How Reading Distractors Are Built
Almost every wrong option in this section is true. Learning the small number of ways they are built is faster than learning to read faster.
Wrong does not mean false
The most useful single fact about CELPIP Reading is that three of your four options are usually true. They are real details from the real text. They were not invented, they were relocated.
This is why the question most candidates ask themselves is the wrong one. Is this true? eliminates almost nothing, because most of the options pass it. The question that works is does this answer the stem in front of me?, and usually only one option passes that.
It also explains a common and demoralising experience: reviewing a test, seeing the correct answer, and thinking that your answer was also correct. It probably was, as a statement about the text. It just was not the answer to the question, and no amount of rereading the passage will reveal that. The problem was in the stem.
Distractors in this section come in a small number of families. Learning to recognise them is quicker than any improvement in reading speed, because each family has a specific check that defeats it in a couple of seconds.
The true-but-not-asked option
The commonest family across the whole section. A genuine detail from the text, attached to a question it does not answer. The stem asks why the council delayed the vote; the option gives a real reason the project was criticised, which is a different thing.
These work because recognition feels like knowledge. You read the option, something in you says yes, that was in there, and the sensation of familiarity gets mistaken for the sensation of correctness. It is the single most reliable trap in the section because it exploits accurate memory rather than poor reading.
The check. Read the stem again with your chosen option inserted, as one sentence, and see whether it makes a claim the text supports about the thing being asked. If the resulting sentence is true but changes the subject, you have this family.
The attribution swap
Part 4 runs on this one and Part 1 uses it too. A real position is moved to a different person. The critic's cost objection appears under the champion's name. The expert's caution is handed to the local resident. The letter writer's partner's reluctance is attributed to the letter writer.
It is cheap for the test writer to build, because they do not have to invent an opinion: four real ones are already on the page. And it is hard for a candidate to catch, because tone attaches to a text as a whole rather than to the sentence it came from, so a fast reader retains the argument and loses the owners.
The check. Say the person's name before you evaluate the option, then find their words. In Part 4 this is what the ledger is for. In Part 1 it means asking, for every option mentioning a feeling or an intention, whose feeling and whose intention.
A variant swaps the option rather than the person. That is the Part 2 version: a real feature belonging to a different column of the chart. The check is the same one in different clothes, which is to name the subject before you read the option.
Read every option to its full stop and treat a two-clause option as two questions. Half-right options work because the eye stops reading at the clause it can confirm.
The overstatement
The text hedges and the option hardens. She is thinking about applying becomes she has decided to apply. The study suggests a link becomes the study proves a link. Several residents complained becomes residents generally oppose it.
Nothing false has been added. A degree of certainty or scope has been added, and certainty is invisible to a reader who is checking content rather than modality.
The check. Look at the strong words. All, never, always, only, proves, decided, refused, must. When an option contains one, go back and find the corresponding word in the text. If the text said often and the option says always, the option is wrong however true the rest of it is.
The reverse also exists and is rarer: an option that understates a text that was categorical. It catches people who have learned to distrust strong words and now distrust them everywhere. The rule is not that strong options are wrong, it is that the option's strength must match the text's.
The paraphrase trap and its mirror
Two opposite constructions, and they catch opposite kinds of reader.
The word-match trap. An option that reuses the text's exact vocabulary while saying something the text did not. It is built for people who match words, and it is most common in Part 3, where the correct answer shares almost no vocabulary with the statement and a wrong paragraph shares one conspicuous word.
The paraphrase that is correct. The mirror image. The right answer restates the text in completely different words, so it feels unfamiliar and gets rejected. Candidates who have been warned about word matching sometimes overcorrect into rejecting any option that sounds like the text, which is its own error.
The check. Decide what the text actually says, in your own plain words, before you look at the options. Then compare each option to your own sentence rather than to the passage. Vocabulary overlap becomes irrelevant, which is exactly the point.
When an option contains all, never, always, only, proves or decided, go back and find the matching word in the text. The option's certainty has to match the passage's, in both directions.
Partial truths and the half-right option
The last major family. An option with two clauses, one of which the text supports and one of which it does not. The pass is the cheapest option and allows weekend travel. Cheapest, yes. Weekend travel, no, that is the other pass.
These are the hardest to catch under time pressure, because reading confirms the first clause and confirmation ends the reading. The eye stops at the point of satisfaction, and the second clause goes past unchecked.
The check. Read every option to its full stop, and treat an option with two claims as two questions. Both have to be true. This is mechanical and slow-sounding, and it takes about two seconds per option once it is a habit.
A related construction supplies a true cause and a false effect, joined by because or so. The Part 4 version is the concession that becomes a change of mind: the critic accepts the savings, so the critic supports the scheme. Accepting is real, supporting is not, and the join is where the option was built.
Using this in review rather than on the day
You cannot run six checks on four options inside sixty seconds, and you should not try. The value of knowing the families is mostly in review.
After any practice set, do not just note which questions you got wrong. Name the family for each one. If four of your six errors were true-but-not-asked, your problem is stem discipline and the fix is rereading the stem before committing. If four were attribution swaps, your problem is that you read Part 4 without a ledger. If four were overstatements, you need to start looking at the strong words.
Three or four sets of that kind of review usually show one family dominating, and one habit fixes it. That is a much better use of an hour than another full reading test, because a test tells you your score and a review tells you what to change.
In review, label every wrong answer with the family it belonged to rather than just marking it wrong. One family usually dominates, and naming it turns a vague weakness into a single habit to change.
Practise it
Five questions on distractor construction and the check that defeats each family.