A sampling method has two parts to write about: why it fits the question, and what it cannot show. This lesson sets up a plan, then writes both parts.
It belongs to designing a small ethical geographical investigation. For the wider choice between random, systematic and stratified, see choosing a sampling approach.
How do I set out a stratified plan?
First list the groups, then their shares, then share the points out in proportion. Round so the points add up to your total.
Write the reason in one clause: the groups differ in the thing you measure, so each must be represented fairly.
Worked example: ground temperature on a school path
The numbers here are made up for illustration. A class investigates whether surface type changes ground temperature on a school path at 1 pm. The path is 50% grass verge, 30% concrete and 20% tar. There is time for 10 readings.
Points: grass 10 × 0.5 = 5, concrete 10 × 0.3 = 3, tar 10 × 0.2 = 2. These add up to 10.
The readings (°C) are:
| Surface | Readings | Mean |
|---|---|---|
| Grass | 30, 32, 31, 33, 34 | 160 ÷ 5 = 32 |
| Concrete | 38, 40, 39 | 117 ÷ 3 = 39 |
| Tar | 44, 46 | 90 ÷ 2 = 45 |
The tar mean is 45 − 32 = 13 °C higher than the grass mean. The overall mean is (160 + 117 + 90) ÷ 10 = 36.7 °C, which matches the share-weighted sum of 16 + 11.7 + 9 = 36.7.
How do I write the limitation?
Move from a vague sentence to a specific one. Compare three levels:
| Level | Wording |
|---|---|
| Vague | The sample was small. |
| Better | Only 10 points were used, so the results may not be representative. |
| Specific | Only 2 readings were taken on tar, so the tar mean of 45 °C could change a lot with one more point, and all readings were taken at 1 pm, so the pattern at other times is unknown. |
The specific version names the number, the group it affects, what could change, and a second weakness in timing.
The mistake that loses marks
A common slip is to state a limitation that blames effort: “I did not have enough time.” That explains why the sample is small but says nothing about the effect on the result.
The fix is one more clause: what the weakness does to the conclusion. In this plan, the claim “tar is hotter than grass” survives a 13 °C gap with 2 readings, but the exact tar mean does not. A marker credits that distinction.
Check your limitation against this test: does it say what is weak, by how much, and what it changes? Practise with the graph evidence and fair-comparison lab.
Check yourself
Original scenario: a class measures wind speed on a 400 m beach, where 75% of the beach is open sand and 25% is behind dunes. They have time for 12 readings. Share the points under stratified sampling and write a specific limitation.
Answer
Open sand: 12 × 0.75 = 9 points. Behind dunes: 12 × 0.25 = 3 points.
A specific limitation: “Only 3 readings were taken behind the dunes, so the mean wind speed there could change a lot with one extra reading, and all readings were taken in one hour, so the pattern on other days is unknown.”
What to study next
Continue with recording observations without collecting identifiable student or household data, which shows how to record your points safely.
To have a teacher test your plans and limitations on a new topic, see online one-to-one Geography tuition.