Assessing the Accuracy of Automated Writing Evaluation in Predicting English Language Arts Proficiency for Middle-Grade English Language Learners and Non-English Language Learners
Zhang, F., & Wilson, J. (2025). Assessing the accuracy of automated writing evaluation in predicting English language arts proficiency for middle-grade English language learners and non-English language learners. Language Teaching Research Quarterly, 51, 247-272. https://doi.org/10.32038/ltrq.2025.51.04
This study examined how well fall, winter, and spring MI Write benchmark writing scores predicted non-proficiency on the Smarter Balanced ELA assessment for middle-grade students. Using ROC curve analyses, the authors compared classification accuracy across grade levels and language status for English learners and non-English learners.
Key takeaways
- MI Write showed acceptable overall classification accuracy for predicting ELA non-proficiency.
- Accuracy was strongest for non-English learners and grade 7 students.
- Classification accuracy was more inconsistent for English learners.
- D-based cutpoints provided the best balance of sensitivity and specificity across grades and subgroups.
Why it matters
Provides validity evidence for using an automated writing evaluation system as a middle-school universal screening tool, with attention to subgroup differences and practical classification accuracy.
Coauthors
Fan Zhang, Joshua Wilson
Topic tags
External links