MAT Exam: Format, Scoring, and How to Prep Smart

The Miller Analogies Test (MAT) is a high-level analytical exam used by hundreds of graduate programs across the United States and Canada to gauge a candidate's reasoning, vocabulary breadth, and conceptual flexibility. Unlike multi-section standardized tests built around math, reading, and writing blocks, the MAT is a single 60-minute sprint composed of 120 analogy questions.

The format sounds simple on paper, but every analogy demands that you decode a relationship, hold it steady, and then map it onto unfamiliar terms from history, philosophy, mathematics, biology, art, literature, music, mythology, and more. You're here because you want a straight answer about what the test is, how scoring works, and whether you can actually prepare for something that pulls from such a wide knowledge pool.

Short version: yes, you can — but only with structured practice, careful timing, and a clear understanding of how analogies are built. This guide pulls together the format, scoring scale, content domains, and study tactics real test-takers use to move from a 400 to a competitive 425+ score.

The MAT practice test on this site mirrors the live exam's structure, timing, and relationship types, so you can rehearse under realistic pressure. Pair it with our MAT study guide for a defensible plan within a week.

What is the MAT exam, exactly?

The Miller Analogies Test is published by Pearson and administered through Controlled Testing Centers (CTCs) at universities nationwide. The exam is delivered on a computer. You'll see one analogy at a time, and the on-screen timer ticks down from sixty minutes.

There are no breaks, no calculators, no scratch paper beyond what the CTC permits, and the scored portion plays out in a single uninterrupted block. You can navigate within the section before submitting, but once you submit, that's final.

Of the 120 items, only 100 are scored. The other 20 are experimental questions Pearson is field-testing for future forms. You won't know which is which. Treat every analogy as scored, because guessing strategically on a question that turns out to be experimental wastes mental energy you can't recover.

Each analogy is presented in this form: A : B :: C : D, with one of the four terms left blank. You select from four answer choices. The relationship between the two paired terms is the heart of the question — synonym, antonym, part-to-whole, cause-effect, function, classification, degree, sequence, and dozens of other patterns appear. Identify the relationship first, and the right answer becomes obvious. Skip that step and you'll waste forty seconds chasing surface similarities.

MAT scoring scale and what counts as a good score

The scaled score ranges from 200 to 600, with 400 representing the mean and a standard deviation of 25 points. That puts roughly 68% of test-takers between 375 and 425. A score of 425 lands you in the top 16% nationally. A 450 sits in the top 2%.

Most graduate programs that require the MAT publish a recommended floor — typically 400 for general admission, 410-420 for competitive programs, and 430+ for elite cohorts. Our MAT score ranges breakdown shows exactly where each band falls.

You also receive a percentile rank, calculated against a rolling norm group of recent test-takers. The percentile is what admissions committees actually compare across applicants, since the scaled score by itself doesn't communicate how rare it is. A 415 might sound modest, but at the 73rd percentile it's a strong signal of analytical capacity.

Pearson sends your official score to up to three institutions you designate on test day, free of charge. Additional score reports cost roughly $25 each. Preliminary scores appear on screen immediately after submission; official reports arrive within 10-15 business days.

Content domains covered on the MAT

The MAT tests four broad knowledge clusters: humanities, social sciences, natural sciences, and mathematics. Within those clusters, the test draws from sub-disciplines that surprise candidates who haven't reviewed the breadth.

Humanities pulls from fine art, music theory, literature (American, British, world), philosophy, religion, and mythology. Social sciences span psychology, sociology, economics, anthropology, political science, geography, and law. Natural sciences cover biology, chemistry, physics, astronomy, geology, and ecology. Mathematics includes arithmetic relationships, algebra, geometry, and basic logic — but rarely computation; the math content tests vocabulary and concept recognition, not calculation.

You're not expected to be a specialist in every domain. You're expected to recognize enough terminology that you can spot a relationship even when one term is unfamiliar. That's why vocabulary breadth matters more than depth. A candidate who recognizes the word entropy as connected to thermodynamics will solve a physics analogy faster than one who can derive the second law but freezes on the word.

Build a personal flashcard set across all four clusters. Aim for 600-800 terms by test day. Review them on rotation so weak areas surface. Our MAT practice questions set groups terms by domain so you can spot gaps quickly.

The relationship types you'll see again and again

Every MAT analogy hinges on a specific relationship between the two terms in the pair. Master the common patterns and you'll cut your average solve time in half. The most frequent relationships include: synonym (large : enormous), antonym (brave : cowardly), part-to-whole (page : book), category (oak : tree), function (scissors : cut), and cause-effect (virus : illness).

You'll also see degree (warm : hot), sequence (infant : toddler), creator-creation (Shakespeare : Hamlet), and definitional (laconic : terse). Less obvious patterns recur too: symbol relationships (dove : peace), measurement (inch : foot), grammatical (run : ran), and characteristic (desert : arid).

Some analogies use cross-cluster bridges — a biology term paired with a literary one, both linked by a Greek root. Those are harder. Slow down, identify the bridge, and you'll usually find the right choice eliminates two distractors instantly.

One trap to watch: the order of terms matters. A : B :: C : D means the relationship from A to B equals the relationship from C to D. If A is the cause of B, then C must be the cause of D — not the effect. Reversed relationships are the most common wrong answer Pearson plants. Always read the relationship in the direction the test gives you.

Time strategy: thirty seconds per analogy is realistic

Sixty minutes divided by 120 analogies gives you exactly 30 seconds per item. That's tight, but workable if you commit to a rhythm. Spend the first 10 seconds identifying the relationship. Spend the next 10 seconds testing it against the answer choices. Spend the final 10 seconds picking the best match and moving on.

If you exceed 45 seconds on a single item, mark a guess and move forward — you cannot afford to lose two future items chasing one. There is no penalty for wrong answers. Always fill in something.

Random guessing gives you a 25% chance of a point; informed guessing (eliminate even one obvious distractor) raises that to 33%. Across 120 items, smart guessing on the hardest 20 questions can add 4-6 scaled points to your final number.

Build endurance with full-length timed sessions. Doing 30 analogies in 15 minutes feels manageable. Doing 120 in 60 minutes feels brutal. The fatigue curve is real, and most candidates hit a wall around question 80. Practice the back half deliberately so you don't crash on test day.

Building a four-week MAT study plan

Four weeks is the sweet spot for most candidates. Less than that and you won't have enough reps. More than that and motivation tends to sag. Here's a battle-tested skeleton.

Week one — diagnostic and vocabulary base. Take one full-length practice test cold. Record the score, the time-per-question, and the relationship types that tripped you up. Build a vocabulary deck of 200 terms drawn from your weak domains. Drill 30 analogies per day at a slow pace, focusing on relationship identification rather than speed.

Week two — relationship patterns. Work through one relationship type per day: synonyms Monday, antonyms Tuesday, part-to-whole Wednesday, and so on. Add 40 vocabulary terms per day to your deck. Take a half-length timed test (60 questions in 30 minutes) on Friday and analyze the misses.

Week three — speed and breadth. Run two full-length timed tests this week. Push your pace toward 25 seconds per analogy on the easier half so you bank time for the harder half. Add 60 vocabulary terms per day, prioritizing humanities and social sciences if those are weaker.

Week four — taper and refine. Cut volume by half. Review your error log every day. Take one final full-length test 72 hours before test day, then rest. Sleep, hydration, and a calm morning routine matter more than last-minute cramming.

Common MAT mistakes that cost real points

Three errors crush more scores than anything else. First, candidates rush past the relationship and pick the choice that feels related to one of the given terms. That choice is almost always a distractor.

Second, candidates ignore part-of-speech. If the relationship pairs two verbs, the answer must be a verb — distractors are often the noun form of the same word. Third, candidates fall for cultural bias. The MAT pulls from a wide canon, but its references skew toward Western academic tradition.

If you've never studied Greek mythology, Renaissance art, or American literature, build at least surface familiarity using a guide like Miller Analogies Test prep. You don't need depth; you need recognition.

A subtler mistake: assuming the analogy works in only one direction. Some relationships are reversible (hot : cold :: tall : short), but many are not (cause : effect). When you build your relationship sentence, lock the direction, and check every answer choice against the locked sentence.

What test day actually looks like

Arrive at the CTC thirty minutes early with two valid IDs (one government-issued photo, one with a signature). Lockers store your belongings; you cannot bring phones, notes, watches, or any electronics into the testing room.

The proctor seats you at a workstation, walks you through the tutorial, and then the timer starts. You have 60 minutes of scored test time, plus an optional brief tutorial period that doesn't count.

The interface is utilitarian. You see one analogy, four answer choices, a navigation panel, and a countdown clock. You can move within the section before submitting, but once you submit, that's final. There's no second section, no break, and no return.

The screen will offer to display your preliminary score immediately — accept it. The number you see is your scaled score (between 200 and 600). Percentile rank arrives later in the official report.

After the MAT: sending scores and retaking

Your three free score reports can be sent on test day or designated afterward through your Pearson account. Programs receive scores typically within two weeks. If you're unhappy with the result, you can retake the MAT after a 30-day waiting period.

There's no cap on retakes, but most programs see your full history, so don't retake casually. If you're prepping for graduate school, pair your MAT prep with sample admissions essays and reach out to your target programs early. A 425 with a clean application beats a 450 with a sloppy one.

The MAT rewards consistent, structured practice more than raw intelligence. If you commit four weeks, drill relationships methodically, and build vocabulary across all four clusters, you'll walk into the testing center confident — not because you've memorized every analogy, but because you've trained your brain to spot patterns under time pressure.

Final tactical notes for high scorers

Candidates targeting 430 and above tend to share three habits. They build vocabulary deliberately rather than passively — sitting down with a deck and a timer rather than hoping to absorb terms through reading.

They drill timed practice tests at least twice a week in the month before the exam, treating each one as a dress rehearsal rather than a learning session. And they keep an error log: not just what they missed, but why they missed it.

Was it a vocabulary gap, a relationship misread, a direction flip, or a part-of-speech trap? Patterns reveal themselves quickly, and the fix is almost always specific. One last edge that costs nothing: the day before the test, do not study.

Take a walk, eat a normal dinner, sleep eight hours, and let the work you've already done consolidate. Stuffing extra vocabulary into your brain the night before crowds out retrieval pathways you actually need. Trust the prep, show up rested, and let pattern recognition do the rest.

The MAT is one of the most learnable graduate admissions tests, which is good news if you're willing to put in four focused weeks of work. There's nothing magical about a 430+ score — it comes from steady reps, an honest error log, and a finishing taper that protects your energy on test morning.

Tens of thousands of candidates clear competitive scores every year using this approach. The difference between a 405 and a 430 is rarely raw ability — it's prep structure. Pick a four-week window, block your study sessions on the calendar, and protect them like you'd protect time on the actual test day.

If you stall partway through, don't restart. Look at your error log, identify the single weakest cluster, and run a focused micro-week on just that area. A four-day deep dive into mythology or art history can lift your raw score noticeably because the MAT pulls evenly across all four content clusters — every cluster you ignore costs you about 25 questions worth of upside.

Finally, treat the MAT as a pattern-recognition trainer rather than a knowledge test. Knowing every Greek god by name won't help if you can't spot that the analogy is testing creator-creation rather than category. Spend half your prep time on relationship classification drills, and the vocabulary work will compound on top of a much stronger foundation.