How to Use Elimination on a Quantitative Reasoning Question

Claim under review: elimination can replace calculation

That claim is too broad. Elimination can reduce the work in a multiple-choice quantitative problem, but it cannot rescue a misunderstood question. The useful claim is narrower: when the wording supplies constraints, those constraints can rule out answers before you calculate an exact value. The population here is ordinary test takers, the instrument is a multiple-choice reasoning item, the outcome is performance on that item, and the time horizon is the current attempt. None of that supports a conclusion about a person's general intelligence.

Quantitative reasoning means using numbers, relationships, quantities, and operations to reach or check a conclusion. Elimination is the deliberate removal of answer choices that cannot satisfy the problem. A raw item result is simply whether you answered that item correctly. It is not a normed score, percentile, or diagnosis. A standard score compares performance with a defined reference group, and a percentile describes the percentage of that group at or below a stated result. You cannot derive either from how many options you eliminated.

The best evidence for the method is practical rather than magical. Mathematics guidance from the Education Endowment Foundation recommends making problem-solving choices visible, then planning, monitoring, and evaluating the approach. A mathematics assessment guide from The Math Learning Center likewise treats estimating, showing a workable method, and checking whether an answer is reasonable as distinct evidence of understanding. Together, these sources support a process: make a prediction, test it against the problem, and inspect the result. They do not establish that elimination raises general cognitive ability.

Why it sounds plausible, and when it works

Multiple-choice options often contain predictable mistakes. One may have the wrong sign because a subtraction was reversed. Another may be ten times too large because place value was mishandled. A third may use the wrong operation. If the question asks for a time, a choice measured in kilograms is impossible. If a quantity is divided into four equal parts, a result greater than the original may be impossible unless the wording includes a multiplier elsewhere.

Before touching the options, translate the question into a small set of tests. Ask: What is being requested? What unit should it have? Should it be positive, negative, whole, or fractional? What is a safe lower or upper bound? Are there divisibility or parity restrictions? These are not tricks. They are properties that any correct answer must have.

Use elimination most confidently when a choice violates a hard condition. Use it more cautiously when the condition is only an estimate. Rounding 48 times 19 to 50 times 20 suggests a result near 1,000. It can reject 90 and 10,000, but it may not distinguish 912 from 1,020. An estimate narrows the field; exact reasoning settles close alternatives. The mistake is treating a rough screen as proof.

Worked example: use bounds before exact arithmetic

Consider an example. A workshop packs 7 trays with 18 screws in each tray and then removes 9 screws for inspection. Which answer is possible for the number left? A. 45, B. 117, C. 135, D. 261. This is an original teaching example, not an item from a protected assessment.

Start with the structure. Seven groups of 18 make a total below 7 times 20, so the starting total must be below 140. Removing 9 makes the final count lower still. That immediately rejects D, 261. The final count must also be a whole number, so no fractional option would survive. Now estimate more closely: 7 times 18 is 126, and 126 minus 9 is 117. B is plausible. A is not impossible from the first bound, but it would require the starting total to be 54, which is inconsistent with seven trays of 18.

Only now do the exact calculation: 7 times 18 equals 126; 126 minus 9 equals 117. The answer is B. Notice what elimination did and did not do. The upper bound removed a grossly oversized answer. The group structure ruled out a much smaller answer. Exact arithmetic confirmed the remaining choice. If the choices had been 116 and 117, elimination would have reduced confidence but not finished the problem.

Write the checks in words when you review the item: seven equal groups create 126, removal decreases the total, and the result stays a whole count. That explanation is more useful than saying, ‘I picked B because it looked right.’ It records the operations that controlled the decision.

A practical elimination sequence for harder questions

Use this sequence under time pressure. First, read the question for the requested quantity, not just the numbers. Second, mark the unit and direction of change. Third, make a rough estimate by rounding one or more values. Fourth, scan the choices for violations of sign, unit, range, divisibility, or scale. Fifth, compare the survivors by calculating the smallest useful piece. Sixth, substitute the selected answer into the original wording and ask whether it satisfies every condition.

Suppose a price of 80 is reduced by 25 percent and the choices are 20, 55, 60, and 105. A reduction must produce a value below 80, so 105 is out. Twenty is a 75 percent reduction, not a 25 percent reduction. The remaining candidates are 55 and 60. One quarter of 80 is 20, so subtracting that discount gives 60. The answer is 60. You did not need a decimal formula, but you did need to understand what 25 percent means.

For ratios, preserve the direction. If 3 notebooks cost 12, then 1 notebook costs 4 and 5 notebooks cost 20. A choice of 6 may come from dividing 12 by 2, but it does not preserve the stated rate. For averages, remember that the average must lie between the smallest and largest values when all values are ordinary positive observations. For proportions, check whether the numerator is being compared with the whole or with a part. Many attractive wrong answers are valid calculations applied to the wrong relationship.

Do not eliminate a choice merely because it is inconvenient to calculate. ‘This takes longer’ is not a mathematical constraint. If two options survive, switch from screening to verification. The most reliable stopping rule is simple: stop eliminating when the remaining choices share all the quick properties, then calculate enough to distinguish them.

Contradictory evidence: a faster answer is not always better

Elimination feels like a transferable reasoning skill, but the evidence requires a narrower conclusion. The Education Endowment Foundation describes problem solving as flexible strategy selection, not a single predetermined method. A learner must choose a suitable approach for the particular problem, monitor progress, and evaluate the result. That means a memorized elimination routine can fail when the structure changes.

Research on test preparation also complicates the story. A 2007 meta-analysis of cognitive-ability retesting found that scores improved across measurement occasions, with larger effects when coaching accompanied practice or when identical forms were used. A 2025 meta-analysis of preparation for large-scale educational tests reports that preparation effects vary with the intervention, domain, and test design. These findings can be consistent with better familiarity, test-taking skill, or relevant knowledge. They do not show that a higher practiced score is a pure measure of a broad underlying ability.

This is the near-versus-far transfer distinction. Near transfer means improvement on tasks that resemble what was practiced. Far transfer means improvement on substantially different abilities or contexts. A second-order meta-analysis of cognitive training found small to medium near-transfer effects and small to null far-transfer effects overall, while noting limits on what populations and programs can be generalized. Practicing bounds, units, and proportional reasoning may make you better at related quantitative questions. It is not a responsible basis for promising a higher general-intelligence score.

The strongest exception is genuine learning of the mathematics itself. If you understand why a rate, bound, or percentage works, you may apply that knowledge across new examples that share the underlying structure. That is different from memorizing that one option is usually too large. The transferable part is the concept and the checking habit, not the surface appearance of an answer set.

What to do after the question

Review the item in three layers. First record the raw outcome: correct, incorrect, or unanswered. Second identify the decisive constraint: unit, bound, operation, ratio, or substitution. Third classify the error. Did you misread the requested quantity, choose a weak estimate, perform the arithmetic incorrectly, or stop before checking? This gives you a practice target without turning one result into a label about who you are.

If you want low-stakes practice, take the free 50-question Test IQ Free reasoning test at /test. It is a fixed-form educational quiz with original questions across pattern, quantitative, verbal, spatial, and logical reasoning. Treat the result as performance under those test conditions. Review the worked answers and look for repeated operations, such as reversing a comparison or failing to check units. The quiz is not a professional intelligence assessment, and its item score should not be converted into a clinical conclusion.

For a more structured review, the optional $9 practice report can organize explanations, error patterns, fresh practice, and a 14-day plan. Its useful promise is feedback on relevant operations, not a guarantee of a higher IQ. If your question concerns disability, gifted identification, school placement, employment, or another consequential decision, seek a qualified professional assessment. Testing standards and assessment guidance emphasize that interpretation and use must match the evidence, administration, and purpose of the assessment.

Return to the opening claim with a changed interpretation: elimination is a way to expose constraints and control arithmetic, not a shortcut around understanding. On your next quantitative question, write one estimate and one hard constraint before looking for the answer. That small action makes your reasoning inspectable and gives you something concrete to improve.

Questions readers ask