Skip to content
SPM Tuition
Mathematics chapter guide

Choosing the right summary of a data set

You can compute every measure, but you are not sure which one the question really wants.

A summary is a single number that stands for a whole data set, and each summary tells a different part of the story. This cluster teaches you to choose the one that fits the question, and to explain why.

It extends dispersion of ungrouped data in SPM Mathematics.

What do the four pages cover?

  1. Comparing two classes with equal means but different spread: why the mean alone is not enough.
  2. Explaining how an outlier changes the mean and median differently: which measure resists an extreme value.
  3. Reconstructing a missing observation from a mean and checking plausibility: using the mean backwards and testing the answer.
  4. Deciding whether two data summaries are genuinely comparable: when a comparison is unfair.

A practice set closes the cluster.

One data set, two stories

Five pupils save RM10, RM12, RM13, RM15 and RM100 in a month. The mean is 150 ÷ 5 = RM30. The median is RM13.

The mean says a typical pupil saves RM30, yet four of the five saved less than RM16. The median tells the truer story of a typical pupil, because the RM100 pulls the mean upward. The choice of summary changed the message.

Who should start where?

A student who can calculate every measure but loses marks on wording should begin at the first page. A student preparing for questions that say “explain” or “justify” can go straight to the outlier and comparability pages.

Use the descriptive statistics explorer to test your predictions on invented data. For a teacher to question your reasoning, see online one-to-one Mathematics tuition.

Common questions

When is the median better than the mean?

When the data has an extreme value that pulls the mean toward it. The median stays in the middle of the ordered list, so it describes a typical value more fairly. Pocket money, house prices and waiting times are usually better described by the median.

Is the standard deviation always the best measure of spread?

Not always. It uses every value, so an extreme value changes it a lot. The interquartile range ignores the extremes. Choose the one that fits the question and state your reason.

Where do I start in this cluster?

Start with comparing two classes with equal means but different spread. Then read the outlier page, the page on reconstructing a missing observation, and the page on whether two summaries can fairly be compared.

Does this go beyond the syllabus?

The pages stay with ungrouped data and the four measures of the chapter. They ask for reasoning about the measures, not new formulas. If you are unsure of the scope, check your school syllabus.

If you want the judgement behind the calculations, one-to-one lessons let a teacher ask you why a measure fits, and push until your reasoning is clear.

  • Online one-to-one lessons for your child with an experienced teacher.
  • Your first class is a one-hour trial, from RM50. The fee is agreed before you book.
  • Happy with the teacher? Continue with lessons of about 1.5 hours. If not, ask for another teacher.