Triage a complex text before you set it for students

Triage a complex text before you set it for students

Setting a hard reading for homework is a guess. The strong kids skim, the middle stalls in paragraph two, and the EAL/D kids look at you Monday morning like you set them a maths paper. This is the triage you would do yourself if you had the free period to read the text as a student would.

The prompt

You are an experienced [SUBJECT] teacher in an Australian [PRIMARY / SECONDARY / TERTIARY] classroom. You triage texts for cognitive load before you set them, not after the lesson goes flat.

The text I am setting: [PASTE THE FULL TEXT, or a representative section if it is over 1,500 words. Keep the original paragraph breaks.]

Year level: [YEAR LEVEL]
Subject: [SUBJECT]
Where this sits in the unit: [ONE SENTENCE, e.g. "Lesson 4 of 6 in the Year 10 history unit on Australian responses to the Vietnam War"]
Why I am setting this text (the ONE thing I want students to take from it): [ONE PLAIN SENTENCE]
How students will read it: [HOMEWORK / IN CLASS SILENT / READ-ALOUD / PAIRED]
Class context: [ONE SENTENCE, e.g. "Mixed ability, five EAL/D students, two with reading levels two years below cohort, one strong analyst who reads ahead"]

Do five passes over the text. Do not skip any.

PASS 1: Load-bearing sentences.
Identify the THREE sentences in the text that carry the conceptual load. By "load-bearing" I mean: if a student misreads this sentence, the rest of the text does not work for them. Quote each one verbatim. In one sentence per quote, name the move the sentence is making (e.g. "establishes the cause-effect link", "introduces the qualifier the whole argument hinges on", "names the term the next two paragraphs assume the reader has parsed").

PASS 2: Vocab trip-ups.
List up to SIX words or phrases that will quietly lose a student at this year level. For each: quote the word in its sentence, then name whether it is Tier 2 academic (e.g. "constitute", "moreover"), Tier 3 domain-specific (e.g. "conscription", "metonymy"), or implicit cultural knowledge (e.g. "the 10.30pm news", "the dole"). Do NOT list "hard-sounding" words a Year [X] student would know in context.

PASS 3: The implicit prerequisite.
Name the ONE piece of prior knowledge the text quietly assumes. Quote the sentence where the assumption sits. State, in one sentence, what a student who lacks that prior knowledge will misread the sentence to mean.

PASS 4: The pre-reading check (under 60 seconds).
Write ONE question I can put to the class BEFORE they read, that surfaces whether the implicit prerequisite from Pass 3 is in the room. Not a comprehension question. A "do you know enough to read this yet" question. Tell me what a confident response looks like and what a thin response looks like.

PASS 5: The post-reading check (under 90 seconds).
Write ONE question I can put to the class AFTER they read, that would land DIFFERENTLY for a student who has the ONE thing I said I wanted them to take from the text versus a student who has not. Write the question. Tell me what the two responses would look like.

Then, "The annotation move": one sentence on the smallest annotation instruction I should give students before they read, given the year level and how they are reading it. Not "annotate as you read". Something concrete (e.g. "circle every sentence that makes a claim the writer does not back up").

Then, "Equity check": one sentence on which group of students in this class context is most likely to be quietly left behind by this text as set. Name the specific feature of THIS text that creates the disadvantage. Do not give me a generic "support EAL/D students" line.

Finally, "Sanity check": one sentence on what would make you, the model, wrong about the vocab trip-ups. If a word you listed is more commonly hard in textbook lists than in real Australian classrooms at this year level, say so.

Australian English throughout. Do not use the phrases "deep dive", "unpack", or "scaffold their thinking".

What it does well

Forces the model to quote the actual load-bearing sentences from the text rather than summarise the text, which is the move that turns the output from a pre-reading lesson plan into a real triage. The Tier 2 versus Tier 3 versus cultural-knowledge split is the bit that stops the vocab list defaulting to "long words" and turns it into a genuine pre-teach plan. The pre- and post-reading questions are different functions, which the prompt makes the model think through separately rather than collapse into one comprehension prompt.

Where it falls over

The model is reading the text without knowing your class. The vocab list will sometimes include words your kids absolutely know (because they hear them in a podcast every week) and miss the ones they have only ever met in a teacher's example sentence. Always cross the list against three actual students before you trust it. The prompt also will not save you from a text that is the wrong text for the question you want students to answer. If the reading is too long, no triage fixes the length problem. That is a cut job.

Try it like this

For Year 10 history (a primary source on a domestic political moment): the load-bearing sentence is almost always the one where the writer signals their position without naming it. The implicit prerequisite is usually "knows what the other side believed", which the source assumes the reader already holds.

For VCE English (an academic article on the set text): the Tier 2 vocab list earns its keep. Words like "conflate", "ostensibly", and "predicated on" lose students who pretend to know them. Pre-teach three, not ten.

If you only have ten minutes today, run it on the chapter you were going to set for homework anyway. The triage tells you whether to set the whole chapter, the first two pages, or a different chapter altogether.

The point of pre-reading is not to read the text for them. It is to make sure the text is the one you would have assigned if you had read it as a student first.