Decision Making (DM) is one of the most intellectually engaging sections of the UCAT — and, since its expansion in 2025, one of the most demanding. With 35 questions in 37 minutes across six distinct question types, it tests your ability to reason logically, handle uncertainty, and make sound judgements under time pressure. These are exactly the skills that medical schools are looking for in future doctors.
In this comprehensive guide, I'll walk you through everything you need to know about the DM section: what it tests, how it's scored, what each question type involves, and the strategies that work best for each one.
DM at a Glance
| Fact | Detail | |------|--------| | Questions | 35 | | Time | 37 minutes | | Time per question | ~63 seconds | | Question types | 6 | | Scoring | 300–900 | | Multi-statement questions | 2 marks (partial credit: 1 mark for ≥4/5 correct) | | No negative marking | Yes |
The 2025 mean DM score was 628, the highest of any UCAT cognitive section, according to official UCAT statistics. Scoring in the 700s puts you in approximately the top 20–30% of candidates.
How DM Changed in 2025
If you're working from older preparation resources, the DM section you'll encounter is significantly different from what they describe:
| Parameter | Pre-2025 | 2025 onwards | |-----------|----------|--------------| | Questions | 29 | 35 | | Time | 31 minutes | 37 minutes | | Time per question | ~64 sec | ~63 sec |
The question-type composition is similar, but the section is substantially larger. Conclusion drawing questions — which carry 2 marks each — now represent an even larger pool of available marks, making them the single most important question type to master.
DM Scoring: How It Works
Most DM questions are single-answer questions worth 1 mark each. You select one correct response from four options.
However, a significant proportion of DM questions are multi-statement questions. These present you with a scenario and ask you to assess 5 statements, each marked as Yes or No. Scoring for these is:
- Both marks (2): All 5 statements correct - Partial mark (1): At least 4 out of 5 correct - No marks (0): Fewer than 4 correct
This partial credit system has important strategic implications. If you've confidently assessed 4 of the 5 statements and are uncertain about the fifth, it's worth selecting any answer for the fifth rather than abandoning the question — partial credit is available.
The Six DM Question Types
1. Syllogisms
What they are: You're given two premises (logical statements) and a proposed conclusion. You must decide whether the conclusion necessarily follows from the premises — not whether it's probably true, but whether it's guaranteed to be true if the premises are correct.
What they look like: > "All medical students study anatomy. Some students who study anatomy become surgeons. Therefore: Some medical students become surgeons." > Does this conclusion necessarily follow? Yes / No
The logical framework:
Valid syllogisms follow specific patterns. The key distinction is between conclusions that must be true (valid) and conclusions that might be true (invalid as a logical deduction).
Common valid patterns: - All A are B + All B are C → All A are C - All A are B + No B are C → No A are C - Some A are B + All B are C → Some A are C
Common invalid traps: - All A are B + All C are B → All A are C (undistributed middle — "B" being shared doesn't link A and C) - Some A are B + Some B are C → Some A are C (two "some" premises rarely guarantee a conclusion)
Watch carefully for: - "Some" — means "at least one," not "most" or "many." Two "some" premises almost never yield a definite conclusion. - Negatives — "No A are B" runs differently from "Not all A are B" - "All" vs "most" vs "some" — these quantifiers are not interchangeable in syllogistic logic
Time target: 40–50 seconds.
2. Logic Puzzles
What they are: Constraint-based arrangement or deduction problems. You're given a set of conditions (e.g., "X sits to the left of Y; Z is not adjacent to W") and asked to determine what must, can, or cannot be true given those conditions.
What they look like: > "Six doctors are seated in a row. Dr Anand is not seated next to Dr Brown. Dr Chen is at one end. Dr Patel sits directly between two others..."
The approach:
1. List all entities and all constraints 2. Identify the most restrictive constraint first — usually the one that fixes a specific position 3. Work from that fixed point to eliminate possibilities 4. For "must be true" questions, test whether there is any valid arrangement where the statement is false — if there isn't, it must be true
Common variants: - Linear arrangements (seating in a row) - Circular arrangements (seating at a table) - Schedule problems (who works on which day) - Assignment problems (which patient is treated by which doctor)
Logic puzzles vary enormously in complexity. A simple one might resolve in 40 seconds; a complex multi-constraint problem might take 90 seconds or more. If you've spent 70 seconds and are not converging on an answer, flag it, guess, and move on.
Time target: 60–90 seconds (variable).
3. Strongest Argument
What they are: A statement or proposition is presented (often a policy or ethical claim), followed by four possible arguments for or against it. You select the strongest argument.
What they look like: > "Should all medical students be required to complete a mental health first aid course?" > A: No, because it would add to an already busy curriculum. > B: Yes, because mental health issues affect one in four people, and doctors frequently encounter patients in crisis. > C: Yes, because mental health is important. > D: No, because some students may already have relevant experience.
What makes an argument "strong": - Directly addresses the proposition (not tangential) - Based on evidence or logic, not just assertion or emotion - Specific — vague claims like "it's important" are weak - Withstands scrutiny — doesn't rely on unstated assumptions
What makes an argument "weak": - Emotional appeal without substance - Restates the conclusion (circular reasoning) - Tangential — addresses a related but different point - Applies equally to many situations (not specific to this proposition)
The elimination approach: Immediately discard emotional/circular/vague options. This typically reduces four options to two, and the final choice is between which of the two is the stronger logical argument.
Time target: 40–55 seconds.
4. Inference and Conclusion Drawing
What they are: The highest-value question type in DM — conclusion drawing questions typically carry 2 marks and require you to assess 5 statements against a given passage or scenario.
What they look like: > [A short passage about a medical or social scenario is provided] > For each of the following statements, select Yes if it follows as a logical conclusion from the passage, or No if it does not. > 1. [Statement A] > 2. [Statement B] > (and so on for 5 statements)
Why these matter so much: Conclusion drawing questions account for approximately 40% of total DM marks. With 2 marks each and partial credit available, they are disproportionately valuable relative to single-answer questions.
The framework for each statement:
A statement is Yes (logical conclusion) only if: - It is necessarily true given everything in the passage - It does not go beyond the information provided - It requires no additional assumptions
A statement is No if: - It is only plausible or likely, not certain - It extends beyond the scope of the passage - It introduces assumptions the passage doesn't make - It contradicts or cannot be confirmed by the passage
Common traps: - Overgeneralisation: Passage says "some patients"; statement says "most patients" → No - Causal overreach: Passage shows a correlation; statement asserts a cause → No - Scope extension: Passage discusses one group; statement claims it applies to everyone → No - Temporal shift: Passage uses past tense; statement claims present or future truth → No (unless the passage explicitly projects forward)
Time target: 70–90 seconds. This is justified — 2 marks warrants more time than a 1-mark question.
5. Venn Diagrams
What they are: Set-theory problems using overlapping circles representing categories. You're asked to place items in the correct region, calculate numbers within regions, or assess statements about the relationships between sets.
What they look like: > A Venn diagram shows three overlapping sets: Doctors (D), Surgeons (S), and Research Fellows (R). Of 120 hospital staff: 40 are Doctors, 25 are Surgeons, 30 are Research Fellows. 10 are both Doctors and Surgeons, 8 are both Surgeons and Research Fellows, 5 are both Doctors and Research Fellows, and 3 are all three. How many staff are in none of the three categories?
The systematic approach:
For 2-circle diagrams: - Label regions: A only, B only, both A and B, neither - Fill in known values; solve for unknowns by subtraction
For 3-circle diagrams: - Label all 7 regions: A only, B only, C only, AB (not C), AC (not B), BC (not A), ABC - Fill the triple intersection first - Work outward: use triple-overlap value to find double-overlap regions - Fill single-category regions last - Verify: sum of all regions = total population
Critical language: - "Only A" = A without B or C (the A-only region) - "At least A and B" = AB region + ABC region - "Exactly A and B" = AB region only (not C)
Always draw the diagram. Even a rough sketch takes 10 seconds and prevents errors that take 30 seconds to untangle mentally.
Time target: 55–70 seconds.
6. Probabilistic Reasoning
What they are: Questions involving probability, likelihood, risk, and statistical inference. These typically present a scenario with numbers — rates, proportions, frequencies — and ask you to calculate or compare probabilities.
What they look like: > "In a population of 10,000 people, 2% carry a genetic variant. Of those with the variant, 60% will develop the condition in their lifetime. Of those without the variant, 5% will develop the condition. What is the probability that a randomly selected person from this population develops the condition?"
The approach:
1. Identify the full population and all relevant subgroups 2. Calculate frequencies for each subgroup (percentages to absolute numbers when possible) 3. Use a frequency tree or contingency table if the problem is complex 4. Apply the formula: Probability = (favourable outcomes) ÷ (total outcomes)
Common types: - Simple probability (one-step calculation) - Conditional probability (given that X, what is the probability of Y?) - Combined probability (P(A and B), P(A or B)) - Expected frequency (in a population of N, how many would...)
Use the on-screen calculator for these — they often involve multi-step arithmetic.
Time target: 50–70 seconds.
DM Time Management Strategy
With 37 minutes for 35 questions and six different question types, time management in DM requires a tiered approach rather than a fixed 63-second rule.
The core principle: Allocate time proportional to marks available.
- Single-answer questions (1 mark) → target 40–55 seconds - Multi-statement questions (2 marks) → target 70–90 seconds
Section checkpoints:
| Questions completed | Target time elapsed | |--------------------|--------------------| | 12 | ~12 minutes | | 24 | ~24 minutes | | 35 | 37 minutes |
If you're at 12 questions and more than 13 minutes have passed, you're behind. Flag any remaining question that is taking more than 55 seconds and move on.
How theMSAG Can Help
Our DM preparation at theMSAG goes deep into all six question types, with particular focus on conclusion drawing (the highest-value type), Venn diagrams, and syllogisms. Our UCAT Question Bank with 6,200+ questions includes DM questions organised by type, so you can drill weak areas efficiently. In our Live UCAT Course, we work through DM logic frameworks live, helping you build the reasoning skills that make this section feel systematic rather than unpredictable.
Last verified by Dr Dibah Jiva — March 2026