earticle

영어

The purpose of this study was to investigate the syntactic text variables related to the item difficulties of CSAT through the indices in the Coh-metrix, the computational tool, using the 299 reading texts from the 2014 to 2018 CSAT and mock CSAT. The 299 texts of CSAT and mock CSAT from The Korea Institute for Curriculum and Evaluation were made into corpora and the percentage of correct answers were obtained from EBSi website. These texts were analyzed by Coh-Metrix, a linguistic computational program, and then, the correlation analysis was conducted to find out what syntactic text variables have influence on the item difficulties. The results of this study are as follows. First, the more sentences with syntactic similarity in the passage, and the longer the distance between new words, the higher percentage of correct answers. Second, frequencies of negative expression was negatively correlated with the percentage of correct answers. Although there are many variables having an effect on the item difficulty and the rate of the correct answer, the objective figures of the text variables offered by linguistic devices is expected to be useful in successful predicting the item difficulty of the reading part in CSAT.