Mensa membership requires scoring at or above the 98th percentile on an accepted intelligence test — a threshold that fewer than one in fifty people meet. If you are preparing to take the Mensa Admission Test or planning to submit prior test scores for qualification, practice is the most direct way to close the gap between your current performance and the score you need. This free mensa practice test PDF gives you a printable collection of IQ-style questions spanning every reasoning domain tested on the Mensa exam: number sequences, matrix reasoning, verbal analogies, spatial visualization, and logical reasoning. Print it out, work through it at your own pace, and use it to identify which question types cost you the most time or accuracy.
Whether you prefer to study offline, want to work through problems without a screen in front of you, or simply find that pen-and-paper practice feels different from clicking through digital tests, this PDF gives you that option. After completing the printed questions, return to our full question bank on the mensa practice test page for timed, scored sessions that simulate the actual test environment and help you build the processing speed the Mensa Admission Test demands.
American Mensa accepts qualifying scores from two sources: a supervised Mensa-administered test, or qualifying evidence from a prior standardized IQ test taken under acceptable conditions. Understanding both pathways helps you choose the right preparation strategy.
The Mensa Admission Test is a timed, in-person test administered at testing sessions held by local Mensa chapters across the United States. The test is supervised by a proctor and delivered in a group setting. It consists of two separate timed sections, each containing approximately 50 questions. The sections measure different cognitive abilities — one tends to be more language-based and the other more visually oriented — and a qualifying score on either section is sufficient for membership. You do not need to score at the 98th percentile on both sections; exceeding the threshold on one is enough.
Mensa does not publish the exact contents of its Admission Test for obvious reasons, but the organization explicitly states that the test measures cognitive ability rather than acquired knowledge. Preparation that improves your pattern recognition, abstract reasoning, and processing speed will transfer directly to test-day performance.
If you have taken an accepted IQ test in the past and received a qualifying score, you can submit that evidence to Mensa in lieu of taking the Admission Test. Accepted tests include the Stanford-Binet Intelligence Scales (5th Edition), the Wechsler Adult Intelligence Scale (WAIS-IV) and Wechsler Intelligence Scale for Children (WISC-V), the Wonderlic Personnel Test (a score of 35+ is typically the qualifying threshold), and many others listed on the American Mensa website. Scores must come from a licensed psychologist or approved testing authority, and some tests are accepted only if administered before a certain age or within a certain time window. Group-administered tests, such as the SAT, ACT, and GRE, are no longer accepted for qualifying submissions as of the most recent Mensa policy.
Number sequence questions ask you to identify the rule governing a series of numbers and predict the next value in the series. These questions test inductive reasoning — the ability to identify a pattern from examples and extend it to a new case. Patterns can involve simple arithmetic progressions (each term increases by a fixed amount), geometric progressions (each term is multiplied by a fixed ratio), alternating rules (even-positioned terms follow one rule, odd-positioned terms follow another), or combinations of operations.
Start by finding the difference between consecutive terms. If the differences are constant, you have an arithmetic sequence — the next term is the last term plus that constant. If the differences are not constant, find the second differences (differences of the differences). Constant second differences indicate a quadratic (squared) sequence. If neither approach works, look for a multiplicative pattern — are terms doubling, tripling, or being multiplied by a fraction? For alternating sequences, separate the odd-positioned and even-positioned terms and analyze each sub-sequence independently. On high-difficulty IQ tests, sequences may involve two interleaved rules operating simultaneously, so always consider whether the series could be two separate sequences merged together.
Speed matters on timed IQ tests. With practice, common sequence types become immediately recognizable — you stop re-deriving the rule from scratch and instead pattern-match to a familiar structure. This is why repeated practice with varied sequences is more valuable than studying any single solving technique.
Matrix reasoning questions present a 3x3 grid of figures (or sometimes a 2x2 grid) with one cell missing, typically the bottom-right. You must identify the rule or rules that govern the rows and columns of the matrix and select the figure that completes the pattern from a set of answer choices. These questions are a standard component of virtually every modern IQ test and are considered among the most culturally fair measures of fluid intelligence because they do not depend on language or factual knowledge.
The most common transformation rules in matrix questions are: quantity (the number of elements in each cell increases or decreases across the row or down the column), shape transformation (shapes rotate by a fixed angle, reflect, or change size), shading (cells alternate between filled and unfilled, or follow a rotation of shading positions), and combination (elements from two cells are combined or subtracted to produce a third cell). On more difficult items, multiple rules operate simultaneously — you must track changes in shape type, number, orientation, and shading all at once. A reliable strategy is to work row by row first: identify what changes from left to right across each row, then verify that the same rule applies down each column. The missing cell must satisfy both the row rule and the column rule.
Verbal analogy questions follow the format "A is to B as C is to ?" and test your ability to identify the relationship between a word pair and apply it to a new pair. The relationship can be semantic (synonyms, antonyms, category membership), functional (tool to its use, part to its whole), or structural (grammatical, degree of intensity). High-difficulty verbal analogies on IQ tests tend to use less common vocabulary and require distinguishing between multiple valid-seeming relationships to find the most precise match.
Create a precise sentence that expresses the A-to-B relationship, then apply that sentence to C and see which answer choice fits. The more specific the sentence, the easier it is to eliminate wrong answers. For example, if the pair is "glove : hand," the sentence "a glove covers a hand" is more useful than "glove and hand go together." Apply it: "a _____ covers a foot" → shoe. Avoid choosing an answer based solely on word association — the answer must fit the precise relationship expressed in your sentence. When two answer choices seem to fit, the one that matches the relationship more specifically and in the same structural way is almost always correct.
Spatial reasoning questions measure your ability to mentally manipulate two-dimensional and three-dimensional objects. Types commonly found on IQ tests include mental rotation (identifying which of several answer choices shows a given 3D figure rotated to a new orientation), paper folding (predicting where holes will appear when a folded piece of paper is punched and then unfolded), pattern folding (determining which 3D shape a flat net will form when folded), and embedded figures (identifying a simple shape hidden within a more complex figure).
Spatial reasoning is more trainable than most people assume. Research consistently shows that practice with spatial tasks produces significant performance gains, particularly for people who start with weaker spatial skills. For mental rotation, practice rotating shapes in your mind without moving your hands — physically rotating the paper is a common crutch that slows you down on timed tests. For pattern folding, practice identifying which edges will join when the net is folded: mark matching edges and track which faces end up adjacent. For embedded figures, use a systematic scanning strategy — trace the target shape's outline with your eyes starting from a distinctive feature (a corner, a long edge) and look for that same feature in the complex figure.
Logical reasoning items on IQ tests present a set of premises and require you to identify a conclusion that must follow (deductive reasoning) or is most strongly supported (inductive reasoning). Deductive problems include syllogisms: "All A are B. All B are C. Therefore, all A are C." These are valid regardless of whether the premises are actually true — validity is about whether the conclusion follows from the premises, not whether the premises describe reality. Common errors include affirming the consequent ("If A then B; B is true; therefore A is true" — this is invalid) and denying the antecedent ("If A then B; A is false; therefore B is false" — also invalid).
Inductive reasoning problems present examples and ask you to choose the most reasonable generalization. The correct answer is the one that is best supported by all the given examples and is not contradicted by any of them. Strong distractors on these items are generalizations that are supported by some examples but not all — always check the answer you are considering against every example in the problem before selecting it.
IQ scores are designed to follow a normal (bell-curve) distribution with a mean of 100 and a standard deviation (SD) of 15 on most modern scales. The 68-95-99.7 rule describes how scores distribute: approximately 68% of people score within one SD of the mean (IQ 85–115), about 95% score within two SDs (IQ 70–130), and about 99.7% score within three SDs (IQ 55–145). The 98th percentile — the Mensa threshold — corresponds to a score that is approximately 2.05 standard deviations above the mean, which equals an IQ of roughly 130–131 on a 15-SD scale.
On the Wechsler scales, a full-scale IQ of 130 places a person at approximately the 98th percentile. On the Stanford-Binet 5th Edition, which also uses SD 15, the same percentile corresponds to the same score. The Wonderlic uses a different scoring system; a score of 35 on the 50-item Wonderlic is roughly equivalent to the 98th percentile on general intelligence. When preparing, focus on the specific test you plan to take — the format, timing, and item types vary meaningfully across instruments, and familiarity with the format reduces the time you spend figuring out what you are supposed to do instead of actually solving problems.
The 98th percentile threshold is fixed, but your score is not — consistent, targeted practice with the specific reasoning types on the Mensa exam reliably raises performance. Use this PDF for offline study and return to the timed question bank on our mensa practice test page to simulate real test conditions, track your accuracy by question type, and identify exactly where to focus your remaining preparation time.