Few writing problems are as common, or as frustrating to grade, as the dropped quotation. In essays about Mountains Beyond Mountains, students often paste a sentence from Kidder's text into a paragraph, add no explanation, and assume the evidence speaks for itself. Teachers respond by writing "explain this" in the margin, again and again, with limited effect. A more structured approach to evidence feedback helps students understand what is missing and how to fix it.

Evidence in a nonfiction essay serves a specific function: it supports a claim the student has made. When a student quotes a passage about Farmer insisting on treatment for a patient others considered a lost cause, the quotation should connect to an argument about his priorities or the cost of his approach. If the connection is not stated, the reader must guess, and graders are rarely generous enough to do the student's thinking for them. Feedback should make this connection explicit.

Nonfiction books also create a distinctive evidence problem, because students must decide whether a detail comes from Kidder's narration, a person he interviewed, or Farmer himself. Mixing these sources can lead to unfair claims, such as attributing a colleague's criticism to Kidder's own opinion. Teaching students to attribute carefully is part of teaching them to read nonfiction responsibly. Feedback can model this by asking who is speaking in the quoted passage.

Diagnosing Evidence Problems Quickly

Most evidence issues fall into a handful of categories, and recognizing them speeds up grading. A quotation may be too long, forcing the reader to hunt for the relevant part. It may be too general to support a specific claim, or it may be accurate but irrelevant to the argument. Sometimes it is simply never explained. Labeling the problem precisely lets the student see exactly what to fix instead of feeling that the essay was judged vaguely.

• The dropped quotation that appears with no introduction and no explanation afterward.
• The overlong block quote that buries the relevant phrase in several sentences of background.
• The mismatched quotation that is accurate but does not actually support the claim being made.
• The vague paraphrase that mentions a scene without naming where or when it occurs in the book.
• The unattributed claim that blends Kidder's narration with the opinions of people he interviewed.

Writing Comments That Teach the Skill

A comment that teaches should show students a pattern they can reuse. For instance, you might suggest a three-part structure: introduce the quotation with context, present it, and follow it with one or two sentences explaining how it supports the claim. Showing a short example drawn from the student's own paragraph makes the advice concrete. Students can then apply the pattern to the rest of the essay without needing a comment on every sentence.

It also helps to praise strong evidence use when you see it, and to be specific about why it works. Telling a student that their short, well-chosen phrase from the chapter on tuberculosis in Peru supported the claim precisely, and that the following sentence explained why, reinforces the habit. Positive examples are easier to imitate than abstract advice. A few such comments per essay balance the corrective feedback.

Where AI Helps With Evidence Feedback

AI grading tools can scan an essay and identify each quotation, check whether it is introduced and followed by explanation, and flag those that are not. This is repetitive work for a human grader but straightforward for a tool applying a rubric. The generated comments can reference specific quotations from the student's paper, which makes them feel individual rather than generic. Teachers can review and adjust them in a fraction of the time it would take to write them from scratch.

There are limits, of course. A tool may not know whether a quotation is accurate to the book unless it has access to the text, and it cannot judge whether a student's interpretation is fair to the author's intent. Teachers should treat flagged evidence problems as prompts for review, not final verdicts. Used this way, AI handles the mechanical identification and the teacher handles the interpretive judgment.

Building Evidence Skills Across the Unit

Evidence skills improve through repetition, so consider short practice tasks before the full essay. Ask students to write a single paragraph with one quotation and a two-sentence explanation, then exchange paragraphs and use the rubric to give feedback. These low-stakes exercises reveal problems early and give students a chance to fix them before the high-stakes grade. They also generate useful samples for class discussion.

By the time the full essay is due, students who have practiced will produce noticeably stronger evidence paragraphs, and grading becomes faster because fewer papers share the same basic flaw. The time invested in teaching the skill is repaid in grading efficiency. More importantly, students leave the unit with a transferable ability to support claims in any subject. That outcome is worth more than any single score.