CAT-ASVAB common mistakes
CAT-ASVAB · Understand the traps and how to avoid them.
After grinding through a multi-step word problem, the candidate selects the first answer choice that matches a number produced mid-calculation, without re-reading the final sentence of the question stem (e.g., computing total cost when the stem asks for change from a $50 bill).
Under time pressure, reading becomes keyword-driven; the candidate locks onto the numbers and the operation and stops processing the stem once an arithmetic path appears, so the requested quantity is never explicitly identified.
What to do instead
Before computing, write the final sentence of the stem as a bare target ('Find: change = 50 - total'). Circle the requested quantity. After computing, do a last-line check: point at the answer choice and say out loud what it represents, confirming it matches the target rather than an intermediate value.
The candidate sinks four or more minutes into a single hard word problem, planning to flag it and come back, only to discover there is no review screen and no way to return to any earlier item.
Habit carried over from linear paper tests and other standardized exams where skipping and returning is allowed; the adaptive format is not rehearsed, so the 'flag and return' reflex fires automatically.
What to do instead
Rehearse the rule 'no item is worth more than ~90 seconds' in practice sets. Set a hard cutoff per item: if no solution path is visible after 60-75 seconds, eliminate what you can, commit to a best guess, and move on. Practise on a timed adaptive-style set so the one-way flow becomes automatic.
The candidate expands a squared binomial as a^2 + b^2, or distributes an exponent across a sum, or cancels terms across an addition bar (e.g., (x+3)/x simplified to 3).
Pattern-matching to the visually simplest rewrite; the rules for exponents and fractions were memorized as procedures rather than as operations with meaning, so 'looks cleaner' substitutes for 'is equivalent'.
What to do instead
Test every simplification with a number: substitute x=2, a=3, b=4 into the original and the rewritten form. If they disagree, the step is illegal. Drill the three specific illegal moves above until they are recognised on sight, and rewrite exponents as repeated multiplication when unsure.
The candidate selects a formula based on the shape of the figure or the sound of the words rather than the quantity asked for - using 2πr when area is wanted, using diameter where radius belongs, or applying perimeter arithmetic to an area question.
Geometry formulas are stored as a jumble of similar-looking strings (πr², 2πr, 4/3πr³) with no attached label for 'what this computes', so retrieval is driven by familiarity rather than by the target quantity.
What to do instead
Rebuild the formula sheet with a 'computes what' column beside each formula, and check units as a diagnostic: area answers are in square units, perimeter in plain units, volume in cubic units. Before plugging in, confirm whether the value given is a radius or a diameter and convert first.
Faced with an unfamiliar word, the candidate chooses the option that shares a root, prefix, or sound with the stem word, or the option that matches their personal association with the word, rather than the option that fits the word's actual meaning.
The test-writer's distractors are deliberately built from morphological and phonetic neighbours (benevolent/beneficial, venal/venial); recognition memory ('that looks related') overrides meaning-based retrieval, especially when the word is only half-known.
What to do instead
Cover the options, put the stem word in a short sentence, and write a one- or two-word synonym of your own first. Then match your synonym to the choices. Also score each option on charge (positive/negative) and intensity before choosing, and eliminate any option whose only link to the stem is spelling.
The candidate answers a comprehension question with a statement that is accurate according to the passage but does not answer what was asked - for example, picking a true supporting detail when the question asks for the main idea, or picking a true fact when the question asks for the author's attitude.
The passage primes those statements, so they feel familiar and correct; the candidate verifies 'is this in the passage?' instead of 'does this answer this question?', which is a different and harder check.
What to do instead
Before looking at the options, paraphrase the question into a target ('I need the author's opinion, not a fact'). Then classify each option as T (true per passage but off-target) or A (actually answers the question) before selecting. Any option that merely repeats passage wording should be treated as suspect, not as confirmation.
The candidate mixes up the roles of voltage, current, and resistance, and plugs the known numbers into whichever formula surfaces first - computing I = V/R when the question asked for power, or adding resistances in parallel with the series rule.
The three quantities are learned as a triangle of interchangeable letters, so the formula is retrieved by shape rather than by asking 'what am I solving for?' and 'what relationship links the knowns?'
What to do instead
Always write the unknown first, then list the knowns and their units (V, A, Ω, W), and only then choose the relation: units tell you the answer's form. For resistor networks, redraw the circuit and collapse it in stages rather than reciting rules; check that a parallel combination comes out smaller than the smallest branch resistor.
The candidate studies auto and shop content as a list of part names and definitions, then in the exam attributes a component to the wrong system - placing the alternator in the starting system, the camshaft's job on the crankshaft, or naming a tool by its handle rather than its function.
Rote name-memorisation produces shallow, unordered traces; without a system-level map, similar-sounding components collapse into each other and the retrieval cue is the word's familiarity rather than its function.
What to do instead
Rebuild knowledge as four labelled maps - engine, fuel/air, electrical, drivetrain - and place each part on its map with what it receives, what it does, and what it sends on. For tools, learn them by the operation they perform, and self-test by drawing the system from memory rather than re-reading the list.
In pulley, gear, and lever items the candidate computes a mechanical advantage using a quick visual count of rope segments or teeth without checking which segments bear the load, and ignores the direction of rotation or motion entirely when the question asks about direction.
Mechanical advantage is treated as a look-and-count exercise rather than a force balance; direction feels like a trivial detail next to the numbers, so the direction clause of the stem is dropped during calculation.
What to do instead
Use a fixed procedure: draw the load, mark every segment that supports it, count only those, then state the direction of the input versus the direction of the load before touching the numbers. For gears, trace one meshed pair at a time outward from the driver and record reverse/forward as you go.
The candidate answers general science items from everyday intuition - predicting that a heavier object falls faster, that a bullet dropped and one fired horizontally land at different times, that mass and weight are the same, or that heat and temperature are interchangeable.
Everyday experience generates plausible but wrong priors that survive because the candidate has never had to state the underlying principle explicitly; the wrong answer is the one that feels obvious rather than the one that follows from a law.
What to do instead
For each classic misconception, memorise the corrective one-liner and the reason (all objects accelerate equally in free fall because acceleration depends on force per unit mass; mass is amount of matter, weight is gravitational force). When an answer feels obvious, state the governing principle first and check that the option follows from it.