NZ Police cognitive test: five sample questions worked through

Read each one with the method covered, decide how you would approach it, then check. What you are looking for is not whether you got it right, but whether you attacked it the same way.

A verbal reasoning example

You are given a passage and a statement, and asked whether the statement is supported by it. The trap is almost always the same: a statement that is true in the world but not established by the passage.

Worked: a passage says that a station recorded more callouts in December than in any other month, and that summer staffing was increased in response. The statement to judge is that callouts rise in summer. It sounds right, and December is a summer month, but the passage establishes one month, not a season, and gives no figures for January or February. Not supported.

The method: answer only from the text. Underline the words that would have to be present for the statement to follow. If they are not there, the statement is not supported: however reasonable it sounds.

A numerical reasoning example

A table of figures and a question about a change, a rate or a proportion. The arithmetic is rarely difficult; the pressure is.

Worked: a table shows incidents by district across four quarters, and you are asked which district had the largest percentage increase between Q1 and Q4. The instinct is to work out every district's change. The faster route is to notice that a large percentage increase needs a small starting figure, scan the Q1 column for the lowest numbers, and check only those two or three.

The method: work out what you actually need before touching any numbers. Most lost time comes from calculating a column that turns out to be irrelevant. Estimate first, then confirm, an estimate often eliminates three options immediately.

An abstract reasoning example

A sequence of shapes with one missing. The rule might be rotation, count, shading, position, or two of those combined.

Worked: four panels, each with a set of small shapes. Trying to see it whole produces nothing. Counting shapes across the four gives 3, 4, 5, 6, so the missing panel has seven. That alone usually removes two options, and the remaining pair differ on one attribute you can then check on its own.

The method: check one attribute at a time across the whole sequence rather than comparing whole images. Count first, then position, then rotation, then shading. Trying to see the pattern all at once is what makes these feel impossible.

A situational judgement example

A colleague asks you to overlook a small procedural shortcut that would save the team time on a busy shift. The options range from refusing outright to going along with it.

Worked: the strongest option declines the shortcut and says so to the colleague directly, because it resolves the immediate issue and keeps the working relationship intact. Going along with it fails on integrity. Reporting it to a supervisor without speaking to the colleague first is defensible but skips a step. Ignoring it defers the problem to whoever is on the next shift.

The method: rank the options against safety, integrity, escalation and communication, in that order. The strongest answers usually address the immediate issue *and* handle the relationship, rather than choosing between them. The weakest are the ones that defer a problem or that escalate before speaking to the person involved.

A written accuracy example

Identifying the error in a sentence, or choosing the clearest version of a passage. Police work generates a great deal of written record, and this section is a proxy for whether yours will hold up.

Worked: "I spoke to the driver of the vehicle who was parked outside the address." Grammatically unremarkable, and genuinely ambiguous: the driver was parked outside, or the vehicle was. In a statement that difference matters, and the version that names which is the correct answer.

The method: read for meaning first and grammar second. The most common error in these items is ambiguity, a sentence that is technically correct but could describe two different events.

What do these questions have in common?

In every case, the difficulty is not the content, it is deciding quickly what the question is asking. That is a skill, it is trainable, and it is most of what practice buys you. It is also why the test is harder than its content suggests.

What should you do with these?

Use them to pick a starting point, not as a syllabus. If one of the five felt noticeably worse than the others, that is your weakest area and the place a preparation plan should begin, which is exactly how the four-week plan is sequenced. For where this stage sits in recruitment and what follows it, see the complete guide to NZ Police recruitment testing. Situational judgement gets its own explainer on how these tests are scored.