Introduction: Why Isn't One Average Enough?
Imagine you are looking for a part-time job. One shop says their "average" pay is \(£15\) an hour, but you find out most workers only get \(£9\). How can they both be right? In Statistics, there are different ways to find the "center" of data. Choosing the wrong one can give a misleading picture, while choosing the right one helps you make smart decisions. In this chapter, we will learn how to pick the best average for any situation and how to use them to compare different groups of data.
Note: If you need a refresher on how to calculate these values, please refer to the chapter "Mode, median and mean for discrete and grouped data".
1. The "Big Three": Pros, Cons, and When to Choose
In your exam, you will often be asked to justify why you chose a specific average. Use the following guide to help you decide:
The Mode (or Modal Class)
The mode is the most common value. When to use it: - Use it for qualitative (categorical) data (e.g., finding the most popular car color). - Use it when you need to know the "most typical" result, like the most common shoe size in a shop. Advantages: It is the only average that can be used for non-numerical data. It is not affected by extreme values (outliers). Disadvantages: A data set might have no mode, or it might have many modes, which makes it unhelpful.
The Median
The median is the middle value when data is in order. When to use it: - Use it when the data has outliers (extreme high or low values that don't fit the pattern). - It is commonly used for house prices or salaries because a few billionaires shouldn't make the "average" person look richer than they are! Advantages: It is not "pulled" away from the center by extreme values. Disadvantages: It doesn't use all the values in the data set, so it can ignore some information.
The Mean (Arithmetic Mean)
The mean is the "total divided by the count." When to use it: - Use it when the data is symmetrical and has no extreme outliers. - Use it when you want to include every single piece of data in your calculation. Advantages: Every value makes a contribution to the final answer. Disadvantages: It is very sensitive to outliers. One very large number will make the mean much higher than the rest of the data.
Quick Summary Table
Mode: Best for popularity and non-numbers.
Median: Best for data with extreme "weird" values (outliers).
Mean: Best for "fair" data with no extremes; uses all information.
2. Higher Tier Only: Special Averages
(Higher Tier Topic Only)
Sometimes, a simple mean isn't enough because some data points are more important than others, or the data is changing at a rate.
Weighted Mean
Use this when certain values have more "weight" or importance than others (e.g., an exam that is worth \(70\%\) of your grade vs. a quiz worth \(30\%\)).
Formula: \( \text{Weighted Mean} = \frac{\sum (\text{value} \times \text{weight})}{\sum \text{weights}} \)
Geometric Mean
The geometric mean is specifically used when dealing with rates of change or growth (like interest rates or population growth).
Formula: \( \text{Geometric Mean} = \sqrt[n]{\text{value}_1 \times \text{value}_2 \times \dots \times \text{value}_n} \)
Exam Tip: The term "geometric mean" will be explicitly stated in the question if you are required to use it.
3. Comparing Data Sets in Context
When an exam question asks you to "Compare two distributions," you must do two things to get full marks:
- Compare a measure of central tendency (Mean, Median, or Mode).
- Compare a measure of dispersion (usually Range or IQR).
How to write a perfect comparison:
Don't just list the numbers. You must interpret them in the context of the question.
Example: "The mean score for Class A (\(65\%\)) was higher than Class B (\(58\%\))."
Better: "On average, Class A performed better in the test than Class B because their mean score was higher."
Key Takeaway: Always use the phrase "On average..." and name the specific average you are using.
4. Skewness and Choosing Averages
The "shape" of the data distribution (skewness) tells you which average to trust.
- Symmetrical Distribution: Mean \(\approx\) Median \(\approx\) Mode. You can use any, but the Mean is usually best.
- Positive Skew (Tail to the right): Mean \(>\) Median \(>\) Mode. The mean is being pulled up by high outliers. The Median is a better "typical" value here.
- Negative Skew (Tail to the left): Mean \(<\) Median \(<\) Mode. The mean is being pulled down by low outliers. Use the Median.
5. Common Mistakes to Avoid
1. Using the mean for shoe sizes: If you calculate the mean shoe size as \(5.42\), a shop cannot stock that size! Use the mode (\(5\)) instead.
2. Ignoring outliers: If a data set is \(1, 2, 2, 3, 100\), the mean is \(21.6\). Does \(21.6\) look like the "middle" of that data? No. The outlier (\(100\)) ruined the mean. Use the median (\(2\)) instead.
3. Forgetting context: Never just say "The mean is bigger." Always say what that means for the story (e.g., "The plants grew taller on average with the new fertilizer").
Quick Review Box
Q: Which average is best for categorical data?
A: The Mode.
Q: Why use the Median instead of the Mean?
A: Because the Median is not affected by outliers.
Q: What does "On average" mean in an exam?
A: It is a general term that refers to the mean, median, or mode. You must specify which one you are using in your answer.