IMAT Logical Reasoning · Data Interpretation
Choosing, Scheduling and Judging Claims from Data
Strike out every option that breaks a condition before comparing what is left, follow timetables in one clock, and accept only the claims the data actually shows.
Why this matters
Choosing the cheapest item that meets several conditions is the commonest shape in this chapter, and the 2023 paper asked it. Timetables, time zones and "which statement does the data support" appear in the Cambridge papers.
Concept 1 of 3: Combining conditions: filter first, then optimise
A question that lists several conditions and then asks for the cheapest (or fastest, or lightest) option has two separate jobs. First remove every row that breaks any condition; only then compare the survivors. Doing it in the other order sends you straight to the cheapest row, which the setter has usually made break one condition.
Definition
- Step 1, filter: write the conditions as a short checklist and strike out each row that fails any of them.
- Step 2, optimise: among the rows left, compute what the question asks (total cost for the whole period, cost for the number needed) and choose the best.
- Boundary words: "at least 8" includes 8; "more than 8" does not. "Less than 1.5" excludes 1.5; "at most 1.5" includes it.
- Packs and bundles: you may need to buy more than you need; try mixtures of pack sizes and compare totals.
Worked example
A student compares five phone plans:
She needs at least 25 GB, 5G, and a contract of at most 12 months. Which plan is cheapest over a year?
| Plan | Monthly fee (€) | Data (GB) | Contract (months) | 5G |
|---|---|---|---|---|
| Alfa | 12 | 20 | 24 | No |
| Beta | 15 | 30 | 12 | Yes |
| Gamma | 10 | 15 | 12 | Yes |
| Delta | 18 | 50 | 1 | Yes |
| Epsilon | 14 | 25 | 24 | Yes |
Practice this conceptself-check · 3 quick reps
The cheapest row usually breaks a condition
Setters put the lowest price on an option that fails one requirement, often by a boundary such as exactly 1.5 kg for "less than 1.5 kg". Filter every row first; compare prices only among the survivors.
Concept 2 of 3: Timetables, schedules and time zones
A timetable is a table where each column is one bus or train and each row is a stop. Follow one column down to travel, and move to a later column when you have to wait. With time zones, the safe method is to convert every time into one clock (one city's local time), solve, and convert back only at the end.
Definition
- Reading: one column is one service; read down it for the arrival time at each stop.
- Connections: add any waiting or task time on arrival, then take the first service that leaves at or after that time.
- Durations: subtract in minutes; across midnight add 24 hours.
- Time zones: local time = reference time + offset. Put everything on one clock, check every condition there, then convert.
- Working hours in several cities: each gives a window on the common clock; the answer must lie in the overlap of all windows.
Worked example
Part of a bus timetable:
Sara reaches the Station at 07:35. She must collect a parcel at the Hospital, which takes 10 minutes, and be at the University by 09:00. Which buses should she take, and when does she arrive?
| Stop | Bus 1 | Bus 2 | Bus 3 | Bus 4 |
|---|---|---|---|---|
| Station | 07:10 | 07:40 | 08:15 | 08:50 |
| Hospital | 07:32 | 08:05 | 08:37 | 09:12 |
| University | 07:51 | 08:24 | 08:58 | 09:31 |
Practice this conceptself-check · 3 quick reps
Allow the waiting and task time before choosing the next service
If you arrive at 08:05 and need 10 minutes, you are ready at 08:15, and any bus before 08:15 has gone. Picking the next bus after your arrival time rather than after your ready time is the usual slip.
Convert to one clock before comparing times
Comparing a Dubai local time with a New York local time directly is meaningless. Put every time on one city's clock, check the conditions there, and convert only the final answer.
Concept 3 of 3: What the data can and cannot support
Some items ask which statement is supported by a table or chart. The right option says no more than the data shows. Wrong options go one step too far: they claim a cause, generalise from a sample, ignore a missing category, or read a ratio off bar lengths. Check each option against the numbers, and reject it if it needs anything the data does not give.
Definition
A statement is supported only if it follows from the numbers shown, with no extra information.
- Data showing two things change together does not show that one causes the other.
- A sample describes the group measured, not everyone.
- Words such as all, never, only are refuted by a single counter-example in the table.
- "Most" or "a majority" means more than half, not just the largest group.
- A rate (per 1,000, per person) can fall while the count rises, and the other way round.
| Claim in an option | What you must check | Common reason it fails |
|---|---|---|
| X increased over the period | Values at the start and end, in the same units | Only part of the period is shown, or the rise is in a different measure |
| X is twice Y | The actual values, not the bar lengths | The chart's axis does not start at zero |
| X caused Y | A controlled comparison where only X differs | The data only shows that X and Y occur together |
| Most people chose X | X is more than half of the total | X is the largest group but under half |
| X always or never happens | Every row of the table | One row is a counter-example |
Use the table as a checklist for each option in a "which statement is supported" item.
Practice this conceptself-check · 3 quick reps
Together is not cause
Two columns rising together (cycle lanes and cyclists) do not show that one causes the other. A statement claiming a cause is not supported by a table of figures alone.
Count and rate can move in opposite directions
Accidents rose from 60 to 65 while the rate per 1,000 cyclists fell from 15 to 10. "More dangerous" is about the rate, so read which one a statement is about.
Summary — formulas & gotchas at a glance
A revision cheat-sheet for the formulas and gotchas above. Click any concept name to jump back to its full explanation.
Reference tables (1)
What the data can and cannot support5 rows
| Claim in an option | What you must check | Common reason it fails |
|---|---|---|
| X increased over the period | Values at the start and end, in the same units | Only part of the period is shown, or the rise is in a different measure |
| X is twice Y | The actual values, not the bar lengths | The chart's axis does not start at zero |
| X caused Y | A controlled comparison where only X differs | The data only shows that X and Y occur together |
| Most people chose X | X is more than half of the total | X is the largest group but under half |
| X always or never happens | Every row of the table | One row is a counter-example |
Use the table as a checklist for each option in a "which statement is supported" item.
Watch out for (5)
- The cheapest row usually breaks a condition→ Combining conditions: filter first, then optimise
- Allow the waiting and task time before choosing the next service→ Timetables, schedules and time zones
- Convert to one clock before comparing times→ Timetables, schedules and time zones
- Together is not cause→ What the data can and cannot support
- Count and rate can move in opposite directions→ What the data can and cannot support
Drill the past-year questions on this topic
Open the bank with this topic pre-filtered.