# GMAT Tables: Weighted Averages and Rate Traps | topin

On this page

## Identify the denominator before using the percentage

GMAC describes Table Analysis as sorting and analysing data to judge relevant information and conditions. The original tables here focus on a narrower issue than the [general Table Analysis guide](/gmat/data-insights/table-analysis): unequal denominators. Sorting is useful, but sorting a percentage column does not calculate a combined rate.

A rate is successes divided by opportunities, or another explicitly defined numerator and denominator. For a combined rate, add successes across the included groups and divide by the combined opportunities. The denominator is the weight attached to each group’s percentage. A small group should not automatically count as much as a large group.

Original example: one class has 9 passes among 10 candidates, another 45 among 90\. Rates are 90% and 50%. Combined passes are 54 among 100, or 54%. The simple average 70% would give the tiny class the same weight as the much larger class.

Write the included population before the formula. “All candidates in these two classes” defines the denominator. “The average class pass rate” could ask a different statistic: an unweighted average of the two class rates. The wording, not a general preference for weighting, determines the calculation.

## Original case mix: better subgroup rates, lower overall rate

__Invented service outcomes; opportunities are disjoint and exhaustive within each provider__
| Provider and task group | Successes | Opportunities | Success rate |
| ----------------------- | --------- | ------------- | ------------ |
| A, routine              | 81        | 90            | 90%          |
| A, complex              | 1         | 10            | 10%          |
| B, routine              | 19        | 20            | 95%          |
| B, complex              | 16        | 80            | 20%          |

Provider B has the higher success rate for routine tasks: 95% against 90%. It also has the higher rate for complex tasks: 20% against 10%. Yet A has 82 successes among 100 opportunities overall, or 82%, while B has 35 among 100, or 35%. Both the subgroup comparisons and the reversed overall comparison are correct.

The difference is case mix. A handles mostly routine tasks, where both providers succeed more often. B handles mostly complex tasks, where both succeed less often. “B has higher rates in both categories” therefore does not logically establish “B has a higher combined rate” when the category weights differ.

This is an arithmetic demonstration, not evidence that either provider is better in every practical sense. The table does not establish how tasks were assigned, whether quality standards match or what caused the rates. Separate the valid numerical comparison from a causal or managerial conclusion.

To compare the providers under the same hypothetical case mix, use common weights explicitly. At 50% routine and 50% complex tasks, A’s weighted rate is 0.5 × 90% + 0.5 × 10% = 50%. B’s is 0.5 × 95% + 0.5 × 20% = 57.5%. B is higher under that chosen common mix, consistent with its stronger subgroup rates.

This hypothetical standardisation is not the actual overall rate in the original table. Label it as a new comparison with imposed equal category weights. A statement asking about reported overall results still uses A’s 90/10 and B’s 20/80 opportunity mix. Substituting the adjusted comparison would answer a different question.

When reviewing, write whether you calculated an actual aggregate or an explicitly reweighted scenario. Both can be mathematically valid while supporting different claims. A neat percentage without its population and weighting definition is incomplete evidence.

## Judge each statement using its own population

Original statement one: “B has a higher success rate in each task category.” Yes, supported by 95 > 90 and 20 > 10\. Statement two: “B has more total successful tasks.” No, B has 35 while A has 82\. Statement three: “A has the higher overall success rate.” Yes, 82% exceeds 35%.

A student can get the first statement right and still answer the third incorrectly by extending the subgroup trend to the total. Use a separate numerator and denominator for each claim. Do not let a correct qualitative observation substitute for the requested aggregate calculation.

If a statement asks “all routine tasks across both providers”, combine 81 + 19 successes and 90 + 20 opportunities: 100/110 = 10/11, about 90.91%. Do not use either provider’s overall rate. The new population is a cross-provider subgroup, so both dimensions of the table matter.

Our [Multi-Source exception guide](/articles/gmat-multi-source-rules-exceptions) handles applicability across tabs. Here the analogous skill is deciding which rows belong in the requested population. Mark the row labels or jot a short inclusion rule before calculating when the table contains several overlapping categories.

## Apply weights to averages as well as percentages

Original question: a small team of four people has a mean monthly output of 30 units; a team of six has a mean of 50\. What is the combined mean? Total output is 4 × 30 + 6 × 50 = 420 units. Divide by ten people to get 42 units per person, rather than the unweighted mean 40.

The weights are people because the requested result is output per person. If the question instead asks for the mean of the two team means, 40 is the relevant unweighted statistic. Read the requested object carefully: mean employee output and mean team-average output are different calculations.

For a second example, suppose two equal-sized teams have means 30 and 50\. Now equal weighting gives 40 correctly. Equality of the means’ denominators is the condition that makes the simple average valid. The same condition explains why averaging two percentages can sometimes work without being universally safe.

Keep units through the calculation. Four people multiplied by 30 units per person gives 120 units, and dividing combined units by combined people restores units per person. Unit tracking can expose a formula that adds percentages to counts or divides by the number of rows instead of the relevant population.

## Know when the table cannot determine the combined result

Suppose a table lists pass rates of 80% and 60% for two centres but omits candidate counts. The overall rate is not determined unless additional weighting information is given. With equal groups it is 70%; with nine times as many candidates in the first centre it is 78%. Both fit the two stated rates.

If all groups have positive sizes and the rates are 60% and 80%, the combined rate lies between them. That bound can be supported even when the exact result is unknown. Do not treat missing weights as permission to invent equal weights; decide whether the question requests a value, a bound or a comparison.

A change in success rate is also different from a change in success count. A branch with 20 successes among 40 opportunities has a 50% rate. Another with 30 among 100 has more successes but a lower 30% rate. Our [percentage-change guide](/articles/gmat-percent-change-reverse-percentages) covers the separate growth calculations.

For rate comparisons with fractions, cross-multiplication can avoid awkward decimals when denominators are positive. Compare 18/40 and 27/70 through 18 × 70 = 1,260 versus 27 × 40 = 1,080\. The first rate is higher. This is an original arithmetic shortcut; choose precision according to the actual answer task.

## A table-review routine that catches denominator errors

Before answering, state the population and measure. Then identify whether sorting alone settles the question or a calculation is needed. For a maximum count, a sorted count column may be enough. For a maximum rate, you need its denominator unless the table already supplies the correctly defined rate.

In multi-part questions, verify all requested comparisons rather than trusting a first impression of the table. Treat each multi-part item as all-or-nothing, since prep providers widely report no partial credit. A correct combined calculation does not protect a response that accidentally compares raw counts where the statement asks for percentages.

After an error, save the numerator and denominator you used and the correct pair. This makes the repair specific. “Used provider totals instead of routine-task opportunities” is more useful than “bad at tables”. Change the group sizes in this article’s example and predict whether the aggregate direction changes before calculating again.

topin’s [free full-length GMAT mock](/gmat/practice-test), marked on the official scale, can test that denominator habit among unfamiliar DI tasks. Use the original reversal example to understand the principle, not as a score forecast. Readiness means selecting the right population and weights without needing to recognise a memorised table.

## Try a full GMAT mock free

Timed like test day, every section scored, every answer explained.

[Take the free mock ](/gmat/login)

## FAQs

Can I average two percentages in a GMAT table?

Only when the requested statistic and weights justify it. For a combined success rate, add successes and opportunities first.

Can a higher rate in every subgroup produce a lower overall rate?

Yes, when subgroup weights differ. The original provider table demonstrates the reversal.

Does sorting solve weighted-average questions?

Sorting may identify relevant rows, but it does not replace the numerator-and-denominator calculation.

What if the table omits group sizes?

An exact combined rate may be undetermined. You can sometimes infer a bound, but should not assume equal weights.

Is a higher success count the same as a higher success rate?

No. The number of opportunities matters. Compare the measure the statement actually asks about.

## Sources (checked 5 October 2026)

* [mba.com: official Table Analysis purpose (checked 5 October 2026)](https://www.mba.com/exams/gmat-exam/about/exam-content)

## Related articles

[All GMAT study guides ](/gmat)

* [GMATStudy plans & strategyReview and Edit decisionsUse GMAT Review and Edit with a clear bookmark system, worked correction examples and a practical review budget that protects first-pass completion.Updated 5 Oct 2026](/articles/gmat-review-edit-strategy)
* [GMATStudy plans & strategyChoose section order and breakCompare all six GMAT section orders using fatigue, warm-up and mock evidence, then choose where the optional break fits your actual testing routine.Updated 5 Oct 2026](/articles/gmat-section-order-break-strategy)
* [GMATStudy plans & strategyReverse percentage problemsSolve GMAT reverse percentage and successive-change problems with original worked examples, multiplier equations and checks for discounts, margins and bases.Updated 5 Oct 2026](/articles/gmat-percent-change-reverse-percentages)
* [GMATStudy plans & strategyRates, work and unitsSolve GMAT rates and work problems by tracking units, combining constant rates and handling changing teams, with original questions and checked solutions.Updated 5 Oct 2026](/articles/gmat-rates-work-units)