Use these selected concepts to connect a rule with its application: examine the worked example, explain the conclusion, and identify the mistake it avoids. The guide supports preparation for the current UK UCAT across Verbal Reasoning, Decision Making, Quantitative Reasoning and Situational Judgement.
Understanding passages
1. Passage evidence sets the boundary
A verbal reasoning conclusion is justified only by what the passage states. The UCAT explicitly tells candidates not to use prior knowledge, so a remembered fact cannot fill an evidential gap. If the passage omits information needed to decide a statement, the safe judgement is that the passage does not settle it, rather than importing outside context.
Worked example: A passage reports that a library extends loan periods during exam weeks but lists no dates; whether a stated March return was extended is left undetermined by the passage, so recalling a university calendar cannot justify a definite answer.
Mistake to avoid: Answering a statement as definitely true because a remembered real-world policy fills the passage's evidential gap.
2. Main ideas differ from supporting details
A summary reduces the whole passage to its main ideas, so a single vivid illustration should not be treated as the central message. Test each option against the full passage: a detail that supports one benefit among several is supporting material. Choosing a memorable example as the overall point misrepresents the author's scope and the passage's actual emphasis.
Worked example: A passage argues that cycling infrastructure improves health, air quality and journey times, illustrating air quality with one street's monitoring data; the main idea is the range of benefits, so a summary naming only that street's readings narrows the author's claim.
Mistake to avoid: Selecting a summary centred on the passage's most memorable illustration rather than the claim the whole passage develops.
Reference: [3]
3. Paraphrases preserve the original meaning
A paraphrase restates the text in different words while keeping the same meaning, including restrictions, scope and degree of certainty. Simplifying language must not widen a claim: if the original limits something to a subset, the paraphrase must keep that limit. Checking subject, qualifiers and certainty against the original prevents a reworded option from quietly changing the claim.
Worked example: A passage states that reminders reduced missed appointments among patients aged over seventy; a faithful paraphrase keeps the age restriction, while 'reminders reduced missed appointments' drops it and overstates the finding beyond what the trial reported.
Mistake to avoid: Dropping a stated restriction, such as an age limit, when rewording a sentence into simpler language.
4. A sentence's role depends on its context
Analyse a detail in relation to the whole passage, because the same sentence can support, qualify or motivate a neighbouring claim. A described failure may justify a proposed change rather than signal rejection of the approach. Reading the surrounding argument first prevents attributing to the author a position that one isolated sentence appears to suggest.
Worked example: A passage criticises a clinic's initial triage checklist and then proposes a revised version with extra red-flag questions; the criticism motivates the revision, so concluding that the author rejects triage checklists altogether reads one sentence apart from its argumentative purpose.
Mistake to avoid: Treating a sentence that introduces a proposed fix as the author's outright rejection of the whole approach.
Reference: [4]
5. Purpose concerns what the author seeks to achieve
A passage's purpose is what the author wants to achieve, such as informing or persuading, and it may differ from the subject both passages share. In UCAT verbal reasoning, identify the aim by asking what the author does with the topic, then check that the passage's wording and structure actually support that reading.
Worked example: A council meeting summary notes that parking meters will be replaced, while a residents' letter argues the meters should be scrapped; both concern parking charges, but the first reports a decision whereas the second campaigns for change.
Mistake to avoid: Assuming that any passage discussing a controversial proposal must support it, without checking whether the author's language reports debate or advocates a position.
Reference: [3]
6. Tone requires evidence from language
Tone is inferred from the author's word choices, not from how you feel about the subject. A passage praising a clinic's new hours but noting staffing pressures suggests measured approval, since the qualification tempers the praise; your own enthusiasm for longer opening times is irrelevant to the inference.
Worked example: A review calls a museum's renovation 'competent but unimaginative' and says children 'may enjoy' the new gallery; the hedged phrasing supports cautious, qualified approval rather than wholehearted praise, because the reservation is part of the author's wording.
Mistake to avoid: Choosing an enthusiastic tone because the topic interests you personally, even though the passage's wording raises doubts or states reservations about the development.
Reference: [4]
Evaluating written claims
7. Facts and opinions provide different support
Verifiable claims can in principle be checked against records or measurement, while opinions express personal judgements that may lack stated criteria. Confident delivery does not turn an assessment into a fact, and a checkable figure can still be incomplete, so passages may invite you to question the evidence behind either kind of statement.
Worked example: A report states that a library lent 12,000 books last year, a checkable record, and adds that it is the district's most valued service, an unquantified judgement; the lending figure can be verified, while the valuation rests on the author's assessment.
Mistake to avoid: Treating a strongly worded opinion as fact because it sounds authoritative, without asking whether any verifiable evidence or stated criteria support it.
Reference: [3]
8. An inference needs an evidential connection
A valid inference links passage evidence to a conclusion without stretching beyond what that evidence covers. If a survey found that all responding clinics employed a pharmacist, you may conclude only about responding clinics, not every clinic; the connection must come from the passage, and set boundaries limit the conclusion.
Worked example: A passage states that every completed questionnaire came from a registered practice, which supports the conclusion that every responding practice was registered; it does not show that all practices in the region responded, since non-responding practices fall outside the evidence described.
Mistake to avoid: Generalising a conclusion to the whole population when the passage's evidence covers only a subset, such as booked seats rather than all seats.
9. A claim is distinct from its supporting reasons
Separate the statement an argument asks you to accept from the statements offered to support it. A single passage may contain a proposed action, a reason, and further backing, and the same sentence can serve different roles in different arguments, so judge each sentence's function within that specific argument.
Worked example: A council proposal reads 'The car park should close at six because staff costs rise after that hour.' The conclusion is the closing proposal; the staffing-cost statement is the reason offered for it, so questions about the conclusion target the action, not the cost claim.
Mistake to avoid: Treating the stated reason as the conclusion being defended, and then answering questions about the wrong statement in the argument.
Reference: [3]
10. Relevant evidence must also be sufficient
Evidence that bears on a claim may still fall short of establishing it. Judge sufficiency against the exact strength and scope of the claim: a narrow claim about one case needs less support than an unrestricted claim about every case, so match the amount and reliability of evidence to what is asserted.
Worked example: A gym cites one member who reports better sleep, offered as proof that membership improves members' sleep generally. The report is relevant to sleep outcomes but covers a single member, so it illustrates a possible experience without establishing the general conclusion about all members.
Mistake to avoid: Counting a single anecdote as adequate support for a sweeping conclusion that covers every case in the study.
Reference: [3]
11. Missing information limits comparisons
A comparison requires the information that makes it fair, such as a shared denominator or matching time period. When that information is absent, the passage cannot settle the comparison, and you should identify what is missing without assuming any particular value for it.
Worked example: A report states that the evening clinic saw forty patients while the morning clinic saw thirty, but gives no session lengths. Without knowing how long each clinic ran, the passage cannot establish which saw more patients per hour, so the per-hour comparison stays undetermined.
Mistake to avoid: Silently inserting an assumed denominator, such as equal opening hours, to complete a comparison the passage never states.
Reference: [3]
12. Alternative explanations test causal claims
A causal claim weakens when the evidence also fits other plausible causes that the argument leaves unresolved. Timing alone rarely isolates one factor's effect, because several changes can precede the same outcome. An unresolved alternative limits how strongly the conclusion is supported; it does not show the proposed cause had no effect.
Worked example: A shop's sales rose after it both extended its hours and launched a loyalty card, and the report credits the extension alone. Because the card could also have lifted sales, the sequence does not isolate the extension's effect, so the causal credit remains uncertain.
Mistake to avoid: Attributing the whole observed change to one selected factor while ignoring the other plausible causes the evidence permits.
Reference: [3]
Statements and logical arguments
13. Negation reverses a statement's truth value
A logical statement is a complete sentence making a claim that can be judged true or false; its negation states the opposite, so exactly one of the pair is true. Negate 'The desk is occupied' as 'The desk is not occupied', keeping the same subject, context and time. Requests, questions and directives make no claim and cannot be negated this way.
Worked example: A hospital notice stating 'Ward 4 is closed for cleaning' is a logical statement, because it can be checked as true or false; its negation, 'Ward 4 is not closed for cleaning', has the opposite truth value.
Mistake to avoid: Treating a polite request such as 'Please occupy the desk' as a negation, even though requests make no claim that can be true or false.
Reference: [5]
14. Universal membership has a direction
In standard set logic, 'All A are B' means A is a subset of B: every member of A belongs to B, while B may contain members outside A. All archive volunteers wear badges, yet badge holders need not be volunteers. The claim also stays true if category A happens to be empty, so universality alone establishes no actual members.
Worked example: A library records that every member of its rare-books team completed preservation training. From this, a trainee list longer than the team is unsurprising, because other staff may also have completed the training without belonging to that team.
Mistake to avoid: Reversing the subset relationship by inferring from 'all volunteers have badges' that every badge holder must be a volunteer.
Reference: [5]
15. Existential claims establish an overlap
In standard set logic, 'Some A are B' means the intersection of A and B is not empty: at least one member belongs to both categories. It asserts nothing about proportion, so most cyclists might garden, and it does not imply that some cyclists necessarily do not garden. Any definition stated in the UCAT question takes precedence.
Worked example: At a sports club, one member who both cycles and gardens is enough to establish that some cyclists garden. The remaining cyclists' hobbies stay unknown, so nothing follows about whether most cyclists garden or whether any cyclist avoids gardening.
Mistake to avoid: Reading 'some cyclists garden' as a proportion, concluding that most cyclists garden or that some cyclists must avoid gardening.
Reference: [5]
16. Quantified negations change the claim's scope
In standard set logic, quantified statements are not negated by merely inserting 'not'. 'All A are B' is negated by 'Some A are not B', which breaks the subset relationship, while 'Some A are B' is negated by 'No A are B', which empties the intersection; the reverse pairings hold equally. Keep the same categories throughout.
Worked example: A courier firm finds one late parcel on Monday. This single exception is enough to disprove 'All Monday deliveries were punctual', because the negation only requires some deliveries to be late; claiming 'No Monday deliveries were punctual' would overstate the evidence.
Mistake to avoid: Replacing 'All deliveries were punctual' with the stronger 'No deliveries were punctual', which goes beyond breaking the subset relationship.
Reference: [5]
17. Validity and soundness answer different questions
Validity asks whether a conclusion follows logically from the premises, while soundness additionally demands that every premise be true. An argument can have a flawless structure yet rest on a false assumption, so candidates should separate structural correctness from factual accuracy when evaluating deductive arguments.
Worked example: Suppose a puzzle states that every amber key opens the storeroom, and a particular key is amber; the conclusion that it opens the storeroom is valid, but the argument is unsound if the puzzle's first premise is actually false.
Mistake to avoid: A common error is accepting a conclusion as factually true simply because the argument's structure is valid, ignoring whether premises hold.
Reference: [6]
18. Inductive patterns differ from deductive guarantees
Inductive reasoning generalises from a pattern of observed cases towards a broader conclusion, whereas a valid deduction has a conclusion that follows from its premises. Inductive arguments are judged strong or weak rather than valid, so observations that neither exhaust the population nor establish a separate proof support a pattern without guaranteeing it.
Worked example: A laboratory logs four consecutive fault-free batches. That pattern may support expecting the next batch to pass, but it does not deductively prove every future batch will pass, since four inspections neither exhaust the population nor establish a separate proof.
Mistake to avoid: Treating a run of consistent observations as deductive proof of an unrestricted rule, confusing inductive strength with certainty.
Reference: [6]
19. A satisfied condition yields its consequence
The law of detachment states that when a conditional 'if P, then Q' is true and P is true for the same case, Q must follow. The conclusion is limited to the consequence actually stated in the conditional, including any qualifications it carries.
Worked example: If a parcel is flagged fragile, a handling sticker is attached; this parcel is flagged fragile, so a handling sticker is attached. Nothing permits concluding that the parcel is also insured, because insurance is not part of the stated consequence.
Mistake to avoid: A frequent slip is inferring extra consequences beyond the conditional's stated outcome, such as assuming additional benefits not mentioned anywhere.
Reference: [6]
20. An absent consequence rules out its condition
The law of denying the consequent holds that when 'if P, then Q' is true and Q fails, P cannot hold. Because the conditional is equivalent to its contrapositive, the absence of the stated consequence rules out the condition under that rule.
Worked example: If a membership is renewed, a confirmation appears in the member's account; no confirmation appears in this member's account, so the membership was not renewed under that rule. The conclusion is non-renewal, not an unspecified delay in the system.
Mistake to avoid: Candidates sometimes weaken the conclusion to a possible delay, when the valid inference is that the condition itself did not occur.
Reference: [6]
21. Conditional chains require a matching middle
A valid chain links two conditionals only when the second condition of the first premise matches the first condition of the second premise in meaning. The conclusion is itself a conditional, so it does not assert that the starting condition actually occurred. Check that the shared middle term carries the same sense in both premises.
Worked example: Suppose anyone attending a placement signs its insurance form, and anyone signing that form receives a receipt. Chaining gives: anyone attending the placement receives a receipt. This stays conditional; it does not establish that anyone actually attended.
Mistake to avoid: Replacing actual attendance with mere permission to attend, so the middle terms no longer match and the chain fails.
Reference: [6]
22. A one-way rule does not work backwards
A single conditional states that membership in the first set guarantees membership in the second, like a subset relation. Affirming the consequent and denying the antecedent are invalid, whereas denying the consequent is valid. Extra premises could establish a two-way link, but the conditional alone licenses neither the converse nor the denial of the antecedent.
Worked example: A hospital rule says every on-call doctor carries a bleep. A bleep holder might be a porter, and a doctor off duty might still carry one. Without further information, seeing a bleep identifies neither role, so the converse inference fails.
Mistake to avoid: Reading 'if' as 'if and only if', then concluding that a badge holder must be a workshop leader.
Sets and probability
23. Intersection requires both; union permits either
The intersection of two sets collects only elements belonging to both sets, matching a logical AND. The inclusive union collects elements belonging to at least one set, and shared members still appear once. Unless the question explicitly uses exclusive wording, an element in both sets belongs to the union.
Worked example: A library records fiction titles as {3, 8, 12} and translated titles as {8, 12, 19}. Their intersection is {8, 12}, while the inclusive union is {3, 8, 12, 19}, listing each shared title once rather than excluding it.
Mistake to avoid: Dropping the shared element from an inclusive union, as if 'or' meant the sets could not overlap.
Reference: [7]
24. Union counts remove duplicated overlap
When both category counts include their shared members, adding the counts double-counts the overlap, so subtract the shared count once to obtain the number in at least one category. This formula covers two sets; disjoint sets need no subtraction because their shared count is zero.
Worked example: At a sports centre, 21 members attend yoga, 17 attend pilates and 5 attend both. The number attending at least one class is 21 + 17 − 5 = 33, because the five dual attendees were counted in both class totals.
Mistake to avoid: Adding 18 and 15 without subtracting the 6 who do both, reporting 33 swimmers or runners.
Reference: [7]
25. Complements depend on the stated population
A complement contains only members of the stated universal set that lie outside the selected set, so the universal set's total equals the subset's size plus its complement's size. Define the universal set first; otherwise outsiders may be wrongly counted, and 'neither activity' may be confused with 'not both activities'.
Worked example: A survey of 75 commuters finds 52 take the bus or train, so 23 use neither; the complement counts only surveyed commuters, not residents who never travel, and 'neither' differs from 'not both'.
Mistake to avoid: Counting townspeople who never joined the club as part of the complement, inflating the 'neither' total beyond the stated membership.
Reference: [7]
26. Brackets determine compound set operations
Parentheses are evaluated before other set operations, so any bracketed intersection or union is resolved first and its result combined with the remaining set. Preserve the grouping exactly as written; re-associating the expression can produce a different set than the stated grouping implies.
Worked example: With X = {1, 4}, Y = {4, 7}, Z = {7, 9}, the expression X union (Y intersection Z) yields {1, 4, 7}, whereas regrouping as (X union Y) intersection Z yields only {7}.
Mistake to avoid: Computing the outside union before the bracketed intersection, which silently replaces the stated grouping and can change the resulting set.
Reference: [7]
27. Conditional probability restricts the eligible outcomes
A conditional probability recalculates likelihoods after the conditioning event is known, shrinking the outcome space to cases where that event occurred. The condition must be possible, and the example assumes uniform selection within the restricted group; dividing by the unconditioned total ignores the updated denominator and distorts the probability.
Worked example: In a class of 30 pupils, 18 joined the orchestra and 6 of those orchestra members play the cello; choosing uniformly among orchestra members, the probability of selecting a cellist is 6/18 = 1/3, not 6/30.
Mistake to avoid: Dividing the favourable count by the whole class instead of the conditioned group, giving 6/30 rather than the correct 6/18.
Reference: [9]
28. Reversing a condition changes the denominator
P(A given B) and P(B given A) use different denominators, so the two conditional probabilities generally differ even when computed from the same population. The counts must describe one population with uniform selection, and although coincidence can make them equal, assuming equality without checking invites reversed-condition errors.
Worked example: At a clinic, 20 of 80 patients received physiotherapy and 15 of 80 received acupuncture, with 5 receiving both; P(acupuncture given physiotherapy) equals 5/20 = 1/4, while P(physiotherapy given acupuncture) equals 5/15 = 1/3.
Mistake to avoid: Swapping the condition and target event, reporting 1/3 as the probability of acupuncture given physiotherapy instead of 1/4.
Reference: [9]
29. Successive probabilities use the updated second stage
When one stage follows another, multiply the first-stage probability by the second-stage probability conditional on the first outcome. In draws without replacement, the pool shrinks, so the second denominator changes; the relevant numerator may stay the same, and the updated probability may increase or decrease, so reusing the first fraction is unreliable.
Worked example: A tray holds 5 green and 3 yellow bulbs. Two bulbs are drawn uniformly at random without replacement. The chance both are green is (5/8) × (4/7) = 20/56 = 5/14, because only 4 green bulbs remain among 7 after the first draw.
Mistake to avoid: Reusing the full first-stage fraction, such as 5/8, for the second draw even though one green bulb has already been removed.
Reference: [9]
30. Unordered outcomes combine separate tree paths
Some outcomes can occur in several distinct orders, and each order forms its own path through a tree diagram. Multiply the probabilities along each path, assuming each remaining item is equally likely to be selected, then add the path probabilities, because different orders are mutually exclusive. This applies whenever the requested outcome, such as one of each colour, ignores order.
Worked example: A box holds 4 orange and 3 purple beads, drawn uniformly at random twice without replacement. One of each colour has probability (4/7 × 3/6) + (3/7 × 4/6) = 4/7, adding the orange-then-purple and purple-then-orange paths.
Mistake to avoid: Counting only one drawing order, such as orange first, and omitting the equally valid reversed path from the total.
Reference: [9]
Ratios, proportions and rates
31. Ratio order identifies the comparison
A ratio compares two quantities in a stated order, and the requested order determines which number is written first. Reading the question carefully tells you whether the first quantity belongs in the numerator or denominator. Reversing the positions produces the reciprocal, which answers a different comparison even though the same two counts appear.
Worked example: A garage stocks 15 diesel engines and 10 petrol engines. The ratio of diesel to petrol engines is 15:10, which simplifies to 3:2, while the ratio of petrol to diesel engines is 2:3; each answers a different question.
Mistake to avoid: Writing the reciprocal ratio, such as 2:3 for diesel to petrol, simply because both counts appear in the question.
Reference: [8]
32. Part-to-part ratios need a total for shares
A part-to-part ratio compares two categories with each other, while a share of the whole compares one category with the total. To convert, add the ratio parts to find the total number of equal shares, then place the relevant part over that total. This works when the categories do not overlap and cover everything.
Worked example: A shelter houses dogs and cats in the ratio 4:3. If these are the only animals, dogs form 4/(4 + 3) = 4/7 of the residents, not 4/3, because the seven equal shares cover all animals.
Mistake to avoid: Reporting the ratio 4:3 as the fraction 4/3 of the total, confusing a part-to-part comparison with a part-to-whole share.
Reference: [8]
33. Equal ratios allow a missing value to be solved
When two ratios are set equal and both denominators are nonzero, the cross-products are equal, so a missing fourth value can be found. Corresponding quantities must occupy matching positions in the two ratios, and the equation should first be reduced to exactly two equal ratios. The worked example assumes a constant price with no fixed charge or quantity discount.
Worked example: At a fixed unit price, 5 pens cost £12; setting 5/12 equal to 18/x gives 5x = 216, so 18 pens cost £43.20, with matching quantities in matching positions.
Mistake to avoid: Cross-multiplying an equation that still contains three ratios or added terms, before it has been reduced to two equal ratios.
Reference: [8]
34. Unit rates determine multiplication or division
A unit rate expresses an amount per one unit of the denominator, so multiply by it to find an amount and divide by it to recover units. Attaching units to the rate distinguishes amount per item from items per amount. The example assumes a fixed unit price, and its cost and quantity calculations follow from the stated rate.
Worked example: At a fixed price of £6 for each mug, 8 mugs cost £48 by multiplying, and £42 buys 7 mugs by dividing, with units attached throughout.
Mistake to avoid: Dividing when multiplication is needed, or multiplying when division is needed, because the unit rate's direction was reversed.
Reference: [8]
35. Proportionality requires a constant ratio
A relationship is proportional only when corresponding quantities share one constant quotient, so divide paired values and compare before applying any constant-rate model. Stated rounding or measurement precision should be allowed for when comparing empirical ratios, and a constant-rate assumption needs support from the problem rather than following from values merely increasing together.
Worked example: Deliveries costing £5 for 1 item and £8 for 2 items give rates of £5 and £4 per item, so no single constant-rate model fits both sets of figures.
Mistake to avoid: Treating any relationship where both quantities increase as proportional without first checking that the quotients stay constant.
Reference: [8]
36. A scale links representation and reality
A stated linear scale converts represented lengths into real lengths and back, so keep the represented and real quantities clearly separate when calculating. The concept concerns linear distances only, and the example uses its own stated scale rather than a real map. Do not read a represented measurement as a real distance.
Worked example: A model railway uses a scale of 1 cm to 2 m, so a 9 cm carriage represents 18 m in reality; the represented length and the real length stay distinct throughout.
Mistake to avoid: Reading a represented model or map measurement directly as the corresponding real-world distance without applying the stated scale.
Reference: [8]
37. Time units must match a rate
Before multiplying by a rate, express the elapsed time in the units the rate uses. A speed quoted in kilometres per hour requires hours, so 2 hours 15 minutes becomes 2.25, not 2.15. This holds only while the rate stays constant across the interval; journeys with varying speed need the actual distance covered instead.
Worked example: A courier cycles at a steady 15 km per hour for 2 hours 45 minutes; converting 45 minutes to 0.75 hours gives 15 × 2.75 = 41.25 km, assuming the pace stays constant throughout.
Mistake to avoid: Reading 3 hours 20 minutes as 3.20 hours and multiplying the hourly rate by that figure, which understates the distance travelled.
Reference: [8]
Interpreting numerical displays
38. Category proportions use the dataset total
A category's proportion equals its frequency divided by the total number of units in the whole dataset, expressed as a fraction, decimal or percentage. The total must cover the same population the display represents; when two datasets have different totals, raw counts cannot be compared directly as shares.
Worked example: In a survey of 120 commuters, 45 travel by tram; the tram's share is 45/120 = 0.375, or 37.5%, because the denominator is every respondent surveyed, not the 75 who chose other options.
Mistake to avoid: Dividing a category's frequency by another category's count instead of the dataset total, producing a share that can exceed 100%.
Reference: [10]
39. Bar heights must be read against their axis
A bar chart's vertical axis may show either frequencies or proportions, so the printed label and tick intervals determine what a height means. A bar reaching 30 on a percentage axis indicates 30%, not 30 respondents; without a labelled axis, no numerical value can be inferred from height alone.
Worked example: A hospital audit chart labels its vertical axis in percentages with gridlines every 10%; a bar stopping midway between 20% and 30% is read as roughly 25%, and converting it to patient numbers needs the audit's total.
Mistake to avoid: Treating a bar's height as an exact respondent count when the axis is labelled in percentages, ignoring the stated units entirely.
Reference: [10]
40. Pie slices express shares of one whole
Each wedge's area shows its category's proportion of one complete dataset, so a labelled percentage converts to a count only when the chart's total is known. Slices from different pie charts with different totals cannot be compared as absolute amounts; use labels or a legend to identify each category.
Worked example: A library survey of 200 visitors shows a 12% wedge for audiobooks, corresponding to 24 visitors; the same 12% wedge in a 50-visitor survey would represent only 6, so the totals differ.
Mistake to avoid: Comparing a 20% slice from one pie chart with a 20% slice from another as equal absolute counts, despite different dataset totals.
Reference: [10]
41. Histogram bins give counts over intervals
In an equal-width frequency histogram, each bar summarises all observations falling within its interval, and its height reports only how many values lie there. The chart does not reveal the individual values inside a bin, so you cannot know where they sit within the interval or reconstruct an exact overall total from midpoints.
Worked example: A clinic's waiting-time histogram shows a frequency of 9 in the 15–20 minute bin; exactly nine patients waited somewhere between 15 and 20 minutes, and their individual times cannot be recovered from the chart.
Mistake to avoid: Reading a bin's frequency as the number of observations equal to that bin's midpoint value.
Reference: [10]
42. Stem-and-leaf plots retain individual observations
A stem-and-leaf plot lists every observation separately: each leaf is one data value, decoded with the plot's key, and repeated leaves mean repeated values. Count every leaf when finding totals, orderings or extremes, and never collapse a row into a single observation or ignore duplicated digits.
Worked example: With key 3|6 = 3.6, a row reading 3|2 2 8 records measurements of 3.2, 3.2 and 3.8, so the dataset contains three observations, two of which are identical repeated values.
Mistake to avoid: Counting each stem row as one observation or ignoring repeated leaves when totalling the data.
Reference: [10]
43. Rounded percentages may not sum to exactly 100
Exhaustive, mutually exclusive categories must sum to one before rounding, but displayed percentages rounded for presentation can total slightly above or below 100%. A small shortfall or excess is therefore a normal rounding artefact, not evidence of a hidden category, though large discrepancies deserve closer inspection.
Worked example: A hospital's discharge destinations are displayed as 41.7% home, 33.3% rehabilitation and 25.0% care home, summing to 100.0%; with three equal categories shown as 33.3% each, the 99.9% total reflects rounding alone.
Mistake to avoid: Treating a small 0.1% shortfall in displayed percentages as proof that a category was omitted.
Reference: [10]
44. A labelled value chart is not a frequency distribution
Some bar charts plot a measured quantity for each label, such as a branch or a year, so the vertical axis carries units rather than the word frequency. Bar heights are then compared as values, and differences are differences in that quantity, never counts of observations.
Worked example: A chart labels four clinics on the horizontal axis and shows scan costs of £60 and £75 at two of them; the £15 difference is a cost gap between clinics, not fifteen extra scans.
Mistake to avoid: Interpreting every labelled bar chart's vertical axis as a frequency count of observations per category.
Reference: [10]
Communication and consent
45. Communication should fit the individual
Effective communication means adapting to each patient's needs rather than relying on professional jargon or acronyms. GDC guidance asks the dental team to take individual views and communication needs into account, for example by using plain language or arranging an interpreter. A communication difficulty should never be mistaken for unwillingness to engage.
Worked example: A student notices a patient looking lost while she explains a filling using abbreviations. She restarts in plain language, asks which format suits him, and raises arranging interpreter support with her supervising dentist, addressing the barrier instead of treating it as disinterest.
Mistake to avoid: Repeating the same jargon-filled explanation more loudly and assuming a quiet patient is simply refusing to engage.
Reference: [11]
46. Understanding needs checking, time and questions
Providing information is only the start; patients also need time to consider it, opportunities to ask questions, and checks that they have understood. GDC guidance suggests asking questions and summarising the main points of the discussion. Silence or a quick nod does not establish understanding, and students should involve supervision where appropriate.
Worked example: After describing a crown option, a trainee dental nurse notices the patient summarised it incorrectly. She apologises for the confusion, restates the key points in simpler terms, invites further questions, and allows the patient time before any decision, without implying any fixed waiting period.
Mistake to avoid: Handing over an information leaflet, receiving a quick nod, and recording that the patient fully understood everything.
Reference: [11]
47. Honest answers support informed choices
When patients ask about their options, they deserve full and honest answers rather than reassuring evasions. GDC guidance requires explaining options, including delaying treatment or doing nothing, with the risks and benefits of each. Inventing certainty to make a choice easier undermines informed decision-making, and someone lacking knowledge should involve the responsible clinician.
Worked example: A patient asks a dental student whether delaying an extraction would cause problems. The student does not know the answer, so she avoids guessing, explains she will check with the supervising dentist, and returns with an honest account of the risks and benefits of waiting.
Mistake to avoid: Telling a worried patient that every option is equally safe just to end the conversation quickly.
Reference: [11]
48. Consent depends on discussion, not just a signature
A signature on a consent form verifies that consent was given, but the discussions with the patient determine whether it is valid. GDC guidance warns against assuming someone else obtained consent and requires checking and documenting understanding. Paperwork alone is never sufficient, and this guidance is not a complete legal test.
Worked example: Before a procedure begins, a dentist finds a signed form but no record that options were discussed. She pauses, checks what the patient understands about the treatment and its alternatives, documents the conversation, and continues once she is satisfied the consent reflects a genuine discussion.
Mistake to avoid: Proceeding with treatment because a signed form exists, without checking who discussed the options with the patient.
Reference: [12]
49. Consent must remain valid as care changes
Consent is a continuing process rather than a single event, so the responsible professional must confirm specific agreement for what is planned at each appointment. If the agreed treatment or estimated cost changes, the patient's consent to that change must be obtained and the discussion documented. Earlier agreement should never be treated as unlimited permission for later actions.
Worked example: Halfway through a course of treatment, a dentist proposes a different crown material that costs more than originally estimated. She explains the reasons and the new price, obtains the patient's agreement before proceeding, and records the discussion in the notes.
Mistake to avoid: Assuming a signed consent form covers every future appointment without confirming what the patient agrees to each time.
Reference: [12]
50. A patient's withdrawal of consent must be respected
Patients may withdraw consent at any time, even after treatment has started, and the professional must acknowledge that right and follow the patient's wishes. The consequences or risks of not continuing should be explained, including responsibility for future problems, and the whole discussion recorded in the notes. Pressure to continue or reliance on a signed form is inappropriate.
Worked example: During a filling, a patient raises a hand and asks the dentist to stop. The dentist stops immediately, explains the risks of leaving the treatment incomplete and the patient's responsibility for future problems, then records the conversation in the notes.
Mistake to avoid: Telling a patient who asks to stop that the signed form means treatment must continue regardless.
Reference: [12]
Records and confidentiality
51. Records and corrections must remain clear
Patient records must be contemporaneous, complete, accurate and readily understood by others, and any amendment must be clearly marked and dated. A dated, clearly identified correction keeps the account transparent and auditable, whereas silently altering or overwriting an entry conceals the change and undermines the reliability of the record.
Worked example: A dental nurse notices she entered the wrong tooth number in a record. She adds a clearly marked amendment stating the correct tooth, dates and initials it, so the change is transparent and the record remains accurate and understandable.
Mistake to avoid: Overwriting an incorrect entry so the amendment cannot be identified or dated, concealing the change.
Reference: [13]
52. Confidentiality extends beyond treatment details
Confidentiality covers all information learnt in a professional role, including personal details, treatment and costs, not only clinical findings. Professionals must not discuss patients where unauthorised people could overhear, and the duty continues after a patient dies. Identifying a patient and their appointment publicly breaches confidentiality even when no diagnosis is mentioned; exceptional disclosures need their own justification.
Worked example: A receptionist mentions a patient's full name and upcoming appointment while answering a call at the front desk, within hearing of the waiting room. Even without discussing any diagnosis, identifying the patient and the visit there breaches confidentiality.
Mistake to avoid: Discussing a patient's personal details in a public area, believing only diagnoses and clinical findings are confidential.
Reference: [13]
53. Secondary uses require permission and restraint
When patient information is wanted for teaching, research or similar secondary purposes, explain how it will be used, check the patient's understanding, obtain and record consent, and disclose only the minimum necessary. If identification is unnecessary, anonymise the material. Patients may withdraw permission at any time, and images or recordings receive the same protection as written records.
Worked example: A dentist asks a student's help with a teaching presentation, explains the purpose, records the patient's written consent, crops the photograph so no identifying features remain, and reminds the patient that permission can be withdrawn at any time.
Mistake to avoid: Treating a patient's consent for one teaching image as permission to share the entire record with an unrestricted audience.
54. Removing a name may not anonymise a case
Deleting a patient's name does not by itself make a case anonymous. Before discussing cases professionally on social media, assess whether remaining details, such as images, dates, locations or rare conditions, could still identify the person. The permission to discuss anonymised cases is narrow, applies only to best-practice discussion, and other consent and confidentiality duties continue to apply.
Worked example: A dental nurse drafts a professional-forum post about a rare condition, removing the patient's name but keeping a recognisable facial photograph, so colleagues could identify the patient and the post fails the anonymity condition.
Mistake to avoid: Assuming that deleting the patient's name makes a post safe when a recognisable photograph or other contextual details could still identify them.
Reference: [13]
55. Information security includes storage and transmission
Confidentiality can be broken without any intentional disclosure, so records must be stored where patients, unauthorised staff and the public cannot see them. Confidential information sent by any route should use a secure method, and electronic sending or storage should be encrypted, following organisational procedures and authorised systems. Encryption supports, but does not replace, checking recipients and restricting access.
Worked example: A receptionist finds a patient's printed medical history left on the open counter, returns it to the locked filing cabinet, and when emailing a referral attachment uses the practice's approved encrypted system rather than a personal email account.
Mistake to avoid: Believing confidentiality is secure because nothing was deliberately disclosed, while records sit visibly where unauthorised people can read them.
Reference: [13]
Teamwork and raising concerns
56. Respectful teamwork includes every contributor
Teamwork serves patients only when every member's contribution is valued, so colleagues must be treated fairly and respectfully in all situations and forms of communication, including non-registered staff such as receptionists. Respect does not mean overlooking unsafe behaviour, which should still be raised, and any patient-related discussion within the team must remain appropriately confidential.
Worked example: A receptionist warns the dentist that two children's appointments have been double-booked, delaying one child's fluoride treatment; the dentist thanks her, checks the schedule with her, and rearranges the appointments the same morning without any criticism.
Mistake to avoid: Dismissing a receptionist's scheduling concern as irrelevant because it comes from a non-clinical member of the team.
Reference: [14]
57. Delegating a task does not erase accountability
Under GDC standards, a dental professional who delegates a task retains accountability for its outcome, so delegation requires confidence that the recipient has been trained, is competent and holds appropriate indemnity. Seniority, availability or a colleague's willingness to help do not by themselves make delegation appropriate. The delegator remains answerable if something goes wrong, even though responsibility for performing the task has been transferred.
Worked example: A dentist asks a newly appointed dental nurse to take impressions although the dentist has not checked the nurse's training or indemnity. When the impressions prove faulty, the dentist remains accountable for the delegated task and cannot shift blame to the nurse.
Mistake to avoid: Assuming that handing a task to a senior or available colleague automatically transfers accountability for anything that subsequently goes wrong.
Reference: [14]
58. An assigned task still needs competence and suitability
A team member who receives a referral or delegation must be clear about what is being asked, confirm they are trained and competent, and judge the task appropriate for the patient. Pressure from a colleague does not resolve a skills gap, and an instruction alone never equals competence. If suitability is doubtful, discussing the request with the requesting colleague, or seeking advice, is the appropriate response.
Worked example: A dental therapist is asked to carry out an unfamiliar procedure while the supervising dentist is busy. She explains that she has not been trained for it and asks for supervision or a different arrangement rather than proceeding unaided under time pressure.
Mistake to avoid: Proceeding with unfamiliar work because a colleague pressured you, treating the instruction itself as proof of your competence.
Reference: [14]
59. Safety concerns need prompt attention, not proof
GDC guidance requires dental professionals to act on patient-safety concerns promptly and to raise them even when they cannot control the working environment. A concern does not need to be proven before investigation, and the duty to raise it overrides personal and professional loyalties. Uncertainty is not a reason for silence; when in doubt, the concern should be raised, distinguishing observed facts from allegations.
Worked example: A dental nurse notices a colleague repeatedly skipping sterilisation steps but worries about seeming disloyal. She records the specific observations and reports them promptly through the practice's concerns procedure, without waiting for a patient to be harmed or for proof.
Mistake to avoid: Staying silent about worrying behaviour until you can prove it, letting loyalty to a colleague delay protective action.
Reference: [15]
60. Escalation depends on who can act appropriately
An employer or manager is usually the first route for raising a concern, but that route may be unsuitable when the manager is the source of the concern or fails to act. Escalation then moves to the appropriate external body, which depends on jurisdiction, profession and severity, and severe concerns may justify direct referral to the GDC. Prompt protective measures remain necessary alongside reporting.
Worked example: A dentist reports potentially unsafe equipment to a manager who ignores the report; the dentist promptly escalates through the appropriate reporting route and, pending investigation, takes proportionate protective measures within their role, such as withdrawing the equipment from use.
Mistake to avoid: Relying solely on a manager-only route that the circumstances make inappropriate, or assuming one ignored report completes the response.
Reference: [15]
Sources
- UCAT Consortium: Current Test Format and Scoring
- UCAT Consortium: 2025 Changes and Official Withdrawal of Abstract Reasoning
- OpenStax Writing Guide: Thinking Critically About a Text
- OpenStax Writing Guide: An Author’s Choices
- OpenStax Contemporary Mathematics: Statements and Quantifiers
- OpenStax Contemporary Mathematics: Logical Arguments
- OpenStax Contemporary Mathematics: Set Operations
- OpenStax Contemporary Mathematics: Ratios and Proportions
- OpenStax Contemporary Mathematics: Conditional Probability and the Multiplication Rule
- OpenStax Contemporary Mathematics: Visualizing Data
- GDC Standards for the Dental Team, Principle 2: Communicate Effectively with Patients
- GDC Standards for the Dental Team, Principle 3: Obtain Valid Consent
- GDC Standards for the Dental Team, Principle 4: Maintain and Protect Patients’ Information
- GDC Standards for the Dental Team, Principle 6: Work with Colleagues in Patients’ Interests
- GDC Standards for the Dental Team, Principle 8: Raise Concerns if Patients Are at Risk