The Inference Leak in an Unseen Passage
A 350-word unseen passage on the return of sparrows to a city is used in a 12-mark reading section. The five questions are: two vocabulary items (2 marks), three literal retrieval items (3), one inference item - 'What does the writer imply by the city has begun to listen?' (3) - one negative factual item (2), and one personal-response item (2). The passage contrasts the city's earlier noise with its new listening. In a trial with 42 Class XII students, 31 scored 0/3 on the inference item, 24 chose the option that repeats a sentence from paragraph 3 on the negative item, and 18 copied whole sentences for vocabulary. The reading period is 25 minutes. The teacher has one 40-minute class before the real test and cannot change the passage. She wants to raise the inference score without turning the lesson into answer-giving.
1. Diagnose the likely cause of the three error patterns and choose one 15-minute intervention. Justify the choice with the data.
Read the worked answer
The data point to one bottleneck: students are matching words, not building meaning. The 31 zeros on the inference item cost 93 marks; the 24 negative-item errors cost 48; the 18 vocabulary copies cost at most 18. The inference item therefore deserves the scarce time. A 15-minute say–suggest ladder works: (1) write the quotation on the board; (2) column A: what the city literally does (makes noise, builds, moves); column B: what listen suggests (attention, care, change); (3) model one link with because; (4) pairs complete a second link; (5) two students read out and the class judges whether the link is earned by the words. Spend 10 minutes here and 5 on a negative-item routine: circle NOT, name the paragraph, eliminate the option that copies a sentence. Do not reteach vocabulary as meaning; reteach the instruction use your own words with one example. This is a judgement call: the negative item is cheaper to fix, but it is not the largest loss.
2. Rewrite the negative factual item so that it still tests careful reading but is less likely to be misread. Explain what you changed and what you deliberately kept.
Read the worked answer
The current stem probably reads 'Which of the following is NOT true?' It invites surface matching because students hunt for a sentence that looks familiar. Rewrite as: 'Which ONE statement is NOT supported by the passage? Circle NOT. For your chosen option, write the paragraph number that contradicts it.' Keep the same four options. The capitalised NOT and the contradiction step force verification rather than recognition. Do not convert it to a positive item: negative items are a legitimate reading skill and appear in the exam. The change is to the response format, not the difficulty. If the options include an absolute word such as only or never, keep it; that is part of the test. Check that only one option is genuinely unsupported.
3. Provide a 3-mark model answer to the inference question and explain how the marks should be awarded.
Read the worked answer
Model answer: The phrase suggests that the city has shifted from indifference to attention. Earlier the city is associated with noise and movement, but listen implies deliberate care: people now notice the sparrows and respond to them. The writer is not saying the city is silent; he is saying it has become receptive. Award 1 mark for the implication (attention, care, receptivity), 1 for linking it to the contrast between noise and listening, and 1 for explaining how the word listen carries that meaning. A response that only repeats 'the city has begun to listen' gets 0 because it offers no inference. A response that says 'the city is noisy' gets 0 because it misses the contrast. A response that says 'the city now cares about nature, as shown by listening rather than merely making noise' gets 3.