NCEA Literacy (Reading) common mistakes
NCEA Literacy (Reading) · Understand the traps and how to avoid them.
Choosing an answer because it contains words copied straight from the passage, even though it does not actually answer the question asked.
Under time pressure, matching surface vocabulary feels safe and is quicker than checking whether the option answers the exact wording of the question. Distractors are often built from lifted phrases for exactly this reason.
What to do instead
Before reading the options, restate the question in your own words (for example, 'I need the reason, not the result'). Then test each option against that restatement and reject any option that only repeats wording without supplying the required relationship.
Answering from general knowledge or personal opinion instead of from what the text says.
Familiar topics trigger prior knowledge, and the brain fills gaps automatically. Candidates then select the option that is true in real life but not stated or supported in the passage.
What to do instead
For every answer, point to the line or section that supports it and say the support aloud or write it in the margin. If you cannot locate support in the text, the option is wrong even if it is factually true.
Missing negation, exception and qualifier words such as 'not', 'except', 'least', 'only', 'mainly' and 'always', and answering the opposite of what was asked.
These words are short and visually low-salience, and skimming readers process the content words while skipping the logic words that reverse the task.
What to do instead
Circle or underline every negation, exception and qualifier in the question stem before looking at options. Re-read the stem after choosing, checking that your answer satisfies the exact logical form of the question.
Treating an inference question as a retrieval question, or the reverse: giving a directly stated fact when asked what is implied, or inventing a conclusion when the answer is stated plainly.
Candidates do not distinguish the command words. 'According to the text' and 'The writer suggests' look similar on a quick read but demand different evidence standards.
What to do instead
Classify each question first: retrieval (find it), inference (combine clues and conclude), or evaluation (judge the text). For inference, require at least two textual clues that combine to support the conclusion, and reject any conclusion that needs outside knowledge.
Summarising or integrating across two texts by reporting each text separately, without stating where they agree, conflict or build on each other.
Reading two passages is cognitively heavy, so candidates default to sequential retelling and never perform the comparison the question rewards.
What to do instead
Make a quick two-column or Venn-style note of each text's main claim and key evidence before answering. Then phrase your answer as a relationship: 'Both state X, but Text A claims Y while Text B claims Z.'
Judging a persuasive text as reliable because it is confidently written, uses statistics, or appears in a professional-looking layout, without checking who wrote it, why, and what is omitted.
Confident tone and numbers feel like evidence. Candidates evaluate style and presentation rather than sourcing, and rarely ask what the writer left out or stands to gain.
What to do instead
Run a fixed checklist on every persuasive text: author and their interest, publication and date, audience, purpose, evidence type (anecdote, expert, data, emotional appeal), and what counter-arguments are absent. Base reliability judgements on that checklist, not on tone.
Identifying a language feature (for example, rhetorical question, emotive adjective, metaphor) but not explaining its effect on the reader or its link to the writer's purpose.
Candidates memorise feature names as labels and stop at naming, because naming feels like the complete answer and effect statements require an extra reasoning step.
What to do instead
Use a fixed sentence frame: 'The writer uses [feature] to [effect on reader], in order to [purpose].' Practise until every feature answer includes both the effect and the purpose link, and drop answers that only name the device.
Reading visual or multimodal elements as decoration, so captions, headings, graphs, diagrams, colour and layout are ignored when answering.
Candidates default to reading body prose and treat non-prose elements as optional, even when the question depends on a chart axis, a caption or a heading hierarchy.
What to do instead
On first contact with any text, deliberately scan the non-prose layer: title, headings, captions, labels, axis units, legend, and any image. Note what each contributes before reading the body, and check graph axes and units before quoting any number.
Misreading unfamiliar vocabulary from context by assuming the most common meaning of a familiar-looking word, or by ignoring the surrounding sentence.
Many exam words are familiar in one sense and used in a technical or less common sense in the passage. Candidates substitute the everyday meaning without testing it against the sentence.
What to do instead
Substitute your guessed meaning back into the sentence and check that the sentence still makes sense and matches the surrounding logic. Use morphology (prefixes, roots, suffixes) and part of speech as secondary clues, and prefer the meaning that fits the whole paragraph.
Running out of time because the candidate reads every passage at the same slow speed and answers questions in the order printed.
No reading plan is set before starting, so dense passages absorb disproportionate time and later, easier marks are lost.
What to do instead
Preview questions first and mark easy retrieval items. Skim the passage for structure (headings, topic sentences, first and last paragraphs), answer high-confidence items, then return to inference and evaluation items with remaining time. Practise with a timer to calibrate pace per passage.