UCAT ANZ Decision Making gives you 35 questions in 37 minutes, so just over a minute each. The official description says it tests your ability to apply logic to reach a decision, evaluate arguments and analyse statistical information. Some questions are standard multiple choice. Others give you five statements and ask for a yes or no on each. There's an on-screen calculator.
In 2026 the median DM score was 650 and the 90th percentile was 770. Because the questions take such different amounts of time, a lot of what separates those scores is which questions people spend their minutes on. Here's how we'd approach it.
The question types
You'll see a mix of these:
- Syllogisms: a set of statements, and you decide which conclusions follow.
- Venn diagrams: information about overlapping groups, either as a diagram or in words.
- Probability: short calculations, often about independent events.
- Interpreting information: a passage, table or chart, with conclusions to accept or reject.
- Strongest argument: a question like "should X happen?" with four yes or no answers, each with a reason.
- Logic puzzles: fitting people or things into positions using a set of clues.
Do the quick ones first
Every question is worth the same, but they don't take the same time. Logic puzzles can eat three minutes. A strongest argument question might take 30 seconds. So the biggest gain in DM comes from order.
Many high scorers work through the subtest in one pass, answering quick questions and flagging slow ones. A common order of "quick to slow" is syllogisms, probability, Venn diagrams, interpreting information, strongest argument, then logic puzzles last. Your own order might differ, which is why it's worth timing yourself on each type in practice.
Whatever you flag, put in a best guess before moving on. There's no negative marking, so if time runs out, at least you have an answer down.
Syllogisms: read the words literally
These test whether a conclusion has to be true based only on the statements. The trap is filling in what seems likely. Learn what the key words mean in the test. "Some" means at least one, and it could be all. "Most" means more than half. "All" and "none" mean exactly that.
Here's an example we wrote:
All the students in the choir are in Year 13. Some Year 13 students play hockey.
| Conclusion | Follows? | Why |
|---|---|---|
| Some choir members play hockey. | No | The hockey players could all be Year 13 students who aren't in the choir. |
| No choir members play hockey. | No | Nothing rules it out either. It might or might not be true. |
| Every choir member is in Year 13. | Yes | That's what the first statement says. |
A quick way to test a conclusion is to try to build a situation where the statements are true and the conclusion is false. If you can, it doesn't follow.
Probability: keep it simple
Most probability questions need only a few rules: multiply for "and" with independent events, and use 1 minus the chance of "none" for "at least one".
For example, a bus is late on 20 percent of days and a train is late on 10 percent of days, independently.
- Both late: 0.2 x 0.1 = 0.02, so 2 percent.
- At least one late: 1 - (0.8 x 0.9) = 1 - 0.72 = 0.28, so 28 percent.
For anything with stages, like drawing twice without replacement, sketch a quick tree diagram on your noteboard.
Venn diagrams: easiest statements first
With five yes or no statements about a diagram, check the quick ones first, like a single region or a simple total. Statements that need you to add up several regions are slower, so leave them until last within the question. If a Venn question asks which diagram correctly shows a set of relationships, go through the options and rule out the ones that break any single rule.
Strongest argument: answer the question asked
The strongest argument is the one that directly answers the question with a reason that's relevant, important and not overstated. It's not about whether you agree. Watch for these weaknesses:
- The reason doesn't address the question that was asked.
- The reason is true but trivial.
- It makes a sweeping claim that goes beyond the question, like "this will solve the problem completely".
- It relies on weak words like "might" or "could" when another option gives a firm reason.
Logic puzzles: draw a table
Put the fixed things (like names) on one axis and the variables (positions, colours, days) on the other. Fill in the absolute clues first, like "Ana is in seat 3", then the relative ones, like "Ben sits next to Ana". Most puzzles fall into place once the absolute clues are in. If it hasn't come together in two minutes, flag it.
Practise with official material
The UCAT ANZ website has official practice material, and for DM it's especially worth using. It shows how the real test defines terms like "some" and what counts as a valid inference. Some third-party questions are stricter or looser than the real ones, and that can teach the wrong habits. Use other practice for speed, and the official material to check your logic.
Keep an error log as you go, with the question type and why you got it wrong. If most of your errors are in one type, that type gets extra sessions. Our post on when to start UCAT ANZ prep has a plan for spreading this out.
FAQ
What is a good Decision Making score?
In 2026 the median DM score was 650 and the 90th percentile was 770. For Otago's 2027 entry thresholds, DM didn't have a published threshold, but Otago says thresholds in other subtests may be applied later. See our post on what your 2026 score means.
How do I stop running out of time in Decision Making?
Flag logic puzzles and long calculations, answer everything else first, and put in a guess before you flag. Timing yourself by question type in practice shows where your minutes go.
Should I use the calculator in DM?
For probability and data questions, yes, if the numbers aren't simple. For most other types you won't need it.
If you want timed DM sets that work like the real test, see how Cutline UCAT practice works.
Cutline is not affiliated with or endorsed by the UCAT ANZ Consortium or the University of Otago. UCAT is used here only to describe the test. Format and statistics are from ucat.edu.au, checked 8 October 2026. The examples were written by Cutline for practice.
