Factors Relating to Kanji Reading Accuracy in Kun‐readings: An Item‐Level Analysis

Abstract

AbstractThis study sought to obtain comprehensive data on the difficulty level of kun‐reading (native Japanese reading) of Japanese kanji (Chinese characters adapted for Japanese) learned in junior high and high school. Drawing from an item‐based analysis, it explored the word variables related to the difficulty level of kun‐reading. A reading test comprising 645 kun‐readings of common kanji learned in junior high and high school was administered to 123 university students. Tobit regressions were conducted for the total, okurigana (phonetic kana letters suffix written after kanji), and single kanji items, using the word variables as independent variables and the reading accuracy as the dependent variable. Significant positive contributions were observed for imageability and word frequency for all item categories, kun‐reading ratio for the single kanji items, and okurigana ratio for the total items. A significant negative contribution was observed for okurigana neighborhood size for the okurigana items. The results highlighted the need for education focusing on the meanings of kanji characters, and for a language policy for the use of the okurigana.