University of Birmingham · FACULTY OF STATISTICS

41297 Chap.4 Measures of Centre: Median, Mode and Means

- one subject, every graph, every model, every mark
7 Chapters4-page Bible
Our own words - no uploaded lecturer files
Updated for this semester
Chapter 4 of 7 · 41297

Measures of Centre: Median, Mode and Means

Finding the typical value

Lecture 2 begins descriptive analysis with the centre of a dataset. The median is the middle of the sorted values, the mode is the most frequent value, and the mean balances every deviation to zero.

They coincide only for tidy, symmetric data, so part of the skill is noticing when they part company.

A family of means

Beyond the plain average, the lecture names grouped-data, weighted, geometric, weighted geometric and harmonic means, each tied to a business use: survey frequencies, portfolios and module marks, compound growth, and rates such as speeds or costs per unit.

What the chapter trains

You practise choosing the measure from the variable's level and shape, computing it by hand with the slide formulas, and explaining in one sentence why another measure would mislead.

Worked items cover an outlier-hit average, a weighted portfolio return and a three-year growth rate.

In this chapter

What this chapter covers

  • 01

    Median for odd and even counts

  • 02

    Mode, multiple modes and bimodal data

  • 03

    Mean, deviations and sensitivity to extremes

  • 04

    Sample and population mean notation

  • 05

    Grouped-data and weighted means

  • 06

    Geometric mean for compound growth

  • 07

    Harmonic mean for rates

Worked example · free

Median and mean of daily orders with one outlier

Q [5 marks]. AskSia-authored practice. An online shop received 9, 10, 11, 12, 13, 14, 15 and 76 orders on eight days, the last during a flash sale. Find the median and the mean daily orders and say which better describes a typical day. The marks shown are an AskSia study allocation, not an official University of Birmingham marking scheme.
One extreme day moves the mean.One extreme day moves the mean. Each dot is one day's orders on a scale from 0 to 80; the solid line marks the mean of 20 and the dashed line the median of 12.5, so notice how far the single outlier drags the mean to the right of seven of the eight days.020406080Median 12.5Mean 20Outlier 76Daily orders
One extreme day moves the mean. Each dot is one day's orders on a scale from 0 to 80; the solid line marks the mean of 20 and the dashed line the median of 12.5, so notice how far the single outlier drags the mean to the right of seven of the eight days.
  • 1The values are already in ascending order, and n = 8 is even, so the median sits between the 4th and 5th values.
  • 1Median: (12 + 13) ÷ 2 = 12.5 orders.
  • 1Mean: the total is 160, and 160 ÷ 8 = 20 orders.
  • 1The mean exceeds seven of the eight days because the flash-sale value of 76 pulls it upward.
  • 1Report the median of 12.5 as the typical day, and mention the flash sale separately.
The median is 12.5 orders and the mean 20. Only the flash-sale day lies above the mean, so the median is the honest summary of an ordinary day; the mean describes total volume divided evenly, which no real day resembled.
Sia tip — Whenever the mean lands above most of the values, name the extreme value that put it there before choosing which centre to report.
Glossary

Key terms

Mode
The value that appears most often in a dataset, the only centre usable for nominal data.
Arithmetic mean
The total of all values divided by how many there are, written x-bar for a sample.
Grouped-data mean
An average formed by multiplying each value by its frequency, adding, and dividing by the total count.
Geometric mean
The nth root of the product of n growth factors, giving the typical compound rate of change.
Harmonic mean
The count of values divided by the sum of their reciprocals, suited to averaging rates over equal quantities.
FAQ

Measures of Centre: Median, Mode and Means FAQ

When should I report the median instead of the mean?

Use the median when the data contain extreme values or are clearly skewed, as with incomes, house prices or order sizes. Lecture 2 stresses that the median ignores extremes while the mean is sensitive to them. Many analysts report both, because a large gap between them is itself evidence of skew worth commenting on.

Can a dataset have more than one mode, or none?

Yes. If two values tie for the highest frequency the data are bimodal, and with more ties they are multimodal. If every value appears once there is no mode at all. That is why the lecture warns that the mode may describe the centre poorly, while it remains the natural choice for categories such as payment method.

Why is the geometric mean used for growth rates?

Growth compounds: each period's rise applies to an already changed base. Multiplying the growth factors and taking the nth root finds the single constant rate that reproduces the same overall change. The ordinary average of percentage changes ignores compounding and never understates typical growth, usually overstating it, sometimes by a lot when rates swing between gains and losses.

What is the difference between a weighted mean and a grouped-data mean?

Both multiply values by something before dividing. In a grouped-data mean the multiplier is a frequency, the number of times each value occurs, and you divide by the total count. In a weighted mean the multiplier is an importance weight such as money invested or a component's share of a mark, and you divide by the sum of the weights.

When does the harmonic mean apply?

Use it for rates expressed per unit, such as kilometres per hour or items per hour, when every rate applies to an equal distance or amount. A van covering two equal legs at different speeds spends longer on the slow leg, so its true average speed sits below the simple average, and the harmonic mean captures exactly that.

Study strategy

Assessment move

Make a one-page decision card from the lecture's table of means: one line per mean with its formula, a business use and a trigger phrase such as 'growth over several years' or 'equal distances at different speeds'. Then take a column from your own dataset and compute the mean, median and mode by hand on the first twenty rows before checking them in a spreadsheet.

If mean and median differ noticeably, sort the column and find the values responsible. Finish by writing one sentence that reports the centre you would choose and the reason, because that interpretation, not the arithmetic, is what the report criteria reward.

Working through Measures of Centre: Median, Mode and Means in 41297? Sia is AskSia’s AI Statistics tutor — ask any 41297 Measures of Centre: Median, Mode and Means question and get a clear, step-by-step explanation grounded in how 41297 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.

A+Everything unlocked
Unlocks this Bible + all 17 of your subjects at University of Birmingham, plus 1,000+ more Bibles across the AskSia library.
Sia - your 41297 tutor, unlimited, worked the way the exam marks it
The full 4-page Bible + practice bank with worked solutions
Chrome extension - sync your LMS so Sia knows your deadlines
Bilingual EN / Chinese on every Bible and every Sia answer
$0.99 Trial
30-day money-back · cancel in one tap · how it works