Retrospective Study of English Achievement at a High School Using an IRT Scale

Abstract

Item response theory (IRT) makes it possible to compare the academic achievement obtained from different versions of tests on a common scale. There have been few studies to compare scores from large-scale academic achievement tests administered in schools using IRT in Japan. Therefore, in order to compare scores on a common scale, two kinds of research were undertaken for English achievement tests administered in the past (Saida, 2003; Saida and Hattori, 2004). The purpose of this research was to investigate trends of English achievement at one high school selected from the previous two studies. The English IRT scale scores of 6498 high school students attending the school from 1995 to 2002 were investigated in terms of courses and grade levels. In addition, the trends in English achievement from first to third grade levels were examined. This research found that the overall trends in English achievement at this high school declined, mirroring the general prefectural decline (Saida, 2003). The present study demonstrates the advantages of applying IRT in tracking trends in academic achievement and for school assessment.