英文摘要 |
The purpose of this study is to conduct standard setting in Chinese for the 2010 Taiwan Assessment of Student Achievement (TASA) of grade eight. The specific definitions of the basic, proficient, and advanced achievement levels are defined by the subject matter experts. The nineteen panelists coming from north, middle, south, and east areas of Taiwan are invited to conduct the standard setting procedures with Yes/No Angoff method. This study evaluates the internal evidence for validity using the extreme value method, standard setting consistency within and between panelists. Such internal evidence then reports that the results of the standard settings in this study are consistent. Meanwhile, we also evaluate the procedural evidence of validity in the choice and execution of the standard settings which collect the assessment of panel meeting, the recruitment of panelists, standard setting methods, and the impact of feedbacks. And the procedural validity of the standard setting in Chinese for grade eight performs quite well as an entirety. Finally, the external evidence of validity is collected from the consistency between the cut-scores setting and the classification. Yes/No Angoff’s method may tend to set low cut-scores for the basic level while it poses high cut-scores for the advanced level. Based on the Yes/No Angoff method, its classification would be consistent with the three other methods. Based on the internal, procedural, and external evidences, the standard setting procedures for the 2010 TASA of gradeeight has its validity. This study shall provide a great reference for those who wish to conduct standard setting methods into practical use. |