Median

What Is the Median in Statistics?

The median is the middle value in a set of numbers arranged from lowest to highest. It splits the dataset into two halves: half the observations are below it and half are above it.

The median is a measure of central tendency, meaning it describes the center of data. It is also the 50th percentile, so it is useful when extreme values pull the mean, or arithmetic average, upward or downward. For example, a few expensive homes can raise an area's average sale price, while the median home price shows the middle sale price.

In everyday language, median can mean “middle.” In statistics, it refers to a position in ordered numerical or ordinal data.

How to Find the Median

Finding a median is straightforward, but sorting is essential. The original order does not matter.

1. Collect the observations you want to summarize.
2. Sort the values from smallest to largest.
3. Count the observations.
4. If the count is odd, select the middle value.
5. If the count is even, find the two middle values and calculate their arithmetic average.

For an odd-numbered dataset, consider 3, 5, 8, 11, and 20. There are five values, so 8 is the median.

For an even-numbered dataset, consider 4, 7, 9, and 16. The middle values are 7 and 9. Their average is 8, so the median is 8. The median does not need to appear in the dataset.

Median Formula for Odd and Even Numbers of Observations

These formulas identify positions after sorting, not the median value by themselves. Let n represent the number of observations.

These rules apply to individual, ungrouped observations. Frequency tables or grouped data may require another method because raw values may not be available.

Median vs. Mean vs. Mode

Median, mean, and mode describe data differently. Calling each an “average” can hide important differences.

Imagine annual incomes of $35,000, $40,000, $45,000, $50,000, and $1,000,000. The median is $45,000 and better represents the middle person's income. The mean is much higher because one income is exceptionally large. The best measure depends on whether the goal is to describe a typical person or total economic value.

Why the Median Is Useful

The median focuses on rank and position rather than the size of every difference between observations.

• It resists unusually large or small values, making it useful for skewed distributions.
• It can show a more typical household, customer, property, or patient when extreme cases exist.
• It can summarize ordinal data, such as ordered satisfaction ratings, where categories have meaningful order but uneven numeric spacing.
• It is easy to explain: half the observations are on each side of the reported value.
• It supports clear reporting in public-facing dashboard reporting, where readers need an understandable center point.

Common Uses of Median

Many fields use the median to answer, “What does the middle case look like?” State the population and time period.

• Income reporting uses median household income to describe the middle household in the income distribution.
• Housing reports use median sale price to show the midpoint of completed sales, reducing distortion from luxury properties.
• Salary reporting uses median pay when executive compensation or rare specialist salaries create extreme values.
• Healthcare services use median waiting time to show the middle patient's experience while monitoring very long waits separately.
• Schools may use median test scores when score distributions are uneven.
• Customer support teams use median first-response time, often alongside upper-percentile measures that reveal slower cases.
• Survey analysis can use a median rating for ordered scales, such as a five-point agreement question.
• Product teams can include median usage, completion time, or transaction value in an app design and development review to avoid letting a few power users define the apparent norm.

Limits and Common Median Mistakes

The median is useful, but it is not a complete description of a dataset. Poor handling can make a correct calculation misleading.

• Do not skip sorting. The middle item in collection order is not necessarily the median.
• Do not use a median for nominal categories with no natural order, such as color, country, or department name.
• Do not assume the median describes spread. Two groups can share a median but have very different variability or inequality.
• Do not treat median cost as total cost. A median does not add up across people, projects, or transactions like a mean can.
• Decide how missing values are handled before calculating. Excluding records can bias results if missingness is systematic.
• Do not compare medians from differently defined groups, periods, geographic areas, or measurement methods without clear labeling.
• Do not infer that exactly half of observations are strictly below and above when ties occur. Values can equal the median.

Interpreting Median Values with Distribution and Context

A median should rarely stand alone. A median income of $60,000 could describe a compact distribution where most households earn near that amount, or a highly unequal distribution with incomes spread widely above and below it. The same applies to home prices, delivery times, and test results.

Pair the median with the sample size, date range, population definition, and a measure of spread. Quartiles divide sorted data into four parts, and the interquartile range measures the distance between the 25th and 75th percentiles. These show the middle half of the data. When decisions affect different communities or customer groups, report relevant subgroups as well as the overall median.

Median in Grouped Data, Weighted Data, and Large Datasets

An exact median is available when every raw observation is stored and sorted. In grouped data, such as an age table with ranges of 20 to 29 and 30 to 39, the median is generally estimated from the group containing the middle observation. The estimate depends on assumptions about values within the range.

A weighted median is useful when observations have unequal importance. Survey responses may receive weights to better represent a population, or transaction data may weight records by exposure. The weighted median is the point where cumulative weight on each side reaches half the total weight. Software can calculate medians quickly, but reliable workflows still define missing-value treatment, confirm consistent units, sort values correctly, and document ties.

A Practical Checklist for Choosing the Median

Use this checklist before selecting a summary measure or presenting a result.

• Choose the median when data are skewed or include meaningful outliers that should not dominate the typical value.
• Use it for ordered ratings and rankings when order matters but equal numeric spacing is uncertain.
• State who or what is included, the location or subgroup, and the time period covered.
• Pair the median with quartiles, the interquartile range, or a clearly labeled range when spread matters.
• Show the mean alongside the median when totals, budgets, capacity, or resource use matter to the decision.
• Check whether a small sample, heavy ties, or missing records could make the median unstable or incomplete.
• Explain the practical question the figure answers, such as the middle sale price or typical response time, rather than presenting a number without context.