A main-idea question can feel oddly difficult when every sentence in the passage is understandable. The trap is often a choice that repeats a vivid detail and then presents that detail as the whole point. Imagine a short passage about a city replacing some streetlights with warmer bulbs. It mentions migration, a pilot street, residents’ reports, and the cost of maintenance. A choice about the pilot street may be true, but the passage may really be about the trade-off the city is studying. The test is asking for the job of the whole passage, not the most memorable noun in it.
I would read the final sentence and the opening sentence again before looking at the choices. Then I would write a five-word private label, such as ‘city weighs lighting trade-offs.’ This is not a summary to submit. It is a guardrail against attractive options. A defensible answer should cover the central relationship without claiming a result the passage does not establish. If the passage says researchers are testing whether warmer bulbs help, an option that says they have proved the benefit goes too far. If it spends half its space on cost, an option about wildlife alone may be too narrow.
A practical review question is: which sentence would become unnecessary if this option were the main idea? If most of the passage would suddenly look irrelevant, the option is probably too small. The reverse also matters. An answer about ‘the importance of environmental policy’ might be broad enough to fit thousands of texts but not precise enough for this one. You want the narrowest statement that still accounts for the passage as a whole.
Here is another way to see the difference. A passage may begin by describing how an archive digitized old photographs, then spend its remaining lines on the problem of preserving captions. An answer about scanning technology names the setup, not the destination. An answer about all museums modernizing their collections is larger than the evidence. A good answer would mention that digitization improves access while creating a separate need to preserve context. That sentence accounts for both halves without pretending the passage argues against digitization.
The stem matters too. ‘Main purpose’ asks what the author is doing; ‘central idea’ asks what the passage says; ‘a detail’ can be answered by a single sentence. These are related but not identical jobs. If the author compares two explanations without endorsing either, do not pick an option that says the author proves one explanation. If the author introduces a problem and a possible response, a choice describing only the problem leaves the second half behind. I find it helpful to note the passage’s movement with a tiny arrow: problem to response, old view to new evidence, or observation to question.
After answering, test your choice against every sentence, not by asking whether every word appears, but whether each sentence has a role under that claim. One sentence may supply an example; another may qualify the conclusion. If the answer cannot make room for the qualifying sentence, it is probably too strong. This is also why rereading only the first line is risky: openings often introduce a topic, whereas the last line tells you what the writer finally wants to say about it.
A useful error log has room for the phrase that fooled you. Write the rejected choice and underline the exact word that narrowed or enlarged it. ‘Scientists studied three neighborhoods’ is not the same as ‘scientists established a citywide pattern.’ Once you spot how an answer changes a limited observation into a broad claim, you can catch the same move in a different topic tomorrow. The surface subject changes; the reading mistake is often the same.
The College Board places central ideas and details among the Reading and Writing skills in its official Student Question Bank. That is useful for targeted practice: filter to the skill, answer a small set, and write down whether a miss came from a detail, an overstatement, or a vague umbrella answer. The official question bank is at College Board Reading and Writing practice. An invented example helps us describe the method here; the official questions are where you should check whether it works under real wording.