients:
National Institute of Educational Measurement, CITO
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:05 communication studies
Programme:Educational Science and Technology MSc (60023)
Link to this item:http://purl.utwente.nl/essays/773
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page