خلاصة:
Due to deficiencies of the traditional models of standard setting, thisstudy intends to suggest a new method for setting standards employingRasch measurement. Precise and efficient methods for settingperformance standards and linking tests to ability scales is a much-feltneed in today's educational contexts. The introduction of the CommonEuropean Framework of Reference as a common paradigm for languageteaching and assessment stressed the need for such methods. Thesuggested method combines the classic test-centered method of standardsetting with the probabilistic properties of the Rasch model to set severalcut points on the ability continuum. The Wright map which jointlydepicts the difficulty location of items and the ability location of personson a common scale is the cornerstone of this method
ملخص الجهاز:
The authors of this paper believe that Rasch measurement due to its capacity to measure both person ability and item difficulty on a single common scale can be a good method for setting cut-points.
By the same token, Hambleton and Eignor (1978) relate the standard setting to the concept of ‘competency’ and define standard or as they call it ‘a minimum proficiency level’ as a point on a test score scale that is used to divide testees into two categories, that each of them reflects a distinctive level of proficiency relative to the competency that the test has measured.
Also, considering the importance of an appropriate method for standard setting on a given situation, Jeager (1976); Zieky and Livingston (1977) and Popham (1978) have provided some guidelines for selecting such methods based on the merits of different methods (cited in Hambleton, Powell, & Eignor, 1979).
Rasch model, which is often called as a ‘latent trait’ model, has this underlying assumption that views each test continuum as a latent trait upon which learners, items and also criterion levels of ability that is standards can be located (Jones & Saville, 2008).
The Proposed Method The probabilistic properties of the Rasch model when combined with test-centred method of standard setting along with the visual facilities that the Wright map provides can offer great help in setting cut-off scores to indicate different levels of performance.
A comparison of two different methods for setting performance standards for a test with constructed-response items, Journal of practical assessment, Research and Evaluation, 13, 1-12.