Mean, Median and Mode: Which Average Answers Which Question

Mean, median and mode are all called averages, but they do not answer the same question. A student who reaches automatically for the mean can produce a perfectly calculated number that describes the data badly.

This guide treats the three averages as different summaries of a distribution. The aim is to choose the statistic that matches the structure of the data, understand what information it preserves, and recognise when an outlier or unusual distribution makes one average more useful than another.

Mean: redistribute the total equally

The arithmetic mean is found by adding all values and dividing by the number of values. Conceptually, it answers: what value would everyone have if the total were redistributed equally?

mean = total of values ÷ number of values

For 4, 6, 7, 8 and 10, the total is 35, so the mean is 35 ÷ 5 = 7.

Median: locate the middle position

The median is the middle value after the data is ordered. It answers: what value lies at the centre of the ordered data?

For 4, 6, 7, 8 and 10, the median is 7. For an even number of observations, take the mean of the two middle values.

The median depends on position rather than the total, so extreme values affect it much less than they affect the mean.

Mode: find the most frequent value

The mode is the value that occurs most often. It answers: which value or category is most common?

For 2, 3, 3, 3, 5, 8, the mode is 3. A data set can have no mode, one mode or more than one mode.

The same data can give three different summaries

Consider the values 2, 3, 3, 4 and 18.

Mean = (2 + 3 + 3 + 4 + 18) ÷ 5 = 6
Median = 3
Mode = 3

The mean is pulled upward by 18. The median and mode stay near the cluster of smaller values. None of the statistics is automatically “wrong”; they answer different questions about the distribution.

When the mean is useful

The mean uses every numerical value, so it is useful when all observations should contribute to the summary and extreme values are meaningful rather than accidental distortions.

For example, if five machines produce 42, 45, 44, 43 and 46 components in an hour, the mean gives a useful overall production level because the values form a fairly tight cluster.

When the median is useful

The median is often useful for skewed distributions or data with extreme values. Household income is a familiar example: a small number of very high incomes can raise the mean substantially while leaving the median much closer to the experience of the middle household.

The median is not “better” in every context. It is better when the central position is more informative than equal redistribution of the total.

When the mode is useful

The mode is especially useful for categorical data where mean and median may not even make sense. If a shop records shoe sizes or most-purchased product types, the most frequent category may be the operationally useful summary.

Worked example: salaries

Five monthly salaries are $2,800, $3,000, $3,100, $3,200 and $12,000.

Mean = 24,100 ÷ 5 = 4,820
Median = 3,100

If the question is “what is the equal-share salary if the total payroll were redistributed evenly?”, the mean is appropriate. If the question is “what salary lies in the middle of these five employees?”, the median is appropriate.

Worked example: exam marks

A class has marks 58, 61, 63, 64, 65, 66, 68. The mean and median are both close to the centre because the distribution is fairly balanced. In such a set, either may provide a useful centre, though they still have different definitions.

Why “average” can be ambiguous

In everyday language, “average” often means mean. In mathematics, however, mean, median and mode are all measures of central tendency. A careful response identifies which one is being used rather than relying on the word average alone.

A selection checklist

1. Is the data numerical or categorical?
2. Are there extreme values?
3. Does the question care about the total?
4. Does the question care about the middle position?
5. Does the question care about the most common value?
6. Would one statistic hide an important feature of the distribution?

Common errors

Finding the median before sorting. The middle position only has meaning in ordered data.

Dividing the mean by the wrong count. The denominator is the number of observations, not the number of distinct values.

Assuming every data set has one mode. Some have none or several.

Using the mean for categories. A mean of colours or product names is meaningless.

Ignoring outliers. A correct mean can still be a poor description of a skewed distribution.

Practice

1. Find the mean, median and mode of 4, 5, 5, 7, 9.
2. Find the mean and median of 2, 4, 5, 6, 23.
3. A shop records favourite drink flavours. Which measure of central tendency is most directly useful for the most popular flavour?
4. Why might median salary be more informative than mean salary in a highly skewed company payroll?

Answers

1. Mean 6, median 5, mode 5. 2. Mean 8, median 5. 3. Mode. 4. A few very high salaries can pull the mean upward while the median still identifies the middle employee.

Connected routes

Continue with Range and Spread: Why Two Data Sets Can Have the Same Mean to see why centre alone is not enough. Return to the Mathematics Learning Hub for the wider Secondary Mathematics route.