Summarising a Distribution
The text below is a transcript of the video above, edited with AI assistance for clarity and flow. Where the two differ, the video is the authoritative version.
You’ll learn to
In this video you start learning how to summarise distributions. Specifically, you will learn to:
- Interpret the mean, median, and mode as summaries of a distribution’s central tendency.
How good is 80%?
Let’s start with an example that is probably familiar to you. Imagine you are choosing a course to apply for at university, and you see a university advertising a course that has an 80% satisfaction rate according to the National Student Survey (NSS). At first glance, 80% sounds like a pretty good approval rate; it looks excellent.
Data in context
However, when you put that number in context, you start to gain more information. A number like that in isolation can be hard to evaluate. If you have not learned to read data in their context, you might look at the 80% and just think that it is an excellent approval rate. But to judge whether 80% is impressive or just typical, just mediocre, we need to know what is typical for similar courses. This is where summaries of central tendency can be really useful.
The mean
One of the indicators that summarise the central tendency is called the mean. The mean is what you might know as the average value. We get the mean by summing the values of all observations and dividing by the number of observations. For the NSS, imagine all of the courses that we have in the NSS data. We take the approval rate of each course, sum them up, and divide by the number of courses. This informs you about what is typical in a distribution of numerical values, in this case of a ratio variable.
It turns out that for course satisfaction, when we look at the National Student Survey data, the national mean is about 82%.
Interpreting the mean
When you find out that the mean is 82%, you see that an 80% satisfaction rate is actually slightly below the national average. The value itself did not change, but the way you interpret it does change when you gain that type of context: the average satisfaction rate across all courses being evaluated by students answering the National Student Survey. With this new information, 80% is not as impressive as it seems. It is just very close to the approval rate of a typical course.
The median
A different summary of central tendency is the median. It is important to know the mean and the median, the differences between them, and the implications for how we interpret the central tendency of a distribution depending on whether we are using the mean or the median.
The median is the value that divides the distribution in half. The value of the median is such that half of the observations are above it and half of the observations are below it. For course satisfaction, the national median is about 84%. It is not that different from the mean, but there are cases in which the mean and the median will be radically different, and it is important to be careful about which one to choose to get a sense of what a typical value is.
Interpreting the median
When we want a summary that is less sensitive to extreme values, we prefer to use the median. Here is an example. If we take the annual salaries of every Londoner, sum them up, and divide by the number of Londoners, we get a mean annual salary in London (2025) of around £70.3k. However, as you might imagine, there are some people who make a lot of money, and this pulls the mean upward in a way that the median is not as sensitive to. The salary that divides the upper half and the bottom half of income earners in London, the median salary, is actually £49.7k.
So if you want to know what a typical Londoner earns, it is better to use the median. The income of the person who separates the upper half from the bottom half is much lower than the mean, because there are some people whose incomes are so high that they pull the mean upward.
The mode
The mode is useful for nominal variables. Some variables do not have numbers on them. Take, for example, the party that a person voted for in the general election. You cannot calculate the mean or the median for a variable like that. So to get a measure of central tendency we usually look at the mode. The mode is the most frequently occurring value or category. It is especially useful for nominal variables, where we cannot really calculate averages.
Example: General Election vote
Which party did you vote for in the general election? This is from the British Election Study.
| 2024 General Election vote | Freq. | Percent | Cum. |
|---|---|---|---|
| Labour | 895 | 43.74 | 43.74 |
| Conservatives | 408 | 19.94 | 63.69 |
| Liberal Democrats | 224 | 10.95 | 74.63 |
| Scottish National Party (SNP) | 75 | 3.67 | 78.30 |
| Plaid Cymru | 22 | 1.08 | 79.37 |
| Green Party | 148 | 7.23 | 86.61 |
| Reform UK | 221 | 10.80 | 97.41 |
| Other (WRITE IN) | 20 | 0.98 | 98.39 |
| An independent candidate | 33 | 1.61 | 100.00 |
| Total | 2,046 | 100.00 |
By looking at this table, we can tell that the most frequent category here was Labour. The Labour Party received the highest number of votes in the 2024 general election. It is the mode of this distribution.
Recap
- The mean, the median, and the mode summarise what is typical in a distribution.
- They answer related but distinct questions about central tendency.
Why this matters
Summaries of central tendency help us interpret numbers in context rather than in isolation. Once we know what a typical value looks like and then look at a particular observation, we have a benchmark to compare it against: we are adding context to that information. They allow meaningful comparison across cases, such as courses in the NSS, voters, or income levels.