Comparing Distributions and Making Inferences
Welcome to one of the most practical and useful topics in GCSE Mathematics! Have you ever wondered how sports coaches decide which player is more reliable, or how science researchers know if a new medicine actually works better than an old one? They do it by comparing distributions.
In this chapter, you will learn how to look at two sets of data side-by-side, calculate the right statistics, and write clear, sensible conclusions. Don't worry if you find statistics intimidating at first — once you learn the simple Two-Step Rule, you will be able to answer these questions with total confidence!
1. The Golden Rule of Comparing Distributions
Whenever an exam question asks you to "Compare the two distributions" or "Compare these two sets of data", you must always mention two specific things, both written in the context of the question:
1. An Average (Measure of Central Tendency): You must compare either the Medians or the Means.
What it tells you: Which group performed better, scored higher, was faster, or was generally larger.
2. A Measure of Spread (Dispersion): You must compare either the Interquartile Ranges (IQR) or the Ranges.
What it tells you: Which group was more consistent, more reliable, or had results closer together.
Memory Trick: The "A-S-C" Checklist
Keep this simple checklist in your head:
• A - Average (Compare Mean or Median)
• S - Spread (Compare Range or IQR)
• C - Context (Mention units and what the numbers represent, like test marks, reaction times in seconds, or goals scored)
Key Takeaway: Never give just one comparison! To get full marks, you need one sentence for the average and one sentence for the spread, both linked to the real-life story of the question.
2. Choosing the Best Average and Spread
You have a few tools in your toolkit, but which ones should you pick?
Averages: Mean vs. Median
• Mean: Calculated by adding all values and dividing by the total count: \(\text{Mean} = \frac{\sum x}{n}\). It uses every single data value, but it is heavily affected by extreme values (outliers).
• Median: The middle value when data is arranged in order. It is not affected by extreme outliers.
Everyday Analogy: Imagine four friends have £\(10\), £\(12\), £\(15\), and £\(13\). Then a billionaire walks into the room with £\(1,000,000\)! The mean jumps up to over £\(200,000\), which does not represent the group at all. But the median remains around £\(13\), giving a much truer picture of a typical person's money.
Spread: Range vs. Interquartile Range (IQR)
• Range: \(\text{Range} = \text{Highest Value} - \text{Lowest Value}\). Simple to calculate, but like the mean, it is distorted by single extreme values.
• Interquartile Range (IQR): \(\text{IQR} = \text{Upper Quartile } (Q_3) - \text{Lower Quartile } (Q_1)\). The IQR measures the spread of the middle \(50\%\) of the data, ignoring any bizarre extremes.
Rule of Thumb for Exams
• If you are working with Box Plots or Cumulative Frequency Curves: Use the Median and the IQR.
• If you are given raw lists or grouped frequency tables with the mean already calculated: Use the Mean and the Range.
Key Takeaway: The median and IQR are the best pair to use when there are extreme values (outliers) or skewed data.
3. Comparing Box Plots
A Box Plot (or Box-and-Whisker diagram) shows the five-number summary of a distribution at a glance:
1. Minimum value
2. Lower Quartile (\(Q_1\))
3. Median (\(Q_2\))
4. Upper Quartile (\(Q_3\))
5. Maximum value
How to Compare Two Box Plots Side-by-Side
Suppose you have two box plots representing test marks for Class 10A and Class 10B:
• Class 10A: \(\text{Median} = 65\), \(\text{IQR} = 78 - 50 = 28\)
• Class 10B: \(\text{Median} = 74\), \(\text{IQR} = 80 - 68 = 12\)
Here is how you write your two comparison sentences:
Sentence 1 (Average): "Class 10B performed better on average because their median mark of \(74\) was higher than Class 10A's median mark of \(65\)."
Sentence 2 (Spread): "Class 10B's marks were more consistent because their interquartile range of \(12\) was smaller than Class 10A's interquartile range of \(28\)."
Vocabulary Tip for Spread:
• Smaller Spread (smaller IQR / Range) means: more consistent, less variation, more predictable.
• Larger Spread (larger IQR / Range) means: less consistent, more varied, more spread out.
Key Takeaway: When comparing box plots, read the middle line for the median and measure the length of the box for the IQR. Always state the numerical values to back up your statement!
4. Comparing Other Diagrams
1. Back-to-Back Stem-and-Leaf Diagrams
A back-to-back stem-and-leaf diagram shares a central stem to compare two groups easily. Remember:
• For the left-hand group, leaves are read backwards (from the stem outwards to the left).
• Always check the Key! For example, \(1 \mid 4\) means \(14\), or \(1 \mid 4\) could mean \(1.4\text{ kg}\).
• Find the total number of items \(n\) in each group, find the middle value at position \(\frac{n + 1}{2}\) to determine the medians, and find the extremes to get the ranges.
2. Frequency Polygons
When two frequency polygons are drawn on the same set of axes:
• Peak position: Look at where the highest peak lies on the horizontal axis. A peak shifted further to the right indicates higher typical values.
• Width of the polygon: A tall, narrow polygon indicates lower spread (more consistency). A wide, flat polygon indicates a larger spread.
3. Cumulative Frequency Curves
• Read the Median at \(50\%\) of the total cumulative frequency.
• Read the Lower Quartile (\(Q_1\)) at \(25\%\) and Upper Quartile (\(Q_3\)) at \(75\%\).
• A steeper cumulative frequency curve around the middle means data values are tightly clustered together (smaller IQR).
Key Takeaway: Whatever the diagram, your final written response must always fall back on: Average + Spread + Context.
5. Making Inferences and Drawing Conclusions
An inference is a logical conclusion you draw based on evidence from data. Exam questions often ask questions like: "The manager says Group A worked harder than Group B. Is she correct? Justify your answer."
How to Structure Your Answer:
1. Make a clear decision: State "Yes" or "No" (or explain if it is partially true).
2. Provide numerical evidence: Quote the exact medians/means or ranges/IQRs.
3. Add context: Explain what the numbers show in terms of the question scenario.
Worked Example:
Scenario: A running club tests two groups of runners over a \(5\text{ km}\) route.
• Group 1: \(\text{Mean time} = 24.5\text{ minutes}\), \(\text{Standard Range} = 8\text{ minutes}\)
• Group 2: \(\text{Mean time} = 21.0\text{ minutes}\), \(\text{Standard Range} = 14\text{ minutes}\)
Question: Which group would you choose to represent the club in a team competition where consistency is vital? Explain your choice.
Model Solution:
"I would choose Group 1. Although Group 2 was faster on average (lower mean of \(21.0\text{ mins}\) compared to \(24.5\text{ mins}\)), Group 1 is much more consistent because their range of \(8\text{ minutes}\) is smaller than Group 2's range of \(14\text{ minutes}\)."
Did you know? In sports with time (like sprinting, swimming, or running), a smaller average is actually better because it means the competitors are faster!
6. Common Mistakes to Avoid
• Mistake 1: Forgetting the context.
Wrong: "The median is higher and the range is smaller." (0 marks)
Right: "The boys have a higher median score of \(55\) compared to the girls' median of \(48\), meaning the boys performed better overall."
• Mistake 2: Confusing consistency with being better.
A smaller spread does not mean higher marks; it only means results are more tightly bunched together.
• Mistake 3: Stating values without comparing them.
Wrong: "Class A median is \(12\) and Class B median is \(15\)."
Right: "Class B has a higher median (\(15\)) than Class A (\(12\))."
• Mistake 4: Mixing up measures.
Do not compare the Mean of Group A with the Median of Group B. Always compare like with like: Mean with Mean, or Median with Median.
Quick Review Summary
1. Always write two points: One for an average (Mean/Median) and one for spread (Range/IQR).
2. Use the word 'consistent': When discussing a smaller Range or smaller IQR.
3. Include units and numbers: Always quote values calculated from the graphs or tables.
4. Check the direction: Remember that in races, lower times mean faster performance!