This study examines the linguistic characteristics and the text difficulty of the National Assessment of Academic Achievement (NAAA). First, the linguistic features of NAAA conducted in June, September, and November were compared to examine the difference among the tests by the scope of the assessment. Second, the linguistic features of NAAA and the reading texts in six high school English textbooks were compared. Lastly, the correlation between the linguistic features of NAAA and the correct answer rate for each question was analyzed. Corpus analysis was conducted by using Coh-Metrix 3.0. The indices were selected for descriptive analysis, lexical diversity, connectives, syntactic complexity, and readability. One-way ANOVA, Kruskal-Wallis H test, independent t-test, Mann-Whitney U test, and correlation analysis were conducted to investigate the Coh-Metrix data. The result revealed the statistical differences among NAAA tests held in June, September, and November regarding the differences in ‘mean number of words of sentences and readability score.’ In addition, discrepancies in linguistic features between the text in NAAA and the English textbooks were found. Lastly, it was found that ‘number of words, mean number of words, CELEX Log frequency for all words, argument overlap, and readability score' had a correlation with the correct answer rate. Implications and the limitations of the study were discussed.