What is the median in simple terms?

The median is the middle value in a sorted list of numbers, so half the values fall above it and half fall below it.

What is the average in simple terms?

The average, or mean, is the sum of all numbers in a set divided by the total count of those numbers.

What is the main difference between median and average?

The median identifies the central position in sorted data, while the average calculates the arithmetic center by dividing the total sum by the count.

Which is better for skewed data, median or average?

The median is better for skewed data because extreme outliers pull the average away from the typical value, making it misleading.

How does an outlier affect the median versus the average?

An outlier shifts the average significantly because it changes the sum, but it barely moves the median since it only alters the middle position.

Can median and average ever be equal?

Yes, median and average are equal when the data is perfectly symmetrical, such as in a normal distribution or a balanced set like 2, 4, 6.

Is the average always higher than the median?

No, the average is higher than the median only when positive outliers exist, and it is lower when negative outliers dominate the dataset.

Why do home prices use the median instead of the average?

Real estate uses the median because a few luxury mansions inflate the average, while the median better represents what a typical buyer actually pays.

What is the most common mistake when using average?

The most common mistake is using the average on skewed data, which misrepresents the typical value because a single extreme number distorts the result.

Can I switch from using the average to the median?

Yes, you can switch to the median when your data has outliers, but you must report it clearly because each measure answers a different statistical question.

Difference Between

Difference Between Median and Average

Written byNex Virox Team

Editorial Team

Reviewed byVarshal Nirbhavane

Senior SEO & Organic Growth Professional · 5+ years

Published: August 26, 202620 min read

Quick answer

The main difference between Median and Average is that the median resists extreme outliers, while the average is pulled by them. Median is the middle value in an ordered dataset, while Average is the sum of all values divided by the count.

Key takeaways

Core distinction: Median is the middle value in an ordered dataset, while average is the arithmetic sum divided by count.
How each works: Average factors every data point equally, but median ignores outliers by focusing solely on the central position.
Outlier sensitivity: Average skews dramatically with extreme values, whereas median stays stable and resistant to those same extremes.
Best-fit use case: Choose median for skewed income or housing data, but use average for symmetric, evenly distributed measurements.
Common decision mistake: Reporting average alone misleads when outliers exist, so always pair it with median for honest insight.

Table of Contents18 sections

Difference Between Median and Average: Comparison Table

Aspect	Median	Average
Definition	Middle value in an ordered dataset, splitting the data into two equal halves.	Sum of all values divided by the total count of values in the dataset.
Purpose	Identifies the central point of a distribution, resistant to extreme outlier influence.	Provides an overall measure of central tendency representing the typical magnitude.
Core Mechanism	Sorts all numbers ascending, then selects the single middle observation or averages two middle ones.	Adds every observation together, then divides that total by the number of observations.
Formula	For odd n, the (n+1)/2th sorted value; for even n, mean of two middle sorted values.	Calculated as the arithmetic mean: Σx divided by n, where n is sample size.
Outlier Sensitivity	Unaffected by extreme values since only the middle position determines its magnitude.	Shifts dramatically with a single extreme value because every data point carries equal weight.
Data Skewness	Remains near the bulk of data when distribution is heavily skewed right or left.	Gets pulled toward the tail in skewed distributions, often exceeding the typical observation.
Computation Steps	Requires sorting the entire dataset first, then identifying the positional middle value.	Requires only summation and division, with no sorting or ordering of observations needed.
Computational Speed	Sorting large datasets takes O(n log n) time, slower than a single-pass summation.	Computes in O(n) time with one pass, making it faster for very large datasets.
Data Type Suitability	Works with ordinal data and continuous data where ranking is meaningful and possible.	Requires interval or ratio data where addition and division are mathematically valid operations.
Uniqueness	Always exists for any dataset; with even counts, any value between two middles is valid.	Always produces a single unique value for any given set of numeric observations.
Integer Data	Can produce a non-integer result when averaging two middle values in an even-sized dataset.	Typically yields a fractional value even when all original observations are whole integers.
Income Data	Better represents typical household earnings because top earners do not inflate the central figure.	Overstates typical income when a small number of very high earners exist in the population.
House Prices	Reflects the price of a typical home, unaffected by a few luxury mansion sales in an area.	Gets inflated by a handful of ultra-premium sales, misrepresenting the affordable housing stock.
Test Scores	Shows the score of the middle student, useful when a few students score extremely low or high.	Summarises overall class performance but hides the spread and distribution of individual results.
Salary Negotiation	Used to quote typical market pay for a role, resisting distortion from executive compensation packages.	Often quoted by employers but can appear artificially high when senior salaries dominate the figure.
Real Estate	Preferred for reporting neighbourhood property values because it ignores extreme waterfront and fixer-upper sales.	Used for portfolio valuation but misleads when comparing areas with mixed luxury and distressed properties.
Skewed Distributions	Recommended when data shows strong positive or negative skew, such as wealth or insurance claims.	Appropriate only when data is roughly symmetric and free from significant outliers or heavy tails.
Symmetric Data	Equals the average when distribution is perfectly symmetric, such as a normal bell curve.	Matches the median exactly in symmetric distributions, making either measure equally valid.
Missing Values	Unaffected by missing extreme values; only the middle position determines the reported figure.	Becomes biased when missing values occur at the high or low end of the distribution.
Sample Size	Stable with small samples since one middle value resists fluctuation from a single unusual observation.	Highly variable with small samples; one extreme observation can swing the average substantially.
Interpretation	Answers what the middle observation is worth, giving the experience of the typical case.	Answers what each observation would be worth if total value were distributed equally.
Everyday Example	Household income of the middle family in a city of 100,000 families, ranked by earnings.	Total city income divided by 100,000 families, including billionaires in the calculation.
Common Users	Economists and policymakers use it for wage, poverty and cost-of-living statistics.	Engineers and scientists use it for repeated measurements, quality control and experimental data.
Reporting Standard	Government agencies report median household income and median home price as standard practice.	Media outlets report average temperatures, rainfall and batting averages for familiar comparisons.
Statistical Robustness	Resistant statistic with a breakdown point of 50%, meaning half the data can be corrupted before it moves.	Non-resistant statistic with a breakdown point of 0%, as a single infinite value destroys it.
Mathematical Use	Cannot be used in further algebraic operations like weighted averages or regression equations.	Feeds directly into variance, standard deviation, correlation and regression calculations.
Data Distribution	Preferred when distribution is unknown or clearly non-normal, such as exponential or power-law data.	Optimal when data follows a normal distribution where mean equals the most probable value.
Limitation	Ignores the magnitude of all values except the middle one, discarding useful information.	Hides the spread of data; very different datasets can produce identical average values.
Best-Fit Scenario	Best for skewed data like income, home prices and response times where outliers distort the centre.	Best for symmetric data like repeated measurements, grades and physical constants without outliers.

What Is Median?

Median is the middle value in a sorted data set. It splits a list into two equal halves, showing the central point. It exists to resist distortion from extreme outliers, giving a truer picture of a typical value than the average does.

Definition of Median

Median is the numerical value separating the higher half of a data sample from the lower half. To find it, arrange all observations in ascending order and select the middle number. With an even count, the median is the mean of the two central numbers.

Key Characteristics of Median

Characteristic	What It Means in Practice
Order-dependent	Data must be sorted from lowest to highest before the middle value can be identified.
Outlier resistant	One extreme value, like a billionaire in a town, barely shifts the median result.
Positional measure	It relies on rank position, not on the arithmetic size of every data point.
Even-count rule	With an even number of values, the median is the midpoint of the two central entries.
Skewed data safe	It stays stable when income or house prices are heavily skewed upward.
Not algebraic	Medians of separate groups cannot be combined to find the overall median.
Scale independent	Multiplying every value by a constant multiplies the median by that same constant.
Always a data point	For odd-sized sets, the median is always one of the original observed values.
Ordinal applicable	It works on ranked data like survey responses where arithmetic means are meaningless.
Distribution descriptor	It marks the 50th percentile, indicating where half the data falls below.

Common Examples of Median

US Household Income – the Census Bureau reports this figure because a few ultra-rich households would inflate the average.
House Prices – real estate agencies quote median prices since mansions skew the average upward in hot markets.
Test Scores – a class median shows the middle student's performance, unaffected by one perfect or failing score.
Player Salaries – sports unions use median pay because superstar contracts distort the average team salary.
Age of Population – demographers cite median age to show the midpoint citizen, not the arithmetic mean age.
Commute Time – transport surveys use median travel duration because a few two-hour commuters skew the mean.
Rental Costs – city planners track median rent to reflect what a typical tenant actually pays per month.
Life Expectancy – actuaries examine median lifespan to show the age by which half a cohort has died.
Apartment Square Footage – developers quote median unit size because penthouses distort the average floor area.
Small Business Revenue – industry reports use median earnings because a few giant firms inflate the average revenue.

Advantages and Limitations of Median

Advantages	Limitations
Unaffected by extreme outliers, so it represents typical cases in skewed distributions honestly.	Ignores the actual magnitude of every value, discarding useful information about spread and totals.
Works well with ordinal data like rankings where arithmetic averaging is mathematically invalid.	Harder to use in further statistical calculations because it lacks the algebraic properties of the mean.
Simple to explain to non-experts, making it accessible for news reports and public policy.	Requires a full sorted data set, so it is inefficient for streaming or very large real-time datasets.
Stable across repeated sampling, giving consistent estimates even when data is not normally distributed.	Less sensitive to genuine changes in the data, so it can hide real shifts in the bulk of values.
Meaningful for skewed data like income, where the average misleads about the typical person.	Does not reflect the total sum of values, making it useless for budgeting or resource allocation decisions.
Unaffected by measurement errors in the extreme tails of the distribution.	With small samples, the median can jump erratically between very different values.
Clearly identifies the central point, which is intuitive for non-mathematical audiences.	Provides no information about the other values, so two datasets with the same median can look very different.
Robust for skewed distributions, making it the preferred measure for real estate and salary data.	Cannot be used to compute a combined median from subgroup medians without the original raw data.
Resistant to data-entry errors, as a single typo in an extreme value rarely moves it.	Assumes data can be ordered, so it is meaningless for nominal categories like colours or brands.
Always represents an actual observed value in odd-sized sets, making it concrete and verifiable.	Fails to capture the full picture, so it must be paired with range or quartiles for complete analysis.

What Is Average?

Average is a single number that summarizes a set of values by dividing their total by the count. It gives a quick central reference point for comparing groups or spotting trends in everyday data.

Definition of Average

The arithmetic mean is the sum of all observations in a dataset divided by the total number of observations. It represents the central value when every data point contributes equally to the final result.

Key Characteristics of Average

Characteristic	What It Means in Practice
Sum-based calculation	Add every value first, then divide by the count of values to get one result.
Uses all data points	Every single observation influences the final number, including extreme outliers.
Sensitive to outliers	One very high or very low value can pull the average dramatically away from the typical case.
Single summary value	Condenses an entire dataset into one number that is easy to quote and compare.
Requires numeric data	Works only with quantities that can be added and divided, not categorical labels.
Assumes equal weighting	Treats each observation equally, regardless of its frequency or importance in the dataset.
Always between extremes	The result can never fall below the smallest value or above the largest value in the set.
Not always a real value	The result can be a fraction or decimal that does not exist as an actual observation.
Easy to compute	Requires only basic addition and division, making it accessible without special tools.
Affected by sample size	Small samples produce averages that vary more widely than averages from large samples.

Common Examples of Average

Class test scores – a teacher sums all student marks and divides by student count to report class performance.
Batting average – a cricketer's runs scored divided by innings played measures consistent scoring ability.
Household income – total earnings across a group divided by households shows general economic standing.
Daily temperature – high and low readings combined into a mean give a simple daily climate summary.
Fuel consumption – total miles driven divided by gallons used reveals a vehicle's typical efficiency.
Employee salary – total payroll divided by staff count gives a baseline for compensation discussions.
Product ratings – sum of all customer star scores divided by review count indicates overall satisfaction.
Travel speed – total distance divided by total travel time yields an effective average pace.
Monthly rainfall – total precipitation over a season divided by months shows typical wetness.
Grocery spending – total monthly food costs divided by four weeks provides a budgeting reference.

Advantages and Limitations of Average

Advantages	Limitations
Provides a single, intuitive number that anyone can understand and communicate quickly.	Hides the actual distribution of values, so very different datasets can share the same average.
Uses every data point, ensuring no individual observation is ignored in the calculation.	Extreme outliers distort the result, making it unrepresentative of the typical value.
Enables direct comparison between different groups, such as teams, regions, or time periods.	Misleading when data is skewed, such as income where a few huge salaries inflate the mean.
Simple arithmetic makes it fast to compute manually or with basic spreadsheet formulas.	Cannot be used on ordinal or nominal data like rankings, colors, or satisfaction labels.
Forms the foundation for advanced statistics like variance, standard deviation, and regression analysis.	Produces fractional results that have no real-world meaning, like 2.3 children per family.
Stable with large sample sizes, giving reliable estimates when data volume is high.	Small samples yield averages that swing wildly and may not reflect the broader population.
Works well for symmetric distributions where values cluster evenly around the center.	Fails to represent multimodal data where multiple distinct clusters exist within one dataset.
Easy to explain to non-experts, making it a default choice for news and business reporting.	Assumes equal importance of every value, ignoring frequency or relevance of specific observations.
Allows algebraic manipulation, such as combining averages from different subgroups mathematically.	Cannot be calculated when any single value is missing or unknown in the dataset.
Useful for detecting changes over time when tracking metrics like sales or production output.	Provides no information about variability, spread, or how consistent the underlying values are.

Similarities Between Median and Average

Shared Aspect	How Median and Average Are Alike
Central tendency	Median and average both summarize a dataset with a single central value representing its typical magnitude.
Single output	Median and average each produce one numeric result that describes an entire group of data points.
Input data	Median and average both require a set of numerical values as their raw input for calculation.
Statistical tools	Median and average both belong to descriptive statistics used for summarizing quantitative information.
Scale types	Median and average both apply to interval and ratio data where numeric differences have meaning.
Purpose goal	Median and average both aim to locate the center of a distribution for comparison purposes.
Data sorting	Median and average both benefit from ordered data, though the median requires full sorting first.
Sample usage	Median and average both work on samples drawn from larger populations for inferential analysis.
Population usage	Median and average both can describe complete populations when all values are available.
Spread context	Median and average both are interpreted alongside measures of dispersion like range or deviation.
User audience	Median and average both serve analysts, researchers, economists, and business decision-makers daily.
Reporting norm	Median and average both appear in standard statistical summaries like reports and dashboards.
Data cleaning	Median and average both require removal of errors or missing values before calculation begins.
Numeric result	Median and average both yield a finite number that can be rounded for presentation.
Unit preservation	Median and average both retain the original unit of measurement, like dollars or kilograms.
Comparability	Median and average both allow direct comparison across different groups or time periods.
Software support	Median and average both are built into Excel, Python, R, and SQL as standard functions.
Teaching basics	Median and average both are foundational concepts taught in introductory statistics courses.
Decision inputs	Median and average both inform policy, pricing, salary, and performance decisions in practice.
Data reduction	Median and average both compress many values into one representative figure for clarity.
Robustness limit	Median and average both lose accuracy when data quality is poor or sample size is tiny.
Interpretation need	Median and average both require context about the dataset to be meaningfully interpreted.
No zero issue	Median and average both handle zero values without mathematical errors in their formulas.
Negative values	Median and average both can process negative numbers correctly within their calculations.
Grouped data	Median and average both can be approximated from frequency tables when raw data is absent.
Summary role	Median and average both serve as headline figures in executive summaries and news reports.
Trend tracking	Median and average both are tracked over time to reveal shifts in central behavior.
Error sensitivity	Median and average both produce misleading results if input values are recorded incorrectly.
Presentation format	Median and average both are typically rounded to a sensible precision for readability.
Long-term use	Median and average both remain stable metrics across decades of statistical practice and research.

Median or Average: Which Should You Choose?

The single deciding variable is outlier presence. If extreme high or low values skew your data, choose the median. If your data is evenly distributed without extremes, choose the average. The median resists distortion; the average reflects every single value, including the extremes.

When to Use Median

Choose Median when outliers would mislead you. Use it for household income, real estate prices, or rent costs, where a few millionaires or mansions inflate the average. Also choose it for skewed datasets like test scores with extreme highs or lows. The median shows the true middle experience.

When to Use Average

Choose Average when every value matters equally. Use it for manufacturing defect rates, daily temperature readings, or delivery times, where values cluster tightly. Also choose it for calculating total consumption or budget forecasting, because the average preserves the exact sum for planning. It suits symmetrical, normal distributions.

Common Misconceptions About Median and Average

Common Myth	The Reality
The average and the median are always close in value.	The average and the median diverge sharply in skewed data, where extreme outliers pull the average away from the median.
The median is simply the average of the two middle numbers.	The median is the middle value in an ordered set; averaging two middle numbers only applies to even-sized datasets.
Outliers affect the median just as much as the average.	The median ignores outlier magnitudes entirely, while the average incorporates every extreme value directly into its calculation.
You can calculate the median without sorting your data first.	The median requires sorting all values from lowest to highest before identifying the central position in the dataset.
The average is always the best measure of a typical value.	The median better represents typical values in income, housing, and other skewed distributions where the average overstates centrality.
The median and average produce identical results in every dataset.	The median and average only match in perfectly symmetric distributions; they differ whenever data skews left or right.
A single extreme value can shift the median dramatically.	A single extreme value barely moves the median, which only changes when the outlier crosses the central position.
The average tells you what the middle person experiences.	The average describes the total divided by count, not the middle experience; the median captures the middle person's value directly.
Median and average are interchangeable terms for the same statistic.	The median is a positional measure of center, while the average is a computational measure that sums all values first.
Averages are always whole numbers when data contains whole numbers.	The average frequently produces decimal values even with whole-number inputs, whereas the median always returns an actual data point.
The median is harder to understand than the average for beginners.	The median is conceptually simpler because it identifies the literal middle value, while the average requires understanding division and summation.
Adding a zero to your dataset barely changes the median.	Adding a zero can shift the median significantly if it falls below the central position, though it always drags the average downward.
The average is more accurate than the median in all cases.	The median is more accurate for skewed data because the average gets distorted by outliers that do not represent the typical case.
Median household income and average household income measure the same thing.	Median household income reflects the typical household, while average household income is inflated by a small number of very wealthy households.
You need all data points to find the median of a set.	You only need the middle value or two middle values for the median, while the average requires summing every single data point.
The median can be negative even when most values are positive.	The median stays positive if more than half of values are positive, while the average can turn negative from a few large negative outliers.
Averages are always representative of the sample population.	The average misrepresents populations with skewed distributions, where the median provides a more honest picture of the typical member.
The median is affected by every value in the dataset equally.	The median only depends on the central value, so changing any non-central value leaves the median completely unchanged.
Average salary and median salary give identical insights about pay.	Average salary is skewed upward by executives, while median salary shows what a typical worker actually earns in the same company.
The median requires more complex math than the average.	The median only requires ordering and finding a middle position, while the average demands addition and division across all values.
If the average is high, the median must also be high.	The average can be high while the median stays low when a few extreme high values inflate the average without affecting the middle.
Median and average always have the same units as the original data.	Both the median and average share the original data's units, but the median always equals an actual observed value while the average may not.
You should always report the average and never the median in statistics.	Reporting the median alongside the average reveals skewness; relying solely on the average hides important distribution information from readers.
The median is the same as the mode in most real datasets.	The median is the middle value, while the mode is the most frequent value; these two measures rarely coincide in continuous data distributions.
Doubling every value doubles the median but not the average.	Doubling every value doubles both the median and the average, since both measures scale proportionally with linear transformations of data.
The average is always a value that exists in your dataset.	The average often produces a number that appears nowhere in the dataset, while the median always corresponds to an actual observed value.
Median is only useful for large datasets, not small ones.	The median works effectively on small datasets too, providing a robust center even with just three to five ordered observations.
The average handles missing data better than the median does.	Both the median and average require complete data, but the median is more robust when missing values occur at the extremes of the distribution.
Median and average give the same result when data is symmetric.	In perfectly symmetric distributions the median and average coincide, but real-world symmetric data still shows tiny differences due to rounding.
You cannot compare medians across different groups or populations.	You can compare medians across groups directly, and doing so often reveals differences that average comparisons obscure due to outlier effects.

Conclusion

Difference Between Median and Average comes down to outlier resistance. The median ignores extreme values, showing the true middle; the average sums everything, pulling toward extremes. Pick the median for skewed data like incomes or home prices. Pick the average for evenly distributed, symmetric data.

FAQs on Difference Between Median and Average

What is the median in simple terms?: The median is the middle value in a sorted list of numbers, so half the values fall above it and half fall below it.
What is the average in simple terms?: The average, or mean, is the sum of all numbers in a set divided by the total count of those numbers.
What is the main difference between median and average?: The median identifies the central position in sorted data, while the average calculates the arithmetic center by dividing the total sum by the count.
Which is better for skewed data, median or average?: The median is better for skewed data because extreme outliers pull the average away from the typical value, making it misleading.
How does an outlier affect the median versus the average?: An outlier shifts the average significantly because it changes the sum, but it barely moves the median since it only alters the middle position.
Can median and average ever be equal?: Yes, median and average are equal when the data is perfectly symmetrical, such as in a normal distribution or a balanced set like 2, 4, 6.
Is the average always higher than the median?: No, the average is higher than the median only when positive outliers exist, and it is lower when negative outliers dominate the dataset.
Why do home prices use the median instead of the average?: Real estate uses the median because a few luxury mansions inflate the average, while the median better represents what a typical buyer actually pays.
What is the most common mistake when using average?: The most common mistake is using the average on skewed data, which misrepresents the typical value because a single extreme number distorts the result.
Can I switch from using the average to the median?: Yes, you can switch to the median when your data has outliers, but you must report it clearly because each measure answers a different statistical question.