A data sufficiency question gives you a question and two statements, and asks a different thing from every other quantitative question you have met: can this be answered? Not what the answer is.
That sounds like a smaller job and it is, which is why the question type is fast once the habit is built. The difficulty is entirely psychological: you will want to work out the answer, and working it out is wasted time.
A caution first. Nothing has been confirmed for CLAT 2028. The Expert Committee Report 2026 proposes keeping Quantitative Techniques as one of three sections and gives data sufficiency as its first sample question type, alongside "conceptual numeracy" and "quantitative and representational logic" (pp. 43–46). The Consortium has referred the recommendations for further consideration. The CLAT 2028 page separates the proposals from the decisions. The skill below is worth having regardless, because deciding whether given information supports a conclusion is a reasoning skill, not a topic.
The standard format
A question, then two statements, then options along these lines:
- A. Statement 1 alone is sufficient, but statement 2 alone is not.
- B. Statement 2 alone is sufficient, but statement 1 alone is not.
- C. Both together are sufficient, but neither alone is.
- D. Each alone is sufficient.
- E. Both together are not sufficient.
This is the option set the Expert Committee's own sample question uses, adapted from GMAT practice material (Report, p. 44). If the type is ever adopted, read the paper's instructions before assuming the letters mean what they mean here.
The method
Four steps, in this order, and the order matters.
One: read the question and decide what would count as an answer. A single number? A yes or no? This is the step people skip, and it causes most errors, because sufficiency depends entirely on what is being asked.
Two: test statement 1 alone. Cover statement 2 with your hand — literally. Contamination between the statements is the second commonest error and it is almost impossible to notice once it has happened.
Three: test statement 2 alone. Now cover statement 1.
Four: only if both fail alone, test them together.
At every step, the question is "does this determine a unique answer?" — not "what is it?"
The traps
Calculating. You do not need the value. If statement 1 gives you two equations in two unknowns, it is sufficient; solving them is unpaid work.
Carrying information between statements. You read statement 1, form a picture, then evaluate statement 2 with statement 1 still in your head. Cover them physically until the habit is built.
Forgetting that "no" is an answer. For a yes-or-no question, a statement that establishes the answer is definitely no is sufficient. Candidates mark insufficient because the answer was not yes.
Missing the second possibility. A statement that permits two values is insufficient. The classic version: something that yields x² = 16 gives you 4 and −4, and only context rules one out.
Assuming from the diagram or the wording. Nothing is true because it looks true. Integers, positives, distinctness — all have to be stated.
Worked examples
Example 1. What is the value of x?
(1) x² = 49 (2) x > 0
Statement 1 alone: x is 7 or −7. Two values, so not sufficient — the trap, because it feels like it has told you a great deal.
Statement 2 alone: infinitely many values. Not sufficient.
Together: x = 7. Unique. Answer: C.
Example 2. Is n an even integer?
(1) n is divisible by 6. (2) n is an integer, and n² is even.
Statement 1: divisible by 6 means divisible by 2, so yes, even. A definite answer — and the answer being "yes" is not what makes it sufficient; a definite "no" would do equally well. Sufficient.
Statement 2: for an integer, an even square means an even number, so yes. Also a definite answer. Sufficient. Note why the statement has to say "integer": without it, n = √2 gives n² = 2, which is even, and the answer would be no.
Each alone works. Answer: D. Candidates who stop after finding statement 1 sufficient mark A, which is why step three is not optional.
Example 3. A shop sold 200 items. What was the average price?
(1) Total revenue was Rs. 50,000. (2) The most expensive item cost Rs. 800.
Statement 1: 50,000 ÷ 200. Note that you have now established sufficiency — do not do the division. Sufficient.
Statement 2: tells you nothing about the average. Not sufficient.
Answer: A.
Example 4. Is quadrilateral ABCD a square?
(1) All four sides are equal. (2) All four angles are equal.
Statement 1: a rhombus satisfies this and is not a square. Not sufficient.
Statement 2: a rectangle satisfies this and is not a square. Not sufficient.
Together: equal sides and equal angles — a square. Answer: C.
This is the archetypal structure, and it is worth noticing why it works: each statement removes one of the two ways a quadrilateral can fail to be a square.
Practising it
Do the steps out loud at first, including "cover statement 2". The discipline is the skill and it takes about fifty questions to become automatic.
Stop the moment you know. Time yourself and notice how often you kept calculating after the answer was settled. That is the leak.
Keep an error log by type: contaminated statements, forgot the negative root, calculated unnecessarily, misread what counted as an answer. Four categories will cover almost everything, and the distribution tells you what to fix.
Whether to spend time on this now
Sitting CLAT 2027: the pattern is frozen, and the section is data interpretation from passages. Not a priority. The Quantitative Techniques guide covers what is actually on the paper.
Sitting CLAT 2028: nothing is confirmed, and the proposal would make the quantitative section roughly a third of the paper rather than a tenth — which is a reason to be better at the section generally, not to drill an unconfirmed format. The Class 11 plan covers the proportions.
For a current measurement, the free Diagnostic Test includes a full Quantitative Techniques section and shows you afterwards how long it actually took you.