Measures of center and spread summarize a data set, but each responds differently to its values. Understanding how the summaries are built helps you calculate missing values and decide which measure answers a question.
Mean uses every value
The mean is the sum of the observations divided by their count. If six values have mean 12, their sum is This reversible relationship is the key to missing-value problems.
A mean is a balance point: values below it and above it offset one another. It need not be an observed value or a whole number. Zeros count as observations; leaving them out changes both the sum and the count.
For a frequency table, multiply each value by its frequency to find its contribution to the sum. Then divide by the total frequency, not the number of different values.
Example: recover a missing score
Five scores have mean 14. Four are 11, 13, 16, and 18. The required total is 70, and the known scores sum to 58. The missing score is 12.
If a sixth score of 20 is added, the new mean is Do not average 14 and 20 directly: the old mean represents five scores, while 20 represents only one.
More generally, adding a value above the current mean raises the mean, adding one below it lowers the mean, and adding one equal to it leaves the mean unchanged.
Median and mode depend on order and frequency
Sort values before finding the median. For 2, 5, 5, 8, 9, 13, the middle pair is 5 and 8, so median is 6.5. The mode is 5 because it occurs most often.
The mean is These centers need not agree. A data set can have more than one mode or, under the usual school convention for distinct values, no mode.
The range is It uses only two values and therefore says little about where the remaining observations cluster.
Example: an outlier changes the mean
Compare the lists 4, 5, 6, 7, 8 and 4, 5, 6, 7, 28. The median is 6 in both lists. The means are 6 and 10, while the ranges are 4 and 24.
Changing the largest value strongly affected mean and range but left the middle position unchanged. The median is often a more representative typical value when a few extreme observations pull the mean away from most of the data.
Do not infer that the median never changes when a new extreme value is added. Adding an observation changes the count and can change which positions are middle, even if the new value itself is far away.
Combine groups using their sizes
A group of 10 students has mean score 70, and a group of 20 has mean 85. Their totals are 700 and 1,700. The combined mean is
The average of the two means, 77.5, would give both groups equal weight even though one is twice as large. Multiply by each group size, combine totals, then divide by the combined count.
Try it yourself
Four numbers have mean 9. A fifth number is added, and the new mean becomes 11. Find the added number.
Answer: The old total is 36 and the new total is 55. The added value is The mean rose because 19 exceeds the original mean 9.
What to remember
Use total equals mean times count, sort for the median, and count frequencies for the mode. Weight group means by group size and interpret changes according to how each summary is defined.