Why Three Different Averages Exist
Statistics has three different measures of central tendency because no single definition of average works best for all data. The mean (arithmetic average) is mathematically elegant and widely used, but sensitive to outliers. The median (middle value) is robust against extremes but harder to use in further calculations. The mode (most common value) is the only measure that works meaningfully for categorical data and discrete count data.
The distinction matters enormously in practice. When the U.S. Census Bureau reports average household income, they typically report the mean — which is significantly higher than median income because a small number of very high earners pull the mean upward. Median income is a better representation of what a typical household earns. This is why real estate listings use median home price, and why income inequality discussions emphasize the mean-to-median gap as a measure of skew.
The relationship between mean and median also reveals the shape of your data. If mean is greater than median, data is right-skewed — a tail of unusually high values is pulling the mean up. If mean is less than median, data is left-skewed. If mean and median are approximately equal, data is roughly symmetric. This diagnostic is one of the fastest ways to understand a dataset's distribution.
Calculate mean, median, mode
Enter any list of numbers to get all three measures of central tendency plus range, standard deviation, and a full statistical summary.
Calculate My StatisticsHow to Choose and Calculate the Right Average
- 1
Calculate all three measures first, then decide which to report
Before choosing which to report, compute mean, median, and mode. The differences between them reveal your data's shape: symmetric (mean equals median), right-skewed (mean greater than median, driven by high outliers), or left-skewed (mean less than median, driven by low outliers). This diagnostic informs which measure best represents a typical value.
- 2
Identify outliers and determine whether they are real data
An outlier is a value unusually far from the rest. A single extreme outlier dramatically shifts the mean but has no effect on the median. If the outlier represents a genuine but unusual observation (not a data entry error), the median better represents the typical value — but both the mean and the outlier should be reported and explained.
- 3
Match your measure to your data type and distribution
Use the mean for numerical data that is approximately symmetric without major outliers — heights, test scores, manufacturing measurements. Use the median for numerical data with skew or outliers — income, house prices, response times, biological measurements. Use the mode for categorical data (most popular product, most common response) or discrete counts where repetition is meaningful.
- 4
Report standard deviation alongside the mean
A mean without standard deviation is incomplete. Two datasets with the same mean can be completely different in character: one tightly clustered (low standard deviation) and one highly spread (high standard deviation). Always pair the mean with standard deviation to give a complete description of both central tendency and variability.
- 5
Be skeptical of average in headlines and marketing
When you see statistics reported as averages, check whether mean or median is being used. For income, wealth, home prices, and anything with inequality: median is more representative of a typical person. For normally distributed phenomena like height or standardized test scores: mean is usually appropriate. The choice of which average to report is itself an analytical decision that shapes perception.
How Each Measure Is Calculated
The arithmetic mean is the sum of all values divided by the count of values. For the dataset 3, 7, 7, 8, 15: mean = (3+7+7+8+15) / 5 = 40 / 5 = 8.0. The mean is pulled upward by the outlier 15 and does not correspond to any actual data point.
The median is the middle value when data is sorted in order. For 3, 7, 7, 8, 15 (already sorted, 5 values): the median is the 3rd value = 7. For an even count, the median is the average of the two middle values. The median is more representative of where most values cluster in this example.
The mode is the most frequently occurring value. For 3, 7, 7, 8, 15: the mode is 7 (appears twice; all other values appear once). A dataset can have no mode (all values unique), one mode (unimodal), two modes (bimodal), or more (multimodal). The mode is the only average that can be applied to non-numerical categorical data.
Frequently Asked Questions
Can a dataset have no mode?
+
Yes. If every value appears exactly once, there is no mode. Mode is most useful when values genuinely repeat — it is less informative for continuous numerical measurements where exact repetition is rare. For categorical data (colors, categories, survey responses), mode is often the primary or only appropriate measure.
What is the median of an even number of values?
+
When n is even, sort the values and average the two middle ones. For the dataset 2, 5, 8, 11 (4 values): the two middle values are 5 and 8, so median = (5+8)/2 = 6.5. The result may not correspond to any actual data point, which is acceptable — the median is a positional measure.
What is the geometric mean and when is it used?
+
The geometric mean multiplies all values together and takes the nth root — appropriate for multiplicative processes like compound investment returns. The arithmetic mean of returns can mislead: a stock that rises 50% then falls 50% has an arithmetic mean return of 0%, but you have actually lost 25% of your money. The geometric mean correctly gives approximately -13.4% return, reflecting the true compounded result.
What is the weighted mean and when does it apply?
+
A weighted mean gives different values different importance based on their assigned weights. GPA is a weighted mean where credit hours are weights. Course final grades are weighted means where assignment category weights determine importance. The formula: weighted mean = sum(value times weight) divided by sum(weights). This differs from the arithmetic mean when values have unequal importance.
What is standard deviation and why does it matter?
+
Standard deviation measures how spread out values are around the mean. A low standard deviation means values cluster tightly near the mean. A high standard deviation means they are widely dispersed. Two classes with the same mean exam score (say, 75%) can be very different: one with standard deviation of 5 (most scores between 70-80) versus one with standard deviation of 20 (scores ranging from 35 to 100+). Standard deviation is essential context for interpreting any mean.
Why does national average income appear higher than what most people earn?
+
National mean household income in the United States is approximately $100,000-$115,000, while median household income is approximately $74,000-$80,000. The gap is caused by the highly skewed income distribution — a relatively small number of very high earners dramatically raise the mean without affecting the median. For any positively skewed distribution (income, wealth, home prices), median is the appropriate measure of a typical value.
Calculate your data's descriptive statistics
Enter any list of numbers for instant mean, median, mode, standard deviation, range, and full statistical summary.
Calculate Mean, Median, Mode