A Text Difficulty Analysis of English Textbooks for Senior High School and the College Scholastic Ability Test in South Korea Using Coh-Metrix

Abstract

• This study aimed to examine the text difficulty of English textbooks for senior high school and the English reading section of the College Scholastic Ability Test (CSAT) in South Korea. By analyzing the degree of their text difficulty, this study attempted to predict how effectively public education can prepare for the CSAT. Five types of high school English reading and writing textbooks representing the new 2015 curriculum, as well as reading passages from the 2019-2020 CSAT English section, were chosen to create a corpus. The text difficulty was compared with linguistic features measured from Coh-Metrix, a computational language analysis tool on: descriptive indices, word information, lexical diversity, syntactic complexity, standard readability, and cohesion. An independent sample t-test is employed to determine whether there are any statistically significant differences in terms of linguistic aspects between the textbooks and the CSAT. In conclusion, except for the syntactic complexity and semantic cohesion, textbooks and the CSAT indicate statistically significant differences in 10 of the 13 Coh-Metrix measurements: the number of syllables in words, the mean number of words in sentences, age of acquisition, word frequency, concreteness, imaginability, type-token ratio, Flesch Reading Ease, Flesch-Kincaid Grade Level, and co-referential cohesion. This finding implies that students who follow the school curriculum and rely on textbooks would have difficulty achieving their CSAT goals. A balance between the difficulty of textbooks and the CSAT should be established to accomplish the CSAT's purpose of normalizing public education and relieving the strain of excessive learning.