Inference questions in CAT RC: how far is too far?

Inference questions are where most CAT RC marks are lost. The stem sounds generous ("It can be inferred that…", "The author would most likely agree…"), so candidates reach for an option that feels reasonable. Then they find that the correct answer was a much smaller step from the text than the one they chose.

The skill is judging distance: how far an option travels from what the passage actually says. This guide gives you a scale for that distance, a test to apply, and worked passages where every wrong option is explained.

What an inference is in CAT

In everyday speech, an inference can be a clever guess. In CAT, it is something the passage implies so strongly that it is close to certain, even though it is not stated in those words.

A useful way to hold this: a CAT inference is a conclusion that the author would accept without argument, because it follows from what they have already written.

The degrees of support

Every option in an inference question sits somewhere on this scale.

The most dangerous options sit in "probably" and "could be true". They do not contradict the passage, so they feel safe. But the passage does not support them either.

The must-be-true test

For each option, ask: if everything in the passage is true, must this also be true?

• If yes, keep it.
• If it only might be true, it is too weak.
• If it needs outside knowledge, it is out of scope.
• If you need to add a step the author never took, it is over-inference.

Then, for the option you keep, find the line or lines that support it. If you cannot point to them, you have not made an inference; you have made a guess.

A warm-up example

Many cities that built wider roads to reduce traffic found that congestion returned within a few years. Planners describe this as induced demand: when driving becomes easier, more people choose to drive, and the new capacity fills up.

Which can be inferred?

(a) Wider roads always make traffic worse.
(b) Adding road capacity may not reduce congestion in the long run.
(c) Planners oppose building any new roads.
(d) Public transport is the only solution to congestion.

Answer: (b). (a) is too extreme: congestion "returned", which is not the same as getting worse, and "always" goes beyond "many cities". (c) attributes a position to planners that the passage never gives. (d) brings in a solution the passage does not discuss. (b) follows directly, and "may" matches the passage's strength.

A worked passage

Every map is an argument. To draw one is to decide what counts: which rivers are named and which left blank, which settlements appear as dots and which vanish into empty green. Cartographers have long insisted that their work is a neutral record, and in a narrow sense it can be. The distance between two towns is what it is. But the choice to measure distance rather than, say, the time it takes to walk between them during the monsoon is not a neutral one. It reflects the needs of whoever commissioned the map, most often a state that wanted to tax, defend or govern the land it was drawing.

This is why maps made by the people who live in a place so often look strange to outsiders. They may place a well at the centre and a highway at the edge, or show a hill that no survey records because it is too small to measure but too steep to cross with a loaded cart. Such maps are not less accurate than official ones. They are accurate about different things.

Question 1

It can be inferred that the author believes an official map:

(a) is usually less accurate than a local map.
(b) reflects choices shaped by the purposes of those who commissioned it.
(c) deliberately falsifies distances.
(d) is of no use to people who live in the area.

Answer: (b). The first paragraph says the choice of what to measure "reflects the needs of whoever commissioned the map".

• (a) reverses the comparison. The author says local maps are "not less accurate", which does not mean official maps are less accurate.
• (c) contradicts "The distance between two towns is what it is." The author questions the choice of measure, not its truth.
• (d) is extreme and unsupported.

Question 2

What can be inferred about the hill mentioned in the second paragraph?

(a) It is too small to matter to anyone.
(b) Official surveyors were careless in missing it.
(c) Its importance to local people comes from its effect on travel, not its size.
(d) It appears on most official maps.

Answer: (c). It is "too small to measure but too steep to cross with a loaded cart". Its size is minor; its effect on moving goods is what matters.

• (a) contradicts the passage: it matters enough to appear on local maps.
• (b) blames the surveyors, which the author never does. The hill was left out because of what surveys measure, not through carelessness.
• (d) contradicts "no survey records".

Question 3

The claim that "every map is an argument" rests on which assumption?

(a) Making a map involves choices about what to include and what to leave out.
(b) Cartographers intend to mislead their readers.
(c) All arguments are biased.
(d) Local maps are better than official ones.

Answer: (a). The author supports the claim immediately with "To draw one is to decide what counts".

• (b) confuses choice with deceit. The author allows that maps can be neutral "in a narrow sense".
• (c) is far broader than anything the passage says about arguments.
• (d) contradicts the conclusion that the two kinds are "accurate about different things".

The traps, named

Practice set

A company that rewards call-centre staff for keeping calls short will get short calls. It will not necessarily get solved problems. Customers whose issue was not resolved will simply call again, and each of those calls will also be short. On the dashboard, the measure improves month after month; in the queue, the same customers keep returning. The difficulty is not that measurement is bad, but that a single number, once it is tied to rewards, begins to describe the behaviour it rewards rather than the outcome it was meant to track.

1. Which of the following can be inferred from the passage?
   • (a) Call centres should stop measuring call length.
   • (b) A measure can keep improving while the problem it is meant to track does not.
   • (c) Staff deliberately mislead customers to end calls quickly.
   • (d) Long calls are always better than short ones.
2. The author would most likely agree that:
   • (a) measures tied to rewards should be checked against the outcomes they are meant to reflect.
   • (b) numbers cannot describe behaviour.
   • (c) customers prefer long calls.
   • (d) dashboards are misleading by nature.
3. What does the passage imply about the repeat calls?
   • (a) They help the call-length figure look good while the underlying problem continues.
   • (b) They are recorded as solved problems.
   • (c) They are caused by rude staff.
   • (d) They raise the average length of calls.

Answers:

1. (b). The dashboard improves "month after month" while "the same customers keep returning". (a) is extreme; the author says the difficulty "is not that measurement is bad". (c) imputes a motive. (d) uses "always" and is never claimed.
2. (a). It follows from the final sentence: a rewarded number drifts from the outcome it tracks, so it needs checking against that outcome. (b) distorts the claim; numbers do describe behaviour, which is the problem. (c) is out of scope. (d) is too broad; the author blames the link to rewards, not dashboards.
3. (a). Repeat calls are "also short", so they keep the measure looking healthy while problems stay unresolved. (b) is not stated. (c) is outside the passage. (d) contradicts "also be short".

What to do next

• For every inference question you practise this week, write the supporting line next to your answer.
• Place each wrong option on the degrees-of-support scale.
• Revise the full list of RC question types and try critical reasoning questions, which use the same skill.