Abstract
The main purpose of this study is to examine how valid self-assessment (SA) of speech performance can be in a high school setting in Japan. Fukazawa (2009) studied the validity of peer assessment (PA) and found that PA has reasonable degree of validity to support teacher assessment (TA). In the current study, another alternative assessment, SA, is the focus. Fifty-two students and three teachers participated in this study. The validity of SA was investigated from five aspects - the content, substantial, structural, external, and consequential aspect - according to Messick's (1996) concept of validity. The validity of SA in this study means, to be more precise, the validity of interpretation and use based on self-assessment scores. The content and substantial aspects of validity were examined through analyzing a questionnaire, and both aspects were considered to have sufficient validity. The structural aspect was analyzed using Multi-faceted Rasch Measurement, and this aspect of validity was partially supported. In order to examine the external aspect of validity, the scores of SA were compared with that of TA in terms of correlations and mean scores. The result showed insufficient validity from the external aspect. The consequent aspect of validity was investigated analyzing students' comments on SA, and some positive effects were found but validity of this aspect was not sufficient. According to these results, it was concluded that the validity of SA was shown to some extent but not sufficient to use as a part of formal assessment. Pedagogical implications are also suggested.
