GCSE Maths Statistics and Probability - How to Solve: Comparing Distributions and Drawing Conclusions

How to Solve: Comparing Distributions and Drawing Conclusions

GCSE/A-Level (Physics, Chemistry, Biology) – Complete Guide

Introduction

"Mastering this skill lets you compare drug trials, climate data, or even exam results—exactly what examiners test in 6-mark questions worth 10% of your paper. Get this right, and you’ll secure top marks in data analysis."

WHAT YOU NEED TO KNOW FIRST

1. Measures of central tendency (mean, median, mode) – how to calculate and interpret them.
2. Measures of spread (range, interquartile range, standard deviation) – what they tell you about data variability.
3. Basic graph types (box plots, histograms, scatter graphs) – how to read and compare them.

KEY TERMS & FORMULAS

Key Terms

Formulas

1. Mean (Average)
   [
   \text{Mean} = \frac{\sum \text{data values}}{\text{number of values}}
   ]

2. MEMORISE THIS – Used to compare central values.

3. Range
   [
   \text{Range} = \text{Max value} - \text{Min value}
   ]

4. MEMORISE THIS – Measures total spread.

5. Interquartile Range (IQR)
   [
   \text{IQR} = Q3 - Q1
   ]

6. MEMORISE THIS – Measures spread of the middle 50% of data.

7. Standard Deviation (σ)

8. Given on exam sheet – Measures average distance from the mean.

STEP-BY-STEP METHOD

Step 1: Identify the distributions

• Are you comparing two datasets (e.g., Group A vs. Group B) or one dataset over time?
• Note the graph type (box plot, histogram, scatter graph).

Step 2: Compare central tendency

• Calculate/read the mean or median for each distribution.
• Ask: "Is one higher/lower? By how much?"

Step 3: Compare spread

• Calculate/read the range, IQR, or standard deviation.
• Ask: "Is one more spread out? Are there outliers?"

Step 4: Check for overlap

• Look at box plot whiskers or histogram bars.
• If distributions overlap a lot, differences may not be significant.

Step 5: Draw a conclusion

• Use comparative language:
• "Group A has a higher mean (X) than Group B (Y), suggesting…"
• "Group B has a wider spread (IQR = Z), meaning…"
• Link to the context (e.g., drug effectiveness, reaction times).

Step 6: Justify with data

• Quote specific numbers (e.g., "The median for Group A is 20, while Group B’s is 15").
• Mention overlap (e.g., "The IQRs overlap by 5 units, so the difference may not be reliable").

WORKED EXAMPLES

Example 1 – Basic (Box Plots)

Question: Two classes took the same test. Compare their performance using the box plots below.

Step-by-Step Solution: 1. Central tendency: Class A’s median (65) > Class B’s (55). 2. Spread: Class A’s IQR = 75 – 55 = 20. Class B’s IQR = 65 – 45 = 20. 3. Overlap: Class A’s Q1 (55) overlaps with Class B’s Q3 (65). 4. Conclusion: Class A performed better on average, but both classes have similar spread. The overlap suggests some students in Class B scored as high as Class A’s middle performers.

What we did and why: - Compared medians to see which group did better on average. - Checked IQRs to see if one group was more consistent. - Noted overlap to assess if the difference was meaningful.

Example 2 – Medium (Histograms)

Question: Two fertilizers were tested on plant growth. Compare their effectiveness.

Step-by-Step Solution: 1. Central tendency: Fertilizer Y has a higher mean (18 cm vs. 15 cm). 2. Spread: Fertilizer Y has a larger standard deviation (4.5 vs. 2.1), meaning more variability. 3. Overlap: If we sketch the distributions, Y’s lower end (18 – 4.5 = 13.5 cm) overlaps with X’s upper end (15 + 2.1 = 17.1 cm). 4. Conclusion: Fertilizer Y produces taller plants on average, but results are less consistent. Some plants with Y may grow less than those with X.

What we did and why: - Used mean to compare average growth. - Used standard deviation to assess reliability. - Considered overlap to avoid overstating differences.

Example 3 – Exam-Style (Scatter Graph + Context)

Question (6 marks): A scientist measured reaction times (ms) for two groups: one given caffeine and one given a placebo. - Caffeine group: Mean = 250 ms, SD = 30 ms - Placebo group: Mean = 300 ms, SD = 25 ms Compare the distributions and draw a conclusion about caffeine’s effect.

Step-by-Step Solution: 1. Central tendency: Caffeine group has a lower mean (250 ms vs. 300 ms), suggesting faster reactions. 2. Spread: Caffeine group has a slightly higher SD (30 ms vs. 25 ms), meaning more variability. 3. Overlap: Calculate approximate ranges:
- Caffeine: 250 ± 2×30 = 190–310 ms
- Placebo: 300 ± 2×25 = 250–350 ms
- Overlap: 250–310 ms (significant overlap). 4. Conclusion: Caffeine appears to reduce reaction time on average, but the overlap means some placebo users reacted as fast as caffeine users. The effect may not be consistent for everyone.

What we did and why: - Compared means to assess average effect. - Used SD to check reliability. - Calculated overlap to evaluate real-world significance.

COMMON MISTAKES

EXAM TRAPS

1-MINUTE RECAP

"Right, listen up—this is your last-minute cheat sheet for comparing distributions. First, compare the middle: mean or median. Second, compare the spread: range, IQR, or standard deviation. Third, check for overlap—if the distributions share values, the difference might not be real. Finally, write a conclusion that links to the question, using numbers from the data. Example: ‘Group A has a higher median (20 vs. 15) but a wider IQR (10 vs. 5), so it’s less consistent.’ Avoid vague answers—examiners want specifics. Now go smash that question!"