This study investigated the relationship between the linguistic features of English essays and the holistic evaluations conducted by non-native Korean teachers of English, while also examining the applicability of automated essay scoring. A total of 23 essays written by second- and third-year high school students were analyzed using Coh-Metrix 3.0, with particular attention to indices of lexical diversity, syntactic complexity, cohesion, and readability. Spearman’s rank-order correlation analysis was performed using SPSS to examine the relationships between the linguistic indices and the teachers’ evaluation rankings. The results revealed that lexical diversity (LDTTRa) demonstrated a relatively strong positive correlation with teachers’ evaluations, whereas indices related to syntactic complexity, cohesion, and readability exhibited only limited correlations. These findings suggest that lexical features may play a significant role in teachers’ holistic judgments of essay quality. Furthermore, the study indicates that Coh-Metrix-based automated scoring may serve as a supplementary tool to enhance the objectivity and reliability of essay assessment, rather than functioning as a replacement for human raters.