Application Of Many Facet Rasch Model On English Writing Assessment

As a comparatively direct measure of learners' communicative language ability, writing test claims test reliability and a strong power for predictive utility of test scores and is gaining increasing interest both in practice and research in language testing. Despite the seemingly direct relationship between the underlying constructs and scores, an effective but more complicated procedure should be followed in test reliability. Guided by this idea, the present thesis aims to give a general overview of justification for the problems involved in test reliability in a writing test context and illustrate how Many-Facet Rasch Model (MFRM) could be adopted as an effective tool in addressing problems in test reliability.As an illustration, the present study investigates and models possible sources of score reliability within the framework of MFRM. Based upon raw scores of a writing test administrated in May of 2007 in DUT, the results demonstrated that the writing test administrated could successfully separate participants into statistically distinct levels of proficiency and facets in the model such as rater, task, rating domains (items) and participants generally function well within certain limits. However, tendency to give central ratings along the scale, less reliability across the five domains and certain patterns of bias in raters' rating pattern are detected. Fair scores for such candidate are also provided, minimizing the reliability due to facets other than participants' ability.MFRM manifests itself as effective in detecting whether each test method facet functions as intended in such performance test setting and providing useful feedback for test improvement.