Last updated July 07, 2026 · Reviewed by the AnvayaPrep team
Introduction
Data Sufficiency is the most distinctive question type on the GMAT and a core component of the Data Insights section. These questions do not ask for a computed answer. They present a question followed by two statements and ask which combination of those statements provides enough information to answer the question. The test-taker must evaluate the sufficiency of each statement, alone and in combination, without necessarily performing the final calculation.
Topics in this unit include the data sufficiency format itself, statement 1 analysis, statement 2 analysis, combined statements, rephrasing the question, variable constraints, hidden constraints, sufficiency logic, testing values, word problems in data sufficiency format, inequalities in data sufficiency, equality determination, algebra in data sufficiency, number properties in data sufficiency, geometry in data sufficiency, data sufficiency traps, and the five standard answer choices.
Data Sufficiency questions account for approximately 30-40% of the Data Insights section and represent a fundamentally different cognitive task from computation. They test whether a student understands what information is needed to solve a problem, not just whether the student can perform the calculation. This mirrors real business reasoning, where determining what data to collect is often as important as analyzing it once collected.
The five standard answer choices never change and must be memorized: (A) Statement 1 alone is sufficient; (B) Statement 2 alone is sufficient; (C) Both together are sufficient but neither alone is; (D) Each statement alone is sufficient; (E) Neither statement alone nor both together are sufficient.
Learning Objectives
- Identify data sufficiency questions and understand their structure: question stem, Statement 1, Statement 2, and five fixed answer choices
- Apply the AD/BCE decision tree systematically: evaluate Statement 1 first, then Statement 2, then the combination if necessary
- Rephrase complex question stems into simpler mathematical targets before evaluating statements
- Determine sufficiency for value questions (must yield exactly one possible answer) and yes/no questions (must yield a definitive yes or definitive no)
- Recognize hidden constraints in question stems such as "n is a positive integer" or "n is even," which restrict the solution space beyond what the statements explicitly state
- Apply the testing-values technique: try multiple valid numerical examples to determine whether a statement always yields the same answer or varies
- Avoid the five major data sufficiency traps: assuming two statements are always independent, confusing sufficiency with computation, accepting answers that work for some cases but not all, using information from one statement to evaluate the other, and treating "cannot determine" as automatically (E)
- Apply data sufficiency reasoning across algebraic, geometric, statistical, and number-property contexts
High-Yield Concepts
| Concept | Rule or Principle | Common GMAT Trap |
|---|---|---|
| The Five Answer Choices | A, B, C, D, E with fixed meanings; memorize as "12TEN": 1 alone, 2 alone, Together, Either alone, Neither | Mixing up C (both together sufficient) with D (each alone sufficient) |
| AD/BCE Decision Tree | If S1 sufficient, answer is A or D; if not, answer is B, C, or E; saves systematic evaluation time | Evaluating S2 before confirming S1 status, leading to confusion |
| Value Questions | Sufficient means exactly one numerical answer is possible; "could be 3 or 5" is insufficient | Accepting a statement that narrows the answer to two or three values as sufficient |
| Yes/No Questions | Sufficient means always yes or always no; "sometimes yes, sometimes no" is insufficient | Selecting a statement that gives yes for some values and no for others as sufficient |
| Rephrasing | Transform the question to the simplest equivalent form before evaluating statements | Evaluating statements against the original complex form instead of the rephrased version |
| Hidden Constraints | Question stem constraints (integer, positive, even) limit which statement values apply | Ignoring constraints in the question stem that eliminate some potential statement values |
| Testing Values | Try two different values satisfying a statement; if they yield different answers, the statement is insufficient | Testing only one value and assuming the statement works for all |
| Combined Statements | Only evaluate together if each is individually insufficient; never use S1 information when evaluating S2 alone | Using information from both statements when evaluating one statement alone |
Rephrase the question before reading the statements. Many data sufficiency questions contain a complex question stem that becomes simpler when algebraically manipulated. For example, "What is the value of 3x + 6?" rephrases to "What is x?" (since knowing x is equivalent to knowing 3x + 6). This rephrasing makes statement evaluation faster and more accurate.
The most dangerous data sufficiency habit is contaminating the individual-statement evaluation. When evaluating Statement 2 alone, you must pretend Statement 1 does not exist. Students who mentally combine both statements while evaluating each individually will overcount sufficiency and incorrectly select C when the answer is A or B.
Study Strategy
Begin by thoroughly memorizing the five answer choices and the AD/BCE decision tree. These are the procedural scaffolding that makes every other skill faster and less error-prone. Practice retrieving the answer-choice meanings instantly so that during the exam, cognitive resources go to the mathematical analysis, not to remembering what choice (C) means.
Next, study the value question and yes/no question distinction, because the sufficiency standard differs between them. For value questions, a statement is sufficient only if it forces exactly one numerical answer. For yes/no questions, a statement is sufficient if the answer is always definitively yes or always definitively no. Drilling these two cases separately before mixing them prevents confusions.
Rephrasing is the highest-leverage technique for time savings. Practice identifying equivalent forms of question stems across algebra, geometry, and number properties. A question asking "Is x^2 > 25?" rephrases to "Is |x| > 5?", which rephrases to "Is x > 5 or x < -5?" This three-step simplification chain makes the statements much easier to evaluate.
The testing-values technique is the safeguard for any statement involving variables or inequalities. Systematically pick two values that satisfy the statement's constraints and check whether both yield the same answer to the question. If they differ, the statement is insufficient. Choose "nice" and "extreme" values: for an integer constraint, test a small positive integer and a larger one; for an inequality, test the boundary and a value well inside the range.
Common Mistakes
Forgetting the question stem's constraints: Constraints like "n is a positive integer" or "x is a prime" are part of the given information and function exactly like a third statement. Ignoring them leads to testing inadmissible values and incorrect sufficiency conclusions.
Selecting C when A or B is correct: This occurs when students mentally combine statements while evaluating each individually. Each statement must be evaluated in complete isolation. Cover the other statement while evaluating the current one.
Confusing "insufficient" with "I cannot calculate it": If a statement limits the answer to exactly one value, it is sufficient even if that value is complex to compute. The question is never "can I calculate the answer easily?" but "is there only one possible answer?"
Accepting a statement that narrows but does not fix the answer: If Statement 1 tells you x is either 3 or 5, the statement is insufficient for a value question even though it narrows the range. Sufficiency requires exactly one answer, not a reduced set of possibilities.
Selecting E too quickly for yes/no questions: A yes/no question can be answered sufficiently if the answer is "always no." "Always no" is as definitive as "always yes." Students sometimes dismiss statements that consistently produce "no" answers as insufficient.
Exam Tips
Use the decision tree structure consistently. Evaluate S1 first; record whether it is sufficient (AD zone) or not (BCE zone). Then evaluate S2 alone; record its status. Then, if necessary, combine. This procedure eliminates the need to reconsider answer categories and keeps the reasoning organized.
Budget 1.5-2 minutes per data sufficiency question. Rephrasing takes 15-20 seconds. Evaluating each statement individually takes 30-45 seconds each. If you are spending more than 2 minutes, choose the most defensible answer among the remaining candidates and move on.
Watch for answer choice D on "each alone" traps. The GMAT occasionally makes both statements independently sufficient, but each is designed to look insufficient if analyzed carelessly. When you correctly determine A is correct, double-check by asking whether S2 would also work. If yes, the answer is D.
In number property questions, consider odd/even, positive/negative, and integer/non-integer cases. These are the most common value-testing categories. For any question about properties of n, systematically test positive odd, positive even, negative, and zero if applicable to see whether the answer changes.
Memorize the answer choices as "12TEN": (1) alone, (2) alone, (T)ogether, (E)ither alone, (N)either. The decision tree: Start with S1. If S1 sufficient, answer is 1 or E(ither) [A or D]. If S1 not sufficient, eliminate A and D, leaving 2-T-N [B, C, E]. Then check S2. Then, if needed, combine.
Sign up free to keep reading
Create a free AnvayaPrep account to finish this GMAT guide on Data Sufficiency — plus flashcards and practice questions.