Keats writes with archaic vocabulary, inverted syntax, and compressed imagery, and even fluent readers stumble over it. For English language learners, a poem from the 1817 collection can be a double challenge: decoding the language and then writing an analysis in a second language. Grading these essays fairly requires separating what the student understands from how well they can express it in English. That separation is possible, but it takes deliberate planning.

The first step is to decide which skills the assignment is actually assessing. If the goal is close reading and interpretation, then grammatical errors should carry less weight than the quality of the argument. Stating this clearly in the rubric, and telling students, reduces anxiety and helps them focus on ideas instead of worrying about every sentence.

It also helps to provide scaffolds before the assignment. Glossaries of difficult words, paraphrased versions of especially dense lines, and audio recordings of the poem give ELL students access to the text without lowering expectations for analysis. These supports level the playing field and make the assessment more accurate.

Separate language accuracy from content in your rubric

A rubric that assigns a separate score to language accuracy makes it possible to reward strong analysis even when the writing contains errors. This approach also gives you a cleaner picture of where the student needs support. Some students have sophisticated ideas and limited vocabulary, while others write fluently but struggle to develop an argument.

• Score interpretation and evidence independently from grammar
• Note errors that interfere with meaning separately from minor slips
• Accept sentence frames and graphic organizers as legitimate scaffolds
• Allow use of a bilingual dictionary during writing when appropriate
• Give feedback on two or three language patterns at a time, not every error

Give feedback on language selectively

Correcting every error in an ELL student's essay is discouraging and rarely effective. A better approach is to identify patterns, such as verb tense or article use, and address one or two at a time. Students can absorb and apply a small number of corrections, and they gradually build accuracy without feeling overwhelmed.

Explaining errors, rather than only marking them, makes the feedback more useful. A brief note about why a sentence is unclear, along with a model revision, teaches a principle the student can apply elsewhere. Over time, this approach produces more improvement than red ink alone.

Recognize strengths that are easy to overlook

Multilingual students often bring insights that monolingual readers miss, such as sensitivity to word origins, sound patterns, or translation. A student who notices that a word in Keats has a cognate in their first language may make a genuinely interesting point about connotation. Watching for these strengths and rewarding them encourages students to draw on their full linguistic resources.

Feedback should also acknowledge growth over time. A student who moves from summary to analysis while still making grammar errors has made real progress, and it deserves to be named. Highlighting that growth builds confidence and motivates continued effort.

Keep the workload sustainable

Giving thoughtful feedback to ELL students takes time, which many teachers do not have in abundance. Comment banks for common language issues, with clear explanations and examples, help you respond efficiently. Adapting each comment to the student's specific error keeps it relevant.

AI-assisted tools can support this work by identifying recurring language patterns and drafting explanations you can review. They should never replace the teacher's judgment about what a student understands, but they can speed up the identification of issues. That saved time can be redirected toward conferences and individual support.