Making estimates (Edexcel GCSE Statistics): Revision Notes
Making estimates
Introduction to making estimates
When we collect data from a small group (called a representative sample), we can use this information to make educated guesses about a much larger group (the population). This process is called making estimates, and it's incredibly useful in real-world situations where we cannot collect data from everyone.
Representative samples are crucial for accurate estimation - the sample must truly reflect the characteristics of the larger population we want to understand.
The key statistical measures we use for making estimates are:
- Mean (average)
- Median (middle value)
- Range (difference between highest and lowest values)
- Interquartile range (spread of the middle 50% of data)
Understanding how data is distributed
To make accurate estimates, we need to understand how data spreads out in a distribution. Think of this like understanding how marks are spread across a class test.
The median and quartiles
The median is the middle value that splits our data exactly in half:
- 50% of all values fall below the median
- 50% of all values fall above the median
Critical Concept: Quartiles and Percentages
Quartiles divide our data into four equal parts, and understanding these percentages is essential for making accurate estimates.
Quartiles divide our data into four equal parts:
- Lower quartile (LQ): 25% of data falls below this value
- Median: 50% of data falls below this value
- Upper quartile (UQ): 75% of data falls below this value (or 25% falls above it)
- The interquartile range contains the middle 50% of all data (between LQ and UQ)
Worked example: blood pressure estimates
Let's work through a detailed example to see how estimation works in practice.
Worked Example: Estimating Blood Pressure Statistics
The scenario: A hospital recorded the systolic blood pressure of patients admitted in one day. Here are the key statistics from their sample:
| Minimum | Lower quartile | Median | Upper quartile | Maximum |
|---|---|---|---|---|
| 107.5 | 112.4 | 116.7 | 124.9 | 160.1 |
Step-by-step solution:
Question (a): What proportion of the sample had a systolic blood pressure less than 112.4?
Answer: Since 112.4 is the lower quartile, we know that 25% of the sample had a systolic blood pressure less than 112.4.
Question (b): If 41,500 patients were admitted to UK hospitals in one day, estimate how many had:
- (i) A systolic blood pressure less than 112.4
- (ii) A systolic blood pressure greater than 116.7
Solution for (i):
- We know 25% of patients have blood pressure less than 112.4
- Calculate:
Solution for (ii):
- We know 116.7 is the median, so 50% have blood pressure greater than this
- Calculate:
Key calculation method
When making estimates from sample data:
- Identify the relevant statistic (lower quartile, median, upper quartile)
- Convert the percentage to a decimal (25% = 0.25, 50% = 0.5)
- Multiply by the total population (percentage as decimal × total population)
Working with percentiles
Understanding Percentiles and Quartiles
Sometimes data is presented using percentiles rather than quartiles. The relationship is straightforward:
- 25th percentile = Lower quartile
- 50th percentile = Median
- 75th percentile = Upper quartile
For any percentile, that percentage of the data falls below that value.
Common exam tips and traps
Typical problem-solving approach:
- Read the question carefully - identify what statistic you need (mean, median, quartile)
- Check the sample size - make sure you're applying percentages to the correct total
- Show your working - write out the percentage calculation clearly
- Check your answer - does it make sense given the context?
Common Exam Traps to Avoid:
- Mixing up quartiles: Remember LQ = 25%, Median = 50%, UQ = 75%
- Percentage errors: Always convert percentages to decimals before multiplying
- Sample vs population confusion: Make sure you're estimating for the right group size
- Reading tables incorrectly: Double-check which column or row contains the data you need
Key Points to Remember:
- Representative samples allow us to estimate statistics for entire populations
- Quartiles divide data into quarters: 25% below LQ, 25% between LQ and median, 25% between median and UQ, 25% above UQ
- The median splits data exactly in half - 50% above, 50% below
- To calculate estimates: convert percentage to decimal, then multiply by total population size
- Always show your working clearly in exam questions, writing out each step of your calculation