A validation of the scoring of KFL writing assessment based on the many-facet Rasch measurement model
This paper aims to analyze the scores assigned to compositions of Korean as a foreign language (KFL) in terms of the many-facet Rasch measurement (MFRM) model. The MFRM computer program Facets has recently been updated and this research uses it in order to investigate the scoring reliability of KFL writing test. For the research data the Korean compositions written by 270 KFL learners were collected and rated by eleven Korean raters. At first a brief theoretical basis for the Facets is introduced based on the MFRM model. Then focusing on the raters’ characteristics such as rating consistency and fitness in terms of the rating scheme of content/organization and language, the collected scores of KFL compositions are analyzed and the analysis results are interpreted. As a result, a few cases of unreliable scoring of Korean raters are identified, but it appears that the ratings of Korean raters for a standardized KFL writing test are generally reliable. Finally it is suggested that the Facets program of MFRM offer many potential advantages for validating the scoring reliability of Korean writing assessment.