King Street may soon belong to its pedestrians, at least on summer weekends. Next month, city council will vote on a proposal to close six blocks of the shopping street to cars every Saturday and Sunday from June to September, freeing the pavement for walkers, patios, and market stalls. Supporters call the trial cheap and easy to undo; opponents answer that its costs would land on the shoulders least able to carry them.
Marisol Vega, who has run a café on King Street for nine years, collected more than eight hundred signatures in favour of the plan. "When cars disappear, people slow down, and people who slow down spend money," she said. She points to weekend closures in other Canadian cities, where nearby shops reported busier summers, and she has already priced the extra tables she would set out on the road.
A few doors away, Werner Brauer sees the same summer differently. His family has owned Brauer Hardware since 1982, and Saturday morning, he says, is when his regulars collect lumber, paint, and ladders. "Ms. Vega is right that the street would be busier," he concedes, "but busier with browsers. Nobody carries a ladder home on the bus." Without somewhere to load a car, he fears his best customers would drive straight to a supplier on the highway.
Between the two stands Hannah Osei, a transportation researcher who has studied dozens of similar street projects. The evidence, she says, is mostly encouraging, pedestrian streets usually attract more visitors and more spending. "But the wins are not automatic," she cautions. The closures that lasted, she found, arrived together with better bus service and clearly marked loading zones; where cities skipped those details, several were quietly reversed. "The design decides the outcome, not the slogan."
Council votes on the fifteenth. If the proposal passes, the barriers would go up in June, and the street would return to normal in September while the city measures the results.
Part 4 is the hardest part: a news style article quotes several people on a local issue, and you answer 5 questions about their viewpoints, then complete a reader comment with 5 more blanks, all in about 13 minutes. Its 10 questions are the largest block of Reading points.
In Part 4 a news style article presents several people's views on a community issue, and you answer 5 questions about those viewpoints, then complete a reader's comment that has 5 more blanks, all in about 13 minutes. It is the hardest Reading part and the biggest, 10 of the 38 points. The demo above is the exact interface you get on test day.
One point per correct answer and no penalty for guessing, like all of Reading. Part 4 carries 10 of the 38 scored questions, about a quarter of the raw score at the top of the difficulty curve, which makes it the highest leverage block for reaching CLB 9.
All 10 questions are plainly right or wrong, the hardest inference is worth the same as the easiest lookup. A guess keeps a one in four chance, an empty drop-down keeps nothing.
There is no separate Part 4 score. Its points feed the Reading total out of 38, which converts to a CELPIP Level from 1 to 12 with difficulty equating across test versions.
The mapping is one to one, CELPIP 9 equals CLB 9. Level 9 usually needs around 31 of 38, so dropping three or four of Part 4's ten questions is what most often sinks the target.
Officially every Reading question tests General Meaning, Specific Information, or Inference, and Part 4 leans hardest on Inference. Most likely and probably are the official markers that an answer is not stated directly.
Read question stems 1 to 5 before the article, skipping the options, and note the names they mention. The stems roughly follow the article's order and tell you which speakers to track.
For each named speaker note their role, their stake, a plus or minus for their position, and the reason behind it, one line each on the paper provided. Attribution questions are decided at the reason level, not the conclusion.
Verify each option against the speaker's own section, the right one restates their reasoning in fresh words. Keep whole article stems for last and demand support from more than one paragraph before you commit.
Fix the commenter's side first, it usually shows in the opening sentence, then fill each blank to stay consistent with that side and with the article's details. Keep the final 30 seconds to confirm all ten drop-downs are set.
These are the exact on screen instructions and the stem shapes that repeat across Part 4. The names inside the stems are your locator keywords, and most likely or probably signals an inference question.
Correct answers are paraphrases by design, so an option that repeats a speaker's exact phrase is usually bait. Official guidance says never to expect the answer in the very same words as the article.
The favourite distractor family hands one person's view to another, or to the article's writer, or upgrades something merely mentioned into an endorsement. Always ask who actually said it, and whether they meant it.
A speaker who grants that she is right about one point and then adds but is disagreeing, the real stance lands after the but. Distractors are routinely built from the conceded half of exactly such sentences.
All four options for a blank usually fit the sentence, and about half the blanks hinge on the article's details or the commenter's side. A smooth sounding option that contradicts either one is wrong.
Each Reading part runs on its own clock, minutes saved in Parts 1 to 3 vanish, and Part 4's 13 minutes cover the biggest question load. Cap any single question at about 90 seconds and keep every drop-down filled.
Officially the Reading parts grow harder from first to last, and Part 4 tops the curve. The article argues instead of informing, most questions lean on inference, and the load is the largest, 10 questions in about 13 minutes.
Content, almost entirely. All four options tend to fit grammatically, while the blanks hinge on the article's details and the side the commenter has taken. Decide that side first, then reject every option inconsistent with it.
Go back to that speaker's own sentences and test each option against their reasoning, not just their conclusion. Watch for concessions, the stance lands after the but, and distrust options that quote their words while bending the point.
