Calculate Central Tendency — Calculate mean, median, range, and weighted averages. - GMAT Quantitative Reasoning

SECTION 1

Historical Context & Motivation

Long before the advent of modern statistics, scholars and merchants faced a fundamental challenge: how to summarize a collection of observations with a single, representative number. The concept of the arithmetic mean traces its roots to ancient Babylonian astronomy, where observers averaged repeated measurements of celestial positions to reduce observational error. This instinct — to compress variability into a stable central value — has driven the development of central tendency as one of the most foundational ideas in quantitative analysis. For GMAT test-takers, these measures appear in data interpretation, problem-solving, and data sufficiency questions with remarkable frequency, making fluency with them essential for competitive performance.

~3000 BCE • Babylonian Averaging

Babylonian astronomers averaged repeated observations of planetary positions to produce more accurate predictions, establishing the earliest known use of the arithmetic mean.

1599 • Tycho Brahe & the Mean

Tycho Brahe systematically employed averaging techniques in astronomical measurements, formalizing the practice that would later become the arithmetic mean in statistical theory.

1774 • Laplace & Least Squares

Pierre-Simon Laplace demonstrated that the arithmetic mean minimizes the sum of squared deviations, providing a mathematical justification for what had been an intuitive practice for millennia.

1900s • Modern Descriptive Statistics

Karl Pearson and Ronald Fisher codified measures of central tendency — mean, median, and mode — as pillars of descriptive statistics, integrating them into the formal framework that underpins contemporary data analysis and standardized testing.

The persistent question driving this topic is deceptively simple: what single number best represents a data set? As we will see, the answer depends on context — the shape of the distribution, the presence of outliers, and whether different data points carry different levels of importance. The GMAT exploits this nuance regularly, testing not only your computational ability but also your judgment about which measure is most appropriate under given constraints.

SECTION 2

Core Principles & Definitions

Before diving into calculations, it is essential to establish precise definitions for the four measures this lesson addresses. Each measure captures a different aspect of a data set's "center" or spread, and the GMAT expects you to distinguish among them with precision. Understanding these definitions also clarifies when each measure is most informative — a judgment that data sufficiency questions frequently probe.

Arithmetic Mean

The sum of all values divided by the count of values. The mean is sensitive to every data point and thus pulled toward outliers. It is the most commonly tested measure on the GMAT.

Median

The middle value when data are arranged in ascending order. For an even number of observations, it is the mean of the two central values. The median is resistant to outliers.

Range

The difference between the maximum and minimum values. While technically a measure of dispersion rather than central tendency, the GMAT pairs it with mean and median questions to test holistic data understanding.

Weighted Average

A generalized mean in which each value is multiplied by a weight reflecting its relative importance before summing. The result is divided by the sum of the weights. This measure is crucial in mixture and proportion problems.

✦

KEY TAKEAWAY

Think of the mean as the balance point of a data set — like the fulcrum on a seesaw that makes both sides level. The median, by contrast, is the positional center — the seat where exactly half the riders sit on each side, regardless of how heavy they are. In research and on the GMAT, choosing the right center depends on whether extreme values should influence your summary.

SECTION 3

Visual Explanation

The following diagram illustrates a data set of seven values plotted on a number line. The positions of the mean and the median are marked, along with the range spanning from the minimum to the maximum value. Notice how the outlier on the right pulls the mean toward itself while leaving the median unchanged — a critical visual intuition for data sufficiency questions.

Seven data points are plotted on a 0–100 number line. The median (35) sits at the 4th value — the positional center. The mean (≈ 39.3) is pulled rightward by the outlier at 90. The range (80) spans the full distance from minimum to maximum.

This diagram encapsulates a principle the GMAT frequently tests: when a data set is right-skewed (an outlier or cluster of high values extends the tail to the right), the mean exceeds the median. Conversely, in a left-skewed distribution, the mean falls below the median. In a perfectly symmetric distribution, the mean and median coincide. Recognizing these relationships allows you to answer many GMAT questions without performing any calculation at all — a powerful strategic advantage under time pressure.

SECTION 4

Mathematical Framework

With the conceptual groundwork established, we now formalize each measure algebraically. On the GMAT, you will frequently need to manipulate these formulas — solving for a missing value, determining the effect of adding or removing a data point, or combining two groups. Comfort with the algebraic form of each formula is therefore not optional but essential.

ARITHMETIC MEAN

x̄ = (x₁ + x₂ + … + xₙ) / n = Σxᵢ / n

where x̄ is the mean, xᵢ are the individual values, and n is the count of values. A crucial rearrangement: Sum = Mean × Count. This identity is the single most useful tool for GMAT mean problems.

MEDIAN

Median = x₍₍ₙ₊₁₎/₂₎ if n is odd; Median = (x₍ₙ/₂₎ + x₍ₙ/₂₊₁₎) / 2 if n is even

The data must first be sorted in ascending order. For an odd count, the median is the single middle value; for an even count, it is the average of the two middle values. GMAT questions often provide unsorted data — always sort first.

RANGE

Range = x_max − x_min

The range captures the full spread of the data but is highly sensitive to outliers. On the GMAT, range questions often test whether you can determine extremes from partial information — for example, given the mean, the count, and all but one value.

WEIGHTED AVERAGE

x̄_w = (w₁x₁ + w₂x₂ + … + wₙxₙ) / (w₁ + w₂ + … + wₙ) = Σ(wᵢxᵢ) / Σwᵢ

where wᵢ are the weights (quantities, frequencies, or proportions) assigned to each value xᵢ. When all weights are equal, the weighted average reduces to the arithmetic mean. The weighted average always falls between the smallest and largest values being averaged — a property the GMAT tests in mixture problems.

💡

GMAT STRATEGY: THE SUM IDENTITY

Most GMAT mean problems are solved by converting between mean, sum, and count using the identity Sum = Mean × Count. For example, if the mean of 5 numbers is 20, their sum is 100. If a 6th number is added and the new mean is 22, the new sum is 132, so the 6th number is 32. This pattern recurs constantly on the exam.

SECTION 5

Detailed Breakdown: Weighted Averages & the Lever Principle

Weighted averages warrant deeper exploration because they underpin a wide class of GMAT problems — mixtures, combined rates, grouped data, and proportion arguments. The key insight is the lever principle: the weighted average always lies closer to the value with the greater weight, much as a seesaw tilts toward the heavier side. This geometric intuition enables rapid estimation and eliminates the need for full computation in many cases.

Two groups with different sizes and averages are represented as unequal masses on a seesaw. The weighted average (75) is the balance point, located three times closer to the larger group (n = 30) than to the smaller group (n = 10). The lever rule confirms: 30 × 5 = 10 × 15 = 150.

The lever principle yields a powerful shortcut: the ratio of distances from each group's average to the weighted average is the inverse of the ratio of the group sizes. In the diagram, Group A has 3 times the members of Group B, so the weighted average is 3 times closer to Group A's average. This inverse-ratio technique is the fastest way to estimate or compute weighted averages on the GMAT, especially in mixture problems where precise computation under time constraints can be expensive.

💡

MIXTURE PROBLEMS

When a GMAT question asks you to mix two solutions at different concentrations or combine two groups with different averages, you are performing a weighted average. The "weight" is the quantity (volume, count, or proportion) of each component. Remember: the final average must always lie between the two component values, and it will be closer to whichever component has greater weight.

SECTION 6

Worked Example

Consider a GMAT-style problem that integrates multiple central tendency concepts. The problem below requires you to compute the mean, determine the median, calculate the range, and apply weighted averaging — all within a single scenario.

Combined Central Tendency Problem

1
Step 1 — Read & Organize the Data

A marketing analyst records daily website visits over 8 days: {120, 95, 150, 105, 200, 110, 130, 90}. A 9th day's data is missing, but the analyst knows the overall 9-day mean is 125. We need to find: (a) the 9th day's value, (b) the median of all 9 days, (c) the range, and (d) the weighted average if weekday visits (7 days) averaged 115 and weekend visits (2 days) averaged something to be determined.
2
Step 2 — Find the Missing Value Using Sum = Mean × Count

The 9-day sum must equal 125 × 9 = 1,125. The sum of the known 8 days is 120 + 95 + 150 + 105 + 200 + 110 + 130 + 90 = 1,000. Therefore, the 9th day = 1,125 − 1,000 = 125.

9th day value = 125
3
Step 3 — Find the Median

Sort all 9 values in ascending order: {90, 95, 105, 110, 120, 125, 130, 150, 200}. With n = 9 (odd), the median is the (9 + 1)/2 = 5th value.

Median = 120
4
Step 4 — Find the Range

Range = Maximum − Minimum = 200 − 90 = 110.

Range = 110
5
Step 5 — Find the Weekend Average Using Weighted Average Logic

We know the overall mean is 125, with 7 weekdays averaging 115 and 2 weekend days averaging some value W. Using the weighted average formula: (7 × 115 + 2 × W) / 9 = 125. Solving: 805 + 2W = 1,125, so 2W = 320, and W = 160. Verification via the lever principle: the overall mean (125) is 10 above the weekday mean (115) and 35 below the weekend mean (160). The ratio of distances is 10 : 35 = 2 : 7, which equals the inverse of the group-size ratio (2 : 7). ✓

Weekend average = 160

SECTION 7

Strengths & Limitations of Each Measure

The GMAT occasionally presents data sufficiency questions in which you must determine whether a particular measure can be computed from given information. Understanding the strengths and limitations of each measure — and the conditions under which each is most informative — provides the conceptual foundation for these questions.

Measure	Strengths	Limitations
Mean	Uses every data point; algebraically manipulable (Sum = Mean × Count); well-understood theoretical properties.	Highly sensitive to outliers; can be misleading for skewed distributions; requires knowledge of all values to compute exactly.
Median	Robust to outliers; better represents the "typical" value in skewed data; requires only the middle value(s).	Does not incorporate the magnitude of extreme values; less algebraically tractable; harder to combine across groups.
Range	Simple to compute and interpret; gives a quick sense of spread; useful for GMAT "possible values" questions.	Depends on only two data points (max and min); extremely sensitive to outliers; conveys nothing about the distribution between extremes.
Weighted Avg	Accounts for differing group sizes or importance; essential for combining subgroup data; generalizes the arithmetic mean.	Requires knowledge of both values and weights; can be computationally heavier; misapplication when weights don't sum correctly leads to errors.

Comparative analysis of central tendency and spread measures

⚖️

WHEN TO USE WHICH MEASURE

On the GMAT, the choice of measure is often implicit in the question's structure. If a problem gives you totals and counts, it is testing the mean. If it asks about the "middle" value or provides a sorted list, it is testing the median. If it mentions combining groups with different sizes, it is testing the weighted average. Identifying the implicit measure quickly narrows your solution strategy and saves valuable time.

SECTION 8

Connections to Advanced Statistical Concepts

While the GMAT does not test advanced statistics directly, understanding how central tendency connects to broader statistical concepts deepens your intuition and prepares you for the occasional challenging question that requires synthesis. The measures covered in this lesson form the first layer of a hierarchy that extends into variance, standard deviation, and distributional reasoning — topics that sometimes appear in GMAT Quantitative Reasoning at an introductory level.

Concept in This Lesson	Advanced Extension	GMAT Relevance
Mean	Expected value in probability; the mean of a random variable E(X) = Σ xᵢP(xᵢ) is a weighted average where the weights are probabilities.	Occasionally tested in probability questions involving expected outcomes.
Median	Percentiles and quartiles; the median is the 50th percentile. Interquartile range (IQR) provides a robust alternative to range.	Percentile reasoning appears in data interpretation sets.
Range	Standard deviation (σ) measures average distance from the mean, providing a more nuanced view of spread than range alone.	Standard deviation concepts appear at the conceptual level; calculations are rare.
Weighted Average	Conditional expectations and Bayesian updating; portfolio return in finance is a weighted average of individual asset returns.	Mixture and rate-combination problems are direct applications.

How central tendency concepts extend into advanced statistics

A particularly elegant connection exists between the arithmetic mean and variance: variance is the mean of the squared deviations from the mean. Thus, the concept of the mean recursively underpins more advanced measures of spread. For GMAT purposes, the practical implication is that if you can compute means, you already possess the computational building block for understanding standard deviation — should a question require reasoning about it.

SECTION 9

Practice Problems

Question 1: CONCEPTUAL

A set of 11 positive integers has a mean of 40 and a median of 35. If the largest value is removed from the set, what can you definitively conclude about the relationship between the new mean and the new median?

Show answer

Removing the largest value from the set reduces the sum while decreasing the count to 10. The original sum is 40 × 11 = 440. The new mean depends on the removed value; since the original mean (40) exceeds the median (35), the distribution is right-skewed, implying the largest value is above the mean. Removing it will decrease the mean. However, the new median becomes the average of the 5th and 6th values (in the 10-element sorted set), which were the 5th and 6th values in the original sorted set — unchanged from their original positions. Without knowing the exact value removed or the specific data points, we can conclude that the new mean will decrease (since we removed a value above the mean), while the new median may remain 35 or change slightly depending on the 5th and 6th values. We cannot definitively state the new mean-median relationship without more data.

Question 2: BASIC CALCULATION

The mean of five numbers is 18. If the numbers 22 and 10 are added to the set, what is the new mean?

Show answer

Using the Sum = Mean × Count identity, the original sum is 18 × 5 = 90. Adding 22 and 10 increases the sum to 90 + 22 + 10 = 122 and increases the count to 7. The new mean is 122 ÷ 7 ≈ 17.43. Note that the two new numbers have a combined average of (22 + 10) / 2 = 16, which is below the original mean of 18, so the overall mean decreases — consistent with the weighted average principle.

Question 3: INTERMEDIATE

Set S = {3, 7, 10, 15, x}. If the mean of Set S equals the median of Set S, what are the possible values of x?

Show answer

The sum of the known values is 3 + 7 + 10 + 15 = 35, so the mean = (35 + x) / 5. The median depends on where x falls when the set is sorted. Case 1: If x ≤ 7, the sorted set places x among the first two positions, making the median = 7, so (35 + x) / 5 = 7, giving x = 0. Check: {0, 3, 7, 10, 15} — median = 7, mean = 7. ✓ Case 2: If 7 < x ≤ 10, the median = x, so (35 + x) / 5 = x, giving 35 + x = 5x, so x = 8.75. Check: sorted {3, 7, 8.75, 10, 15} — median = 8.75, mean = 8.75. ✓ Case 3: If x > 10, the median = 10, so (35 + x) / 5 = 10, giving x = 15. Check: {3, 7, 10, 15, 15} — median = 10, mean = 10. ✓ The possible values are x = 0, x = 8.75, or x = 15.

Question 4: APPLIED

A graduate program admits students from two pools. Pool A consists of 60 applicants with a mean GRE Quantitative score of 158, and Pool B consists of 40 applicants with a mean GRE Quantitative score of 164. If the program wants the overall mean GRE Quantitative score of admitted students to be at least 162, what is the minimum number of students they must admit from Pool B (assuming they admit all 60 from Pool A)?

Show answer

Let b = number of Pool B students admitted. The weighted average equation is (60 × 158 + b × 164) / (60 + b) ≥ 162. Solving: 9,480 + 164b ≥ 162(60 + b), which gives 9,480 + 164b ≥ 9,720 + 162b, so 2b ≥ 240, meaning b ≥ 120. However, Pool B has only 40 applicants. This means it is impossible to achieve an overall mean of 162 if all 60 Pool A students are admitted. The lever principle confirms this: for the overall mean to reach 162 (only 2 points below Pool B's mean of 164), Pool B would need to heavily outweigh Pool A — specifically, Pool B would need 3× the count of Pool A's distance ratio (4/2 = 2, inverted gives B : A = 2 : 1, so at least 120 from B). This problem illustrates how weighted average constraints can render a scenario infeasible.

Question 5: CRITICAL THINKING

A data set contains n positive integers, each at most 100. The range is 80, the median is 50, and the mean is 60. Prove that n ≥ 4, and construct an example with exactly 4 elements satisfying all three constraints.

Show answer

First, the range of 80 with values at most 100 means the minimum is at most 20 (since max − min = 80 and max ≤ 100, so min ≤ 20 and max ≥ 80). With n = 2: {a, a + 80} where a ≤ 20. The median = (a + a + 80)/2 = a + 40 and the mean = a + 40. For median = 50, a = 10, giving {10, 90}, mean = 50 ≠ 60. ✗ With n = 3: sorted {a, 50, a + 80} (median = 50, range = 80). Mean = (a + 50 + a + 80)/3 = (2a + 130)/3 = 60 ⟹ 2a = 50 ⟹ a = 25. But then max = 105 > 100. ✗ Alternatively, min need not be a: {min, 50, max} with max − min = 80, mean = 60. (min + 50 + max)/3 = 60 ⟹ min + max = 130 and max − min = 80, so max = 105 > 100. ✗ With n = 4: sorted {min, b, c, max} where max − min = 80, median = (b + c)/2 = 50, and mean = (min + b + c + max)/4 = 60. So b + c = 100 and min + max + 100 = 240, meaning min + max = 140 and max − min = 80, giving max = 110... unless we use max = 100, min = 20 (range = 80). Then 20 + b + c + 100 = 240, so b + c = 120, but median = 60 ≠ 50. Try {20, 40, 60, 100}: range = 80, median = 50, mean = 55. Adjust: {10, 40, 60, 90}: range = 80, median = 50, mean = 50. Need mean = 60: {20, 30, 70, 100}: range = 80, median = 50, mean = 55. Try {10, 30, 70, 90}: range 80, median 50, mean 50. {20, 40, 60, 100} gives mean 55. We need sum = 240 with min + max = min + (min + 80) and b + c = 100: 2×min + 80 + 100 = 240, so min = 30 and max = 110 > 100. ✗ Allow max = 100: min = 20. Then b + c = 240 − 120 = 120, median = 60 ≠ 50. So try n = 5: {20, 40, 50, 70, 100}: range = 80, median = 50, mean = 56. {20, 40, 50, 90, 100}: range = 80, median = 50, mean = 60. ✓ This shows n = 5 works. For n = 4, with max = 100 and min = 20: sum = 240 needed, b + c = 120, median = 60 ≠ 50. With n = 4 and max ≤ 100, the constraint set is infeasible, so actually n ≥ 5. A valid 5-element set is {20, 40, 50, 90, 100}: range = 80, median = 50, mean = 300/5 = 60. ✓ This problem illustrates how mean, median, and range constraints interact to limit the minimum size of a data set.

SUMMARY

Lesson Summary

Summary

Central tendency measures distill a data set into a single representative value. The arithmetic mean equals the sum of all values divided by the count, and its most powerful rearrangement — Sum = Mean × Count — is the cornerstone of most GMAT mean problems. The median is the positional center of sorted data, robust to outliers; for odd n it is the middle value, and for even n it is the average of the two middle values. The range (maximum minus minimum) measures spread rather than center, but GMAT questions frequently combine it with mean and median reasoning.

The weighted average generalizes the mean by assigning differing weights to each value, and the lever principle provides an elegant shortcut: the weighted average lies closer to the value with the greater weight, in the inverse ratio of the group sizes. In skewed distributions, the mean is pulled toward the tail while the median remains stable — a relationship the GMAT tests frequently. Mastery of these four measures, their formulas, and their interrelationships equips you to handle the full spectrum of descriptive statistics questions on the exam.