Introduction: Readability formulas are crucial for identifying suitable texts for children’s reading development. Traditional formulas, however, are linear models designed for alphabetic languages and struggle with numerous predictors. Purpose: To develop advanced readability formulas for Chinese texts using machine-learning algorithms that can handle hundreds of predictors. It is also the first readability formula developed in Hong Kong. Method: The corpus comprised 723 texts from 72 Chinese language arts textbooks used in public primary schools. The study considered 274 linguistic features at the character, word, syntax, and discourse levels as predictor variables. The outcome variables were the publisher-assigned semester scale and the teacher-rated readability level. Fifteen combinations of linguistic features were trained using Support Vector Machine (SVM) and Random Forest (RF) algorithms. Model performance was evaluated by prediction accuracy and the mean absolute error between predicted and actual readability. For both publisher-assigned and teacher-rated readability, the all-level-feature-RF and character-level-feature-RF models performed the best. The top 10 predictive features of the two optimal models were analyzed. Results: Among the publisher-assigned and subjective readability measures, the all-RF and character-RF models performed the best. The feature importance analyses of these two optimal models highlight the significance of character learning sequences, character frequency, and word frequency in estimating text readability in the Chinese context of Hong Kong. In addition, the findings suggest that publishers might rely on diverse information sources to assign semesters, whereas teachers likely prefer to utilize indices that can be directly derived from the texts themselves to gauge readability levels. Conclusion: The findings highlight the importance of character-level features, particularly the timing of a character’s introduction in the textbook

Purpose: This study adopted a mixed-method approach, including a classroom experiment and 24 in-depth interviews, to investigate the effects of two feedback techniques (coded focused and unfocused written corrective feedback) on ESL learners' writing in a self-financed tertiary institution in Hong Kong. Methodology: Three intact classes of 47 students served as the experimental and control groups; the control group only received feedback on content and organization, whereas the two experimental groups also received focused and unfocused linguistic feedback, respectively. The feedback intervention was conducted over an eight-week intensive summer course, focusing on three grammar errors (articles, singular/plural nouns and verb forms). Altogether, students wrote seven pieces, four of which were analysed for the present research. Results: The study found that students who received focused written corrective feedback (WCF) significantly outperformed the other two groups, though the effects varied across error types. Meanwhile, no significant differences were found between the unfocused and control groups. In-depth interviews explored how individual learners' metalinguistic understanding and engagement affect their intake of WCF. The results revealed that learners who received focused feedback developed a deeper understanding of the linguistic nature of specific error types. Learners with limited English proficiency were less likely to apply their linguistic knowledge to revise a task or write a new one. Conclusion: Because not all errors deserve equal attention, teachers and students should consider how feedback can be used more effectively, particularly in areas where comprehensive feedback is considered obligatory. When teaching students with limited language proficiency, it is recommended that, rather than providing a wide range of error corrections, teachers provide focused feedback complemented with carefully designed metalinguistic support. Copyright © 2022 National